Multi-source information fusion measurement system combining laser radar data and wind resource evaluation

Abstract

The invention relates to the technical field of wind resource assessment and wind power plant optimization, and discloses a multi-source information fusion measurement system combining laser radar data and wind resource assessment, and the system comprises a data collection module which is used for collecting laser radar data, synthetic aperture radar data, numerical weather forecast data and ground weather station data; the data preprocessing module is used for carrying out noise filtering, time alignment, space alignment and data calibration on the collected data; the information fusion module is used for dynamically fusing the preprocessed multi-source data and outputting high-precision wind speed and turbulence characteristic distribution; and the wind field modeling and prediction module is used for constructing a dynamic three-dimensional wind field based on the fusion data and performing short-time wind speed prediction. Through multi-source data deep fusion, dynamic wind field modeling and intelligent optimization modules, the comprehensive effects of high-precision wind field evaluation, nonlinear wind speed prediction and fan layout global optimization are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind resource assessment and wind farm optimization, and in particular to a multi-source information fusion measurement system combining laser radar data with wind resource assessment. Background Art

[0002] The multi-source information fusion measurement system is an important technical means to integrate multiple data sources to obtain comprehensive and accurate environmental information, and has been widely used in meteorology, transportation, energy and other fields. In the field of wind resource assessment, multi-source information fusion technology can achieve comprehensive measurement of wind speed, turbulence characteristics and wind field distribution on a spatiotemporal scale by integrating multiple data sources such as lidar, synthetic aperture radar (SAR), ground meteorological stations and numerical weather forecasts (NWP), providing scientific support for wind energy resource development and wind farm site selection.

[0003] The application of existing multi-source information fusion measurement systems in wind resource assessment can provide short-term prediction effects of local wind speed and trend by combining lidar and numerical weather forecast models, and use SAR data to measure the spatial distribution of large-scale wind fields. These systems improve the limitations of single data sources in spatial resolution or temporal dynamics through statistically based weighted fusion methods, and can provide reasonable predictions of spatial uniformity of wind speed and short-term wind field changes to a certain extent.

[0004] However, the existing multi-source information fusion measurement system still has great limitations in practical applications. Due to the lack of deep fusion methods for data sources such as lidar, SAR and NWP, the existing system cannot effectively take into account the dynamic coupling of high-resolution local data and large-scale wind field distribution; at the same time, its single prediction model has weak ability to capture nonlinear changes in wind speed, and it is difficult to meet the needs of high-precision wind field modeling under complex terrain or dynamic meteorological conditions. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention provides a multi-source information fusion measurement system that combines lidar data with wind resource assessment, which solves the problem that the existing technology lacks deep fusion of multi-source data, resulting in the inability to effectively couple high-resolution local data with large-scale wind field dynamic characteristics.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multi-source information fusion measurement system combining lidar data with wind resource assessment, the system comprising:

[0007] Data acquisition module, used to collect lidar data, synthetic aperture radar data, numerical weather forecast data and ground meteorological station data;

[0008] A data preprocessing module, used for performing noise filtering, time alignment, space alignment and data calibration on the collected data;

[0009] Information fusion module, used to dynamically fuse pre-processed multi-source data and output high-precision wind speed and turbulence characteristic distribution;

[0010] Wind field modeling and prediction module, used to build a dynamic three-dimensional wind field based on fused data and perform short-term wind speed prediction;

[0011] The intelligent assessment and optimization module is used to conduct wind resource assessment, wind turbine optimization layout and operation scheduling decisions based on modeling results.

[0012] Preferably, the data preprocessing module comprises the following steps:

[0013] Noise filtering: Use discrete wavelet transform to denoise the lidar data;

[0014] Time alignment: Linear interpolation method is used to complete the time alignment of lidar data and synthetic aperture radar data;

[0015] Spatial alignment: The SAR data are resampled to a grid of lidar resolution using bilinear interpolation.

[0016] Data calibration: Use ground-based weather station data to perform error correction on lidar and synthetic aperture radar data.

[0017] Preferably, the discrete wavelet transform in the noise filtering step is performed according to the following formula:

[0018]

[0019] Where: S is the original signal; D k is the detail coefficient; A k is the approximation coefficient; N is the number of decomposition levels.

[0020] Preferably, the information fusion module adopts a dynamic Kalman filtering method to achieve fusion of multi-source data, which specifically includes the following steps:

[0021] Status prediction:

[0022]

[0023] in: is the predicted state; F is the state transfer matrix;

[0024] Measurement Update:

[0025]

[0026] Where: Kk is the Kalman gain; z k is the measurement data; H is the observation matrix.

[0027] Preferably, the Kalman gain is calculated according to the following formula:

[0028] K k =P k|k-1 H T (HP k|k-1 H T +R) -1

[0029] Where: P k|k-1 is the prediction error covariance; R is the measurement noise covariance.

[0030] Preferably, the wind field modeling and prediction module comprises the following steps:

[0031] Dynamic three-dimensional wind field modeling, calculation of turbulence intensity TI and wind shear coefficient α based on fusion data;

[0032] For short-term wind speed forecast, a long short-term memory network (LSTM) is used to model the wind speed time series.

[0033] Preferably, the turbulence intensity TI in the dynamic three-dimensional wind field modeling step is calculated according to the following formula:

[0034]

[0035] Where: TI is the turbulence intensity; σ u is the standard deviation of wind speed; is the average wind speed.

[0036] Preferably, the wind shear coefficient α is calculated according to the following formula:

[0037]

[0038] Among them: u1, u2 are wind speeds at two heights; h1, h2 are two corresponding heights.

[0039] Preferably, the intelligent evaluation and optimization module includes the following sub-steps:

[0040] Wind resource assessment, calculation of wind power density P;

[0041] Optimize the layout of wind turbines and complete the layout of wind turbines based on multi-objective optimization algorithms;

[0042] Decision support and visualization, building a three-dimensional wind farm dynamic display.

[0043] Preferably, the wind energy power density P is calculated according to the following formula:

[0044]

[0045] Where: P is wind power density; ρ is air density; A is wind turbine swept area; V is wind speed; C p is the wind energy utilization coefficient.

[0046] The present invention provides a multi-source information fusion measurement system combining laser radar data with wind resource assessment.

[0047] It has the following beneficial effects:

[0048] 1. The present invention achieves high-resolution dynamic modeling and prediction of wind farms by accurately integrating lidar, SAR and numerical weather forecast data through multi-source information fusion. It can comprehensively and in real time characterize key wind farm characteristics such as wind speed, turbulence intensity and wind shear, thus providing reliable support for the scientific layout and dynamic scheduling of wind farms.

[0049] 2. Through the intelligent evaluation and optimization module, the present invention can combine the wind power density calculation with the multi-objective optimization algorithm to output the wind turbine optimization layout and operation scheduling strategy, improve the power generation efficiency of the wind farm, reduce the impact of turbulent loads, extend the service life of the wind turbine, and optimize the utilization of wind energy.

[0050] 3. By leveraging real-time dynamic update capabilities and a visualization platform, the present invention realizes an intuitive display of the three-dimensional distribution and temporal dynamic changes of the wind farm, provides a scientific basis for the planning and intelligent management of wind farms, and improves the comprehensive benefits of wind resource development and the operability of system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a system structure diagram of the present invention;

[0052] Figure 2 This is a system structure diagram of the data preprocessing module of the present invention;

[0053] Figure 3 This is a system structure diagram of the intelligent evaluation and optimization module of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Please see attached Figure 1 -Attached Figure 3The embodiment of the present invention provides a multi-source information fusion measurement system combining laser radar data and wind resource assessment, the system comprising:

[0056] The data acquisition module is used to collect lidar data, synthetic aperture radar data, numerical weather forecast data and ground meteorological station data. Through multi-source data acquisition, the system can not only capture local turbulence characteristics, but also provide large-scale distribution and dynamic change trends of wind fields, providing high-quality input for subsequent processing and fusion;

[0057] Data preprocessing module, used to filter noise, time align, space align and calibrate the collected data. The data preprocessing module improves the quality and consistency of multi-source data, ensures the accurate combination of different data sources, and provides high-precision input for dynamic fusion;

[0058] The information fusion module is used to dynamically fuse the pre-processed multi-source data and output high-precision wind speed and turbulence characteristic distribution. The information fusion module effectively combines the advantages of multi-source data to generate dynamic and refined wind speed and turbulence characteristic distribution, providing a solid foundation for subsequent wind field modeling;

[0059] Wind farm modeling and prediction module, which is used to build a dynamic three-dimensional wind farm based on fused data and perform short-term wind speed prediction. The wind farm modeling and prediction module provides dynamic and accurate wind farm distribution and short-term prediction results, helping wind farms achieve real-time response and optimize layout;

[0060] The intelligent evaluation and optimization module is used to conduct wind resource evaluation, wind turbine optimization layout and operation scheduling decisions based on modeling results. The intelligent evaluation and optimization module effectively improves the scientific nature of wind resource development, reduces the design risks of wind farms, and improves the efficiency of wind energy utilization.

[0061] Please see attached Figure 1 and attached Figure 2 In a preferred embodiment of the present invention, the data preprocessing module includes the following steps:

[0062] Noise filtering: The laser radar data is denoised using discrete wavelet transform. Noise filtering improves the signal-to-noise ratio of the laser radar data, reduces the error in wind speed and wind direction estimation, and provides a higher quality data basis for subsequent processing;

[0063] Time alignment: The linear interpolation method is used to complete the time alignment of lidar data and synthetic aperture radar data. Time alignment solves the problem of inconsistent timestamps of multi-source data, so that the subsequent fusion steps can align the time dynamic characteristics and avoid the deviation of wind field calculation caused by time error;

[0064] Spatial alignment: The bilinear interpolation method is used to resample the synthetic aperture radar data to a grid of lidar resolution. Spatial alignment makes the spatial distribution of multi-source data consistent, avoids wind field distribution deviation caused by resolution differences, and improves the accuracy of fusion and modeling.

[0065] Data calibration: The error correction of LiDAR and SAR data is carried out using ground-based meteorological station data. Data calibration significantly reduces the measurement errors of LiDAR and SAR data, and improves the physical credibility and data accuracy of subsequent wind field modeling.

[0066] Please see attached Figure 1 and attached Figure 2 In a preferred embodiment of the present invention, the discrete wavelet transform in the noise filtering step is performed according to the following formula:

[0067]

[0068] Where: S is the original signal; D k is the detail coefficient; A k is the approximation coefficient; N is the number of decomposition layers. The core of discrete wavelet transform is to decompose the original signal into components of different frequencies to achieve the separation of signal and noise. The effective signals in the lidar wind speed signal are mostly concentrated in the low-frequency components, while the noise is mainly manifested as high-frequency oscillations. Through decomposition and threshold processing, high-frequency noise is effectively removed and low-frequency signals are retained, thereby improving the quality of wind speed signals, ensuring the data foundation stability and accuracy of the measurement system, and laying a solid foundation for subsequent multi-source data fusion and wind field modeling.

[0069] Please see attached Figure 1 In a preferred embodiment of the present invention, the information fusion module uses a dynamic Kalman filtering method to achieve the fusion of multi-source data, which specifically includes the following steps:

[0070] Status prediction:

[0071]

[0072] in: is the predicted state; F is the state transfer matrix. The state prediction uses the time dynamic model of the system to estimate the current wind speed distribution and turbulence characteristics, and preliminarily integrates the physical characteristics of time continuity. The state prediction provides an initial estimate for measurement update, which can reduce the deviation in multi-source data fusion and improve real-time performance.

[0073] Measurement Update:

[0074]

[0075] Where: K kis the Kalman gain; z k is the measurement data; H is the observation matrix. The measurement update dynamically adjusts the state estimate according to the actual observation data, and achieves the optimal combination of the predicted value and the observed value through the Kalman gain, which improves the accuracy of data fusion. The measurement update dynamically corrects the prediction error, making full use of the local high-resolution characteristics of the lidar, the large-scale characteristics of the SAR and the trend prediction ability of the NWP, ensuring the high accuracy of the fusion result.

[0076] Please see attached Figure 1 In a preferred embodiment of the present invention, the Kalman gain is calculated according to the following formula:

[0077] K k =P k|k-1 H T (HP k|k-1 H T +R) -1

[0078] Where: P k|k-1 is the prediction error covariance; R is the measurement noise covariance. The role of the Kalman gain is to dynamically balance the contribution ratio of the predicted state and the observed data. Its core idea is to minimize the mean square error of the state update. By comparing the prediction error covariance and the measurement noise covariance, the Kalman gain can automatically adjust the weight under different data credibility conditions. This dynamic adjustment mechanism ensures the robustness and accuracy of multi-source data fusion, while improving the accuracy of the fusion results and ensuring the high credibility of key parameters such as wind speed and wind direction.

[0079] Please see attached Figure 1 In a preferred embodiment of the present invention, the wind field modeling and prediction module includes the following steps:

[0080] Dynamic three-dimensional wind field modeling calculates turbulence intensity TI and wind shear coefficient α based on fusion data. Turbulence intensity and wind shear coefficient describe the randomness and high-variability characteristics of wind speed respectively. Combining these parameters can accurately characterize the local characteristics and overall trend of the wind field. The three-dimensional wind field model reflects the spatial distribution of the wind field in real time through grid calculation and dynamic update, thereby improving the accuracy of wind field characterization in complex terrain and turbulent environment, and providing a scientific basis for the optimization of wind farm layout;

[0081] For short-term wind speed forecasting, a long short-term memory network is used to model the wind speed time series. LSTM can capture the long-term dependencies in the wind speed time series, and is particularly suitable for nonlinear and dynamically changing wind speed forecasting scenarios. Through the memory mechanism, LSTM effectively solves the problem of insufficient long-term dependency modeling capabilities in traditional time series models, and provides reliable support for the operation scheduling optimization and short-term energy management of wind farms.

[0082] Please see attached Figure 1 In a preferred embodiment of the present invention, the turbulence intensity TI in the dynamic three-dimensional wind field modeling step is calculated according to the following formula:

[0083]

[0084] Where: TI is the turbulence intensity; σ u is the standard deviation of wind speed; is the average wind speed. Turbulence intensity is an important indicator for measuring wind field fluctuations. Its calculation is based on the time series characteristics of wind speed. The standard deviation σ u : reflects the amplitude of wind speed fluctuation and is a direct representation of the randomness of turbulence; average wind speed A baseline value of turbulence intensity is provided to normalize the degree of fluctuation. By calculating the TI of each grid cell, a turbulence distribution map of the wind field can be constructed to accurately characterize the random fluctuation characteristics of wind speed. The turbulence intensity calculation provides a quantitative description of wind speed fluctuations and improves the level of refinement of three-dimensional wind field modeling.

[0085] Please see attached Figure 1 In a preferred embodiment of the present invention, the wind shear coefficient α is calculated according to the following formula:

[0086]

[0087] Among them: u1, u2 are wind speeds at two heights; h1, h2 are two corresponding heights. By dynamically calculating the wind shear coefficient, the changing characteristics of wind speed with height can be accurately described, which provides important technical support for wind field modeling and wind resource assessment. By combining multi-source data fusion, the calculation results of the wind shear coefficient can be updated in real time, improving the accuracy and adaptability of complex wind field modeling.

[0088] Please see attached Figure 1 and attached Figure 3 In a preferred embodiment of the present invention, the intelligent evaluation and optimization module includes the following sub-steps:

[0089] Wind resource assessment, calculation of wind power density P. Wind resource assessment provides a scientific basis for wind farm site selection, can accurately locate areas with high wind energy potential, optimize wind energy resource utilization efficiency, and reduce wind farm development risks;

[0090] Wind turbine optimization layout, wind turbine arrangement is completed based on multi-objective optimization algorithm. Wind turbine optimization layout optimizes wind turbine arrangement through multi-objective optimization algorithm (such as genetic algorithm or particle swarm optimization algorithm) to balance power generation efficiency and turbulence load. Multi-objective optimization algorithm ensures efficient operation of wind farms by weighing power generation and turbulence load, while reducing the loss of turbulence to wind turbine structure, improving the operating life and economic benefits of wind farms, thereby improving the power generation efficiency of wind farms and reducing the wear of wind turbine structure caused by turbulence, providing reliable support for the long-term stable operation of wind farms;

[0091] Decision support and visualization, building a three-dimensional dynamic display of the wind farm. Visualization transforms complex wind farm modeling and optimization data into an intuitive three-dimensional display. Combined with dynamic analysis functions, it provides strong data support and decision-making basis for the planning and operation of wind farms. The decision support and visualization module enables users to intuitively understand the wind farm characteristics and wind turbine layout optimization results, improves the scientificity and efficiency of wind farm planning and operation, and reduces the risk of misjudgment due to information complexity.

[0092] Please see attached Figure 1 and attached Figure 3 In a preferred embodiment of the present invention, the wind power density P is calculated according to the following formula:

[0093]

[0094] Where: P is wind power density; ρ is air density; A is wind turbine swept area; V is wind speed; C p The wind energy utilization coefficient and wind power density are calculated through the above formula, which accurately quantifies the potential of wind energy resources, provides important technical support for wind farm planning and optimization, and combines multi-source data and dynamic update capabilities to improve the accuracy and real-time performance of wind resource assessment.

[0095] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-source information fusion measurement system combining lidar data with wind resource assessment, characterized in that: The system comprises: Data acquisition module, used to collect lidar data, synthetic aperture radar data, numerical weather forecast data and ground meteorological station data; A data preprocessing module, used for performing noise filtering, time alignment, space alignment and data calibration on the collected data; Information fusion module, used to dynamically fuse pre-processed multi-source data and output high-precision wind speed and turbulence characteristic distribution; Wind field modeling and prediction module, used to build a dynamic three-dimensional wind field based on fused data and perform short-term wind speed prediction; The intelligent assessment and optimization module is used to conduct wind resource assessment, wind turbine optimization layout and operation scheduling decisions based on modeling results.

2. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 1 is characterized in that: The data preprocessing module comprises the following steps: Noise filtering: Use discrete wavelet transform to denoise the lidar data; Time alignment: Linear interpolation method is used to complete time alignment of lidar data and synthetic aperture radar data; Spatial alignment: The SAR data are resampled to a grid of lidar resolution using bilinear interpolation. Data calibration: Use ground weather station data to perform error correction on lidar data and synthetic aperture radar data.

3. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 2 is characterized in that: The discrete wavelet transform in the noise filtering step is performed according to the following formula: Where: S is the original signal; D k is the detail coefficient; A k is the approximation coefficient; N is the number of decomposition levels.

4. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 1 is characterized in that: The information fusion module uses a dynamic Kalman filtering method to achieve the fusion of multi-source data, which specifically includes the following steps: Status prediction: in: is the predicted state; F is the state transfer matrix; Measurement Update: Where: K k is the Kalman gain; z k is the measurement data; H is the observation matrix.

5. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 4 is characterized in that: The Kalman gain is calculated according to the following formula: K k =P k|k-1 H T (HP k|k-1 H T +R) -1 Where: P k|k-1 is the prediction error covariance; R is the measurement noise covariance.

6. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 1 is characterized in that: The wind field modeling and prediction module includes the following steps: Dynamic three-dimensional wind field modeling, calculation of turbulence intensity TI and wind shear coefficient α based on fusion data; For short-term wind speed forecast, a long short-term memory network (LSTM) is used to model the wind speed time series.

7. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 6 is characterized in that: The turbulence intensity TI in the dynamic three-dimensional wind field modeling step is calculated according to the following formula: Where: TI is the turbulence intensity; σ u is the standard deviation of wind speed; is the average wind speed.

8. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 6 is characterized in that: The wind shear coefficient α is calculated according to the following formula: Among them: u1, u2 are wind speeds at two heights; h1, h2 are two corresponding heights.

9. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 1 is characterized in that: The intelligent evaluation and optimization module includes the following sub-steps: Wind resource assessment, calculation of wind power density P; Optimize the layout of wind turbines and complete the layout of wind turbines based on multi-objective optimization algorithms; Decision support and visualization, building a three-dimensional wind farm dynamic display.

10. The multi-source information fusion measurement system combining lidar data and wind resource assessment according to claim 9, characterized in that: The wind power density P is calculated according to the following formula: Where: P is wind power density; ρ is air density; A is wind turbine swept area; V is wind speed; C p is the wind energy utilization coefficient.

Images