Method, device and equipment for evaluating credible capacity of wind power in extreme weather and medium

Abstract

The invention provides a wind power credible capacity evaluation method, device and equipment in extreme weather and a medium, and relates to the technical field of power dispatching. The method comprises the following steps: acquiring a wind speed modal component and a temperature modal component in a current rolling time window, and historical power of a wind turbine generator; inputting the wind speed modal component, the temperature modal component and the historical power into a trained physical information neural network model to obtain an output prediction result of the wind turbine generator; wherein the physical information neural network model takes a wind speed power characteristic model of a wind turbine generator considering an air temperature factor and / or a wind turbine generator output model considering low-temperature protection as physical constraints; and calculating a reliability index of a power system where the wind turbine generator is located based on the output prediction result, and solving by adopting an equal-reliability Monte Carlo method to obtain the dynamic credible capacity of the wind turbine generator in the current rolling time window. The wind power prediction accuracy and universality can be improved, and physical quantity expression of schedulability measurement is provided.

Description

Technical Field

[0001] This invention relates to the field of power dispatching technology, and in particular to a method, apparatus, equipment and medium for assessing the reliable capacity of wind power under extreme weather conditions. Background Technology

[0002] With the increasing global demand for clean energy, wind power, as a renewable and pollution-free energy source, is gradually taking up a larger share of the energy mix. The reliable capacity of wind power can assess the extent to which a wind farm can provide a credible capacity with a reasonable or acceptable confidence probability. A reasonable assessment of the reliable capacity of wind farms is crucial for determining the capacity of some thermal power units that can be replaced by wind power, as well as for power system dispatching and future planning.

[0003] However, in recent years, against the backdrop of global warming, cold waves, characterized by strong winds and low temperatures, have exhibited more extreme climate features. With the increasing proportion of wind power connected to the grid, cold waves pose a serious threat to the normal operation of wind power and the normal supply of electricity. In extremely cold weather, when the ambient temperature falls below a set threshold, wind turbines automatically trigger low-temperature protection and disconnect from the grid, only reconnecting when the temperature returns to the specified value. The question of how much wind power capacity should be included in the power reserve calculation during this period urgently needs to be addressed.

[0004] Current wind power reliable capacity assessment methods include analytical methods, simulation methods (convolutional methods, sequence algorithms, Monte Carlo simulations, etc.), and dynamic reliable capacity assessment. Technically, machine learning and deep learning have improved the accuracy of power prediction, while numerical weather prediction and transmission line fault models have been combined to construct refined reliability assessment models. Application scenarios have also expanded from routine operation to reliability assessment under high-proportion wind power integration and extreme weather conditions.

[0005] However, many problems remain. Wind power prediction accuracy under extreme weather conditions is insufficient; traditional models struggle to capture power losses and have limited generalization capabilities. In reliability assessment, balancing conservatism and accuracy is difficult; for example, sequential algorithms may underestimate wind power contributions, and convolutional algorithms ignore temporal correlations. Dynamic reliable capacity assessment is computationally complex and lacks real-time performance, making it difficult to meet the needs of large-scale power grids and multi-wind farm scenarios. Furthermore, existing models have limited adaptability and versatility, lacking a comprehensive assessment method that combines energy storage and flexibility resources. Summary of the Invention

[0006] This invention provides a method, apparatus, equipment, and medium for assessing the reliable capacity of wind power under extreme weather conditions, in order to solve the problem of poor accuracy and versatility in wind power prediction.

[0007] In a first aspect, embodiments of the present invention provide a method for assessing the reliable capacity of wind power under extreme weather conditions, including: Obtain the wind speed modal components and temperature modal components within the current rolling time window, as well as the historical power of the wind turbine. The wind speed mode component, temperature mode component, and historical power are input into a trained physical information neural network model to obtain the output prediction results of the wind turbine. The physical information neural network model is physically constrained by the wind speed power characteristic model of the wind turbine considering temperature factors and / or the output model of the wind turbine considering low temperature protection. The reliability index of the power system where the wind turbine is located is calculated based on the output prediction results, and the dynamic reliable capacity of the wind turbine within the current rolling time window is obtained by solving the problem using the equal reliability Monte Carlo method.

[0008] In one possible implementation, the wind speed-power characteristic model of a wind turbine considering temperature factors is as follows:

[0009] in, For wind power, The swept area of ​​the wind turbine. For air pressure, The wind energy utilization coefficient of the wind turbine. The molar mass of air, For wind speed, Let be the ideal gas constant. Temperature.

[0010] In one possible implementation, the wind turbine output model considering cryogenic protection is as follows:

[0011] in, For the actual output of the wind turbine unit For wind power, For temperature, The shutdown temperature threshold, To activate the temperature threshold, For runtime, The threshold for the duration of the pause. This is the startup duration threshold.

[0012] In one possible implementation, the power output prediction result of the wind turbine includes the power output probability prediction result; the physical information neural network model includes an Informer model and a probability prediction module. The wind speed mode component, temperature mode component, and historical power are input into a trained physical information neural network model to obtain the wind turbine output prediction results, including: By inputting the wind speed modal components, temperature modal components, and historical power into the Informer model, the predicted power of the wind turbine is obtained. The uncertainty of predicted power is quantified by the probability prediction module to obtain the prediction mean and prediction standard deviation, and the output probability prediction result of the wind turbine is represented by the prediction mean and prediction standard deviation.

[0013] In one possible implementation, the Informer model includes an encoder and a decoder; The encoder includes a first stack layer and a second stack layer, each stack layer including a multi-head sparse self-attention layer and a distillation layer; The decoder includes a fully connected layer.

[0014] In one possible implementation, the reliability index of the power system where the wind turbine is located is calculated based on the output prediction results, and the dynamic reliable capacity of the wind turbine is solved using the equal reliability Monte Carlo method, including: Based on the output probability prediction results of wind turbine units, the non-sequential Monte Carlo simulation method is used to calculate the system reliability index. The equivalent substitution method is used to adjust the wind power output and thermal power output, and the replaced wind power output is taken as the reliable capacity of the wind turbine.

[0015] In one possible implementation, obtaining the wind speed modal components and temperature modal components includes: Get the wind speed and temperature sequences within the current scrolling time window; Successive variational mode decomposition was performed on the wind speed sequence to obtain the wind speed mode components; Successive variational mode decomposition of the temperature series is performed to obtain the temperature mode components.

[0016] Secondly, embodiments of the present invention provide a wind power reliable capacity assessment device under extreme weather conditions, comprising: The acquisition module is used to acquire the wind speed modal components and temperature modal components within the current rolling time window, as well as the historical power of the wind turbine. The prediction module is used to input wind speed mode components, temperature mode components and historical power into a trained physical information neural network model to obtain the output prediction results of the wind turbine; wherein, the physical information neural network model is physically constrained by the wind speed power characteristic model of the wind turbine considering temperature factors and / or the output model of the wind turbine considering low temperature protection. The calculation module is used to calculate the reliability index of the power system where the wind turbine is located based on the output prediction results, and to solve it using the equal reliability Monte Carlo method to obtain the dynamic reliable capacity of the wind turbine within the current rolling time window.

[0017] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0018] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect or any possible implementation thereof.

[0019] The wind power reliable capacity assessment method, device, equipment, and medium provided in this invention under extreme weather conditions, by using a wind speed power characteristic model of wind turbine considering temperature factors and / or a wind turbine output model considering low temperature protection as physical constraints, can accurately describe the power generation characteristics and state of wind turbines at low temperatures, improving the accuracy of wind turbine power prediction in low-temperature scenarios and extreme low-temperature conditions. Based on the Monte Carlo method of equal reliability to solve the reliable capacity, it not only focuses on the output trend, but also emphasizes the contribution of wind power output to system reliability and dispatch operation. By integrating prediction information, system state, and risk constraints, it improves the accuracy and versatility of wind power prediction, while providing a physical quantity expression for the "dispatchability metric" of wind power. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the implementation of the wind power reliable capacity assessment method under extreme weather conditions provided in this embodiment of the invention. Figure 2 This is a system functional implementation block diagram of the wind power reliable capacity assessment method under extreme weather conditions provided in this embodiment of the invention; Figure 3 This is a flowchart illustrating the implementation of wind power reliable capacity assessment provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the wind power reliable capacity assessment device under extreme weather conditions provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] See Figure 1 The document illustrates a flowchart of the implementation of the wind power reliable capacity assessment method under extreme weather conditions provided by an embodiment of the present invention, which is described in detail below: Step 101: Obtain the wind speed modal component and temperature modal component within the current rolling time window, as well as the historical power of the wind turbine.

[0023] In this embodiment, the entire evaluation process is executed using a rolling time window approach. At regular intervals, the prediction time window slides forward, and modeling and simulation predictions are re-performed based on the wind speed and temperature modal components and the historical power of the wind turbines within the current rolling time window, dynamically updating the reliable capacity assessment results. This method can output a reliable capacity curve that changes over time, suitable for system scheduling and backup strategy formulation in scenarios with a high proportion of wind power integration.

[0024] Step 102: Input the wind speed mode component, temperature mode component, and historical power into the trained physical information neural network model to obtain the output prediction result of the wind turbine; wherein, the physical information neural network model uses the wind speed power characteristic model of the wind turbine considering the temperature factor and / or the output model of the wind turbine considering low temperature protection as physical constraints.

[0025] In this embodiment, based on wind turbine power and numerical weather forecast data, low-temperature wind speed and power characteristics and low-temperature protection strategy constraints of wind turbines are introduced into the power prediction process to describe the power generation state of wind turbines at low temperatures. This allows the model to learn the power output characteristics of wind turbines under extreme low temperatures, improving the accuracy of wind turbine power output probability prediction under extreme low-temperature conditions. Then, combined with an Informer-based physical information neural network model, the accuracy of wind turbine power probability prediction in low-temperature scenarios is further improved through physical information constraints.

[0026] like Figure 2 As shown, the architecture for predicting wind turbine output in this embodiment mainly consists of a wind speed prediction module (a), a temperature prediction module (b), physical information constraints (d), and an Informer power prediction module (e). The modal components of predicted wind speed and temperature, along with the historical power of the wind turbine, are used as feature inputs to the power prediction module. The model training is guided by the low-temperature wind power characteristics of the wind turbine and low-temperature protection constraints, ultimately enabling the prediction of wind power output in low-temperature scenarios.

[0027] Step 103: Calculate the reliability index of the power system where the wind turbine is located based on the output prediction results, and solve it using the equal reliability Monte Carlo method to obtain the dynamic reliable capacity of the wind turbine within the current rolling time window.

[0028] In this embodiment, to more accurately assess the equivalent capacity of wind power in the power system, a dynamic reliable capacity assessment process based on equal reliability Monte Carlo is constructed, such as... Figure 3As shown in the diagram, this process, based on the probability prediction results of wind turbine output, performs power flow calculations on power systems containing wind power and calculates reliability indices. It replaces the wind turbine capacity with that of conventional thermal power units and uses the equal-reliability Monte Carlo method to solve for the dynamic reliable capacity of the wind turbines. It considers real-time adjustments based on weather and equipment status, focusing not only on output trends but also on the contribution of wind power output to system reliability and dispatch operation.

[0029] This invention uses wind speed-power characteristic models of wind turbines that consider temperature factors and / or output models of wind turbines that consider low-temperature protection as physical constraints. This can accurately describe the power generation characteristics and state of wind turbines at low temperatures, improving the accuracy of wind turbine power prediction in low-temperature scenarios and extreme low-temperature conditions. Based on the Monte Carlo method of equal reliability to solve for reliable capacity, it not only focuses on the output trend but also emphasizes the contribution of wind power output to system reliability and dispatch operation. By integrating prediction information, system state, and risk constraints, and fully combining numerical weather prediction (NWP), equipment state modeling, and system reliability simulation, it not only improves the accuracy and versatility of wind power prediction but also provides a physical quantity expression for the "dispatchability metric" of wind power, providing effective theoretical support for the dispatch optimization of power systems containing wind power under extreme weather conditions.

[0030] In one possible implementation, the wind speed-power characteristic model of a wind turbine considering temperature factors is as follows:

[0031] in, For wind power, The swept area of ​​the wind turbine. For air pressure, The wind energy utilization coefficient of the wind turbine. The molar mass of air, For wind speed, Let be the ideal gas constant. Temperature.

[0032] In this embodiment, the derivation process of the wind speed-power characteristic model of the wind turbine considering temperature factors is as follows: Wind speed-power characteristics describe the output power of a wind turbine at different wind speeds, reflecting the nonlinear relationship between wind speed and power. According to Betz's law, wind power is proportional to the cube of the wind speed, with the following relationship: (1) Where P represents wind power. Let A be the air density, Cp be the wind turbine's swept area, and V be the wind speed. As the temperature decreases, the air density... This will also change accordingly, thus affecting the relationship between wind power and wind speed. The relationship between air temperature and air density follows the formula: (2) Where B is the air pressure, M is the molar mass of air, R is the ideal gas constant, and T is the Kelvin temperature.

[0033] From the above formulas (1) and (2), we can obtain the wind speed power characteristic model considering temperature factors: (3) In one possible implementation, the wind turbine output model considering cryogenic protection is as follows:

[0034] in, For the actual output of the wind turbine unit For wind power, For temperature, The shutdown temperature threshold, To activate the temperature threshold, For runtime, The threshold for the duration of the pause. This is the startup duration threshold.

[0035] In this embodiment, the low-temperature protection for wind turbines is a safety measure designed to ensure the normal operation of wind turbines in low-temperature environments. When the temperature falls below a set threshold for a certain period of time, the controller will activate corresponding protection measures according to a preset program to shut down the turbine. When the temperature recovers above the threshold for a certain period of time, the wind turbine will restart.

[0036] Most of these turbines employ a protection strategy that shuts down when the ambient temperature drops below -30°C. Turbines with shutdown thresholds of -35°C and -40°C are classified as low-temperature models and are configured in some wind farms. The output of wind turbines with low-temperature protection can be described by the following formula. (4) Where P * The actual output of the unit is T, where T is the air temperature. stop and t stop These represent the shutdown threshold and duration, respectively, T start and t start These are the start threshold and duration, respectively.

[0037] In one possible implementation, obtaining the wind speed modal components and temperature modal components includes: Get the wind speed and temperature sequences within the current scrolling time window; Successive variational mode decomposition was performed on the wind speed sequence to obtain the wind speed mode components; Successive variational mode decomposition of the temperature series is performed to obtain the temperature mode components.

[0038] In this embodiment, wind speed and temperature prediction differ from power prediction. Wind power changes are closely related to wind speed, therefore, wind speed is suitable as the input feature for prediction. Although it's difficult to find strongly correlated features to assist in wind speed and temperature prediction, their changes exhibit certain trends and fluctuations. Therefore, mode decomposition can be performed, and the mode components can be used as features for prediction. To this end, an adaptive mode decomposition method based on successive variational mode decomposition (SVMD) is proposed.

[0039] SVMD employs a sequential decomposition method for the signal, a process that continues until all IMFs are extracted, or the reconstruction error (the error between the input signal and the sum of the modal components) is less than a threshold. The time-domain signal to be decomposed... Decomposed into the first Modal components and residual components ,Right now: (5) residual components in the formula It consists of two parts, including the initial product obtained from the decomposition. Modality and unprocessed parts When performing sequential decomposition, the iterative expression for SVMD is: (6) In the formula For the center frequency, The parameter representing model compactness and data fidelity is set to 20000 here. and The update equation is: (7) (8) In the formula, To update the parameters, the double ascent method is used to obtain the corrected expression: (9) The modal components of wind speed and temperature are extracted step by step using SVMD until the reconstruction error is less than the set threshold.

[0040] In one possible implementation, the Informer model includes an encoder and a decoder; The encoder includes a first stack layer and a second stack layer, each stack layer including a multi-head sparse self-attention layer and a distillation layer; The decoder includes a fully connected layer.

[0041] In this embodiment, as Figure 2 The Informer model's main structure consists of an encoder and a decoder. Features are input into the encoder, and intermediate features are extracted through two stack layers. Each stack layer consists of two parts: an encoding layer and a distillation layer. The encoding layer is mainly a multi-head sparse self-attention layer. The calculation formula for the multi-head sparse self-attention mechanism is as follows: (10) In the formula , and It is a matrix obtained by different linear transformations of the input variables. Through The probability sparse matrix obtained; It is the dimension of the query vector and the key vector. It is a normalized activation function, assuming for The If so, then the first indivual The attention value is: (11) No. indivual The sparsity can be expressed as: (12) The distillation layer consists of a one-dimensional convolutional layer and a pooling layer with a stride of 2. It reduces the size of the network parameters through the "distillation" mechanism, assigns higher weights to the dominant features, and generates a concentrated feature map in the next layer. From the moment Layer to The distillation operation of the layers can be described as follows: (13) In the formula, For self-attention modules with multi-head sparse self-attention, This represents a one-dimensional convolution operation. For activation function, For max pooling operations, the "distillation" operation reduces memory utilization by using a pooling layer with a step size of 2.

[0042] The decoder takes as input a set of implicit intermediate feature sequences about bearing temperature output by the encoder and a combination of 0 values, where 0 is used to place the temperature to be predicted. The decoder uses the intermediate feature sequences output by the encoder to perform multi-head attention operations, adjusts the dimension of the output data through a fully connected layer, and uses a mask mechanism to prevent data leakage on the time scale, and finally obtains the prediction result.

[0043] In one possible implementation, the power output prediction result of the wind turbine includes the power output probability prediction result; the physical information neural network model includes an Informer model and a probability prediction module. The wind speed mode component, temperature mode component, and historical power are input into a trained physical information neural network model to obtain the wind turbine output prediction results, including: By inputting the wind speed modal components, temperature modal components, and historical power into the Informer model, the predicted power of the wind turbine is obtained. The uncertainty of predicted power is quantified by the probability prediction module to obtain the prediction mean and prediction standard deviation, and the output probability prediction result of the wind turbine is represented by the prediction mean and prediction standard deviation.

[0044] In this embodiment, a probabilistic prediction module is introduced based on the Informer model to model the power prediction result as a probability distribution rather than a single predicted value. Specifically, this is achieved by introducing the quantification of uncertainty, representing the prediction result as a mean and variance.

[0045] (14) Where μ is the predicted mean and σ is the predicted standard deviation. It is a random variable that follows a standard normal distribution.

[0046] In one possible implementation, the reliability index of the power system where the wind turbine is located is calculated based on the output prediction results, and the dynamic reliable capacity of the wind turbine is solved using the equal reliability Monte Carlo method, including: Based on the output probability prediction results of wind turbine units, the non-sequential Monte Carlo simulation method is used to calculate the system reliability index. The equivalent substitution method is used to adjust the wind power output and thermal power output, and the replaced wind power output is taken as the reliable capacity of the wind turbine.

[0047] In this embodiment, firstly, based on the wind power output probability prediction results obtained in the previous embodiments, the system reliability index is calculated using the non-sequential Monte Carlo simulation method. The non-sequential Monte Carlo simulation method is often called the state sampling method. This method is based on the premise that a system state is a combination of all component states, and each component state can be determined by sampling the probability of the component appearing in that state. Assume the system consists of… Composed of several components, Representative components The state of the system is then the system state. It depends on the state combination of the system components. The probabilistic characteristic of each component can be represented by a state combination. Describing it as a uniform distribution, assuming each component has only two states: fault and operation, let Indicator element The invalidity is determined by extracting a range of values. Uniformly distributed random numbers Then we have: (15) Extract random numbers Using equation (15), the state of a system can be determined. Repeat the above process. Then, a containing A set of system state samples The non-sequential Monte Carlo method uses the sample mean as an approximate estimate of the expected value of the reliability index, calculated as follows: (16) In the formula Indicates sample size; Represents the sample function; Represents a random function The sample mean, When we choose different test functions that we are interested in This represents different reliability metrics; here, we choose the probability of time without power (LOLP).

[0048] Then, to calculate the reliable capacity of wind power, an equivalent substitution method is used to gradually reduce wind power output while increasing conventional thermal power unit output, constructing substitution scenarios. For each substitution scenario, the system reliability is recalculated to obtain a new LOLP (Reliable Logic Response Level) index. When the new LOLP index is close to the LOLP under the original system, the currently substituted wind power output is considered the reliable capacity at that moment. If the difference between the new LOLP index and the LOLP under the original system exceeds the set tolerance range, the wind power to thermal power output ratio is adjusted until the equivalent reliability condition is met.

[0049] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0050] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0051] Figure 4 A schematic diagram of the wind power reliable capacity assessment device under extreme weather conditions provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 4 As shown, the wind power reliable capacity assessment device 4 under extreme weather conditions includes: The acquisition module 41 is used to acquire the wind speed mode component and temperature mode component within the current rolling time window, as well as the historical power of the wind turbine. Prediction module 42 is used to input wind speed mode components, temperature mode components and historical power into a trained physical information neural network model to obtain the output prediction result of the wind turbine; wherein, the physical information neural network model is physically constrained by the wind speed power characteristic model of the wind turbine considering temperature factors and / or the output model of the wind turbine considering low temperature protection. The calculation module 43 is used to calculate the reliability index of the power system where the wind turbine is located based on the output prediction results, and to solve it using the equal reliability Monte Carlo method to obtain the dynamic reliable capacity of the wind turbine within the current rolling time window.

[0052] In one possible implementation, the wind speed-power characteristic model of a wind turbine considering temperature factors is as follows:

[0053] in, For wind power, The swept area of ​​the wind turbine. For air pressure, The wind energy utilization coefficient of the wind turbine. The molar mass of air, For wind speed, Let be the ideal gas constant. Temperature.

[0054] In one possible implementation, the wind turbine output model considering cryogenic protection is as follows:

[0055] in, For the actual output of the wind turbine unit For wind power, For temperature, The shutdown temperature threshold, To activate the temperature threshold, For runtime, The threshold for the duration of the pause. This is the startup duration threshold.

[0056] In one possible implementation, the power output prediction result of the wind turbine includes the power output probability prediction result; the physical information neural network model includes an Informer model and a probability prediction module. Prediction module 42 is specifically used for: By inputting the wind speed modal components, temperature modal components, and historical power into the Informer model, the predicted power of the wind turbine is obtained. The uncertainty of predicted power is quantified by the probability prediction module to obtain the prediction mean and prediction standard deviation, and the output probability prediction result of the wind turbine is represented by the prediction mean and prediction standard deviation.

[0057] In one possible implementation, the Informer model includes an encoder and a decoder; The encoder includes a first stack layer and a second stack layer, each stack layer including a multi-head sparse self-attention layer and a distillation layer; The decoder includes a fully connected layer.

[0058] In one possible implementation, the computation module 43 is specifically used for: Based on the output probability prediction results of wind turbine units, the non-sequential Monte Carlo simulation method is used to calculate the system reliability index. The equivalent substitution method is used to adjust the wind power output and thermal power output, and the replaced wind power output is taken as the reliable capacity of the wind turbine.

[0059] In one possible implementation, module 41 is specifically used for: Get the wind speed and temperature sequences within the current scrolling time window; Successive variational mode decomposition was performed on the wind speed sequence to obtain the wind speed mode components; Successive variational mode decomposition of the temperature series is performed to obtain the temperature mode components.

[0060] This invention uses wind speed-power characteristic models of wind turbines that consider temperature factors and / or output models of wind turbines that consider low-temperature protection as physical constraints. This can accurately describe the power generation characteristics and state of wind turbines at low temperatures, improving the accuracy of wind turbine power prediction in low-temperature scenarios and extreme low-temperature conditions. Based on the Monte Carlo method of equal reliability to solve for reliable capacity, it not only focuses on the output trend but also emphasizes the contribution of wind power output to system reliability and dispatch operation. By integrating prediction information, system state, and risk constraints, it improves the accuracy and versatility of wind power prediction while providing a physical quantity expression for "dispatchability measurement" of wind power.

[0061] Figure 5This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 5 As shown, the electronic device 5 of this embodiment includes a processor 50 and a memory 51. The memory 51 stores a computer program 52. When the processor 50 executes the computer program 52, it implements the steps in the various method embodiments described above. Alternatively, when the processor 50 executes the computer program 52, it implements the functions of each module / unit in the various device embodiments described above.

[0062] For example, computer program 52 may be divided into one or more modules / units, which are stored in memory 51 and executed by processor 50 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 52 in electronic device 5.

[0063] Electronic device 5 may include, but is not limited to, processor 50 and memory 51. Those skilled in the art will understand that... Figure 5 This is merely an example of electronic device 5 and does not constitute a limitation on electronic device 5. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 5 may also include input / output devices, network access devices, buses, etc.

[0064] The processor 50 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0065] The memory 51 can be an internal storage unit of the electronic device 5, such as a hard disk or RAM. The memory 51 can also be an external storage device of the electronic device 5, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 51 can include both internal and external storage units of the electronic device 5. The memory 51 is used to store the computer program 52 and other programs and data required by the electronic device 5. The memory 51 can also be used to temporarily store data that has been output or will be output.

[0066] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0067] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0068] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0069] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0070] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0071] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for assessing the reliable capacity of wind power under extreme weather conditions, characterized in that, include: Obtain the wind speed modal components and temperature modal components within the current rolling time window, as well as the historical power of the wind turbine. The wind speed mode component, the temperature mode component, and the historical power are input into a trained physical information neural network model to obtain the output prediction result of the wind turbine; wherein, the physical information neural network model is physically constrained by a wind speed power characteristic model of the wind turbine considering temperature factors and / or a wind turbine output model considering low temperature protection. Based on the power output prediction results, the reliability index of the power system where the wind turbine is located is calculated, and the dynamic reliable capacity of the wind turbine within the current rolling time window is obtained by solving the problem using the equal reliability Monte Carlo method.

2. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 1, characterized in that, The wind speed-power characteristic model of the wind turbine considering temperature factors is as follows: in, For wind power, The swept area of ​​the wind turbine. For air pressure, The wind energy utilization coefficient of the wind turbine. The molar mass of air, For wind speed, Let be the ideal gas constant. Temperature.

3. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 1, characterized in that, The wind turbine output model considering low-temperature protection is as follows: in, For the actual output of the wind turbine unit For wind power, For temperature, The shutdown temperature threshold, To activate the temperature threshold, For runtime, The threshold for the duration of the pause. This is the startup duration threshold.

4. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 1, characterized in that, The power output prediction result of the wind turbine includes the power output probability prediction result; the physical information neural network model includes an Informer model and a probability prediction module. The step of inputting the wind speed mode component, the temperature mode component, and the historical power into a trained physical information neural network model to obtain the power output prediction result of the wind turbine includes: The wind speed mode component, the temperature mode component, and the historical power are input into the Informer model to obtain the predicted power of the wind turbine. The probability prediction module quantifies the uncertainty of the predicted power to obtain the prediction mean and prediction standard deviation, and represents the output probability prediction result of the wind turbine based on the prediction mean and prediction standard deviation.

5. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 4, characterized in that, The Informer model includes an encoder and a decoder; The encoder includes a first stack layer and a second stack layer, each stack layer including a multi-head sparse self-attention layer and a distillation layer; The decoder includes a fully connected layer.

6. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 4, characterized in that, The calculation of the reliability index of the power system where the wind turbine is located based on the output prediction results, and the use of the equal reliability Monte Carlo method to solve for the dynamic reliable capacity of the wind turbine, includes: Based on the power output probability prediction results of the wind turbine, the system reliability index is calculated using the non-sequential Monte Carlo simulation method. The equivalent substitution method is used to adjust the wind power output and thermal power output, and the wind power output to be replaced is taken as the reliable capacity of the wind turbine.

7. The method for assessing the reliable capacity of wind power under extreme weather conditions according to claim 1, characterized in that, The acquisition of wind speed mode components and temperature mode components within the current rolling time window includes: Get the wind speed and temperature sequences within the current scrolling time window; Successive variational mode decomposition is performed on the wind speed sequence to obtain wind speed mode components; Successive variational mode decomposition is performed on the temperature sequence to obtain temperature mode components.

8. A reliable wind power capacity assessment device under extreme weather conditions, characterized in that, include: The acquisition module is used to acquire the wind speed modal components and temperature modal components within the current rolling time window, as well as the historical power of the wind turbine. The prediction module is used to input the wind speed mode component, the temperature mode component, and the historical power into a trained physical information neural network model to obtain the output prediction result of the wind turbine; wherein, the physical information neural network model is physically constrained by a wind speed power characteristic model of the wind turbine considering temperature factors and / or a wind turbine output model considering low temperature protection. The calculation module is used to calculate the reliability index of the power system where the wind turbine is located based on the output prediction results, and solve it using the equal reliability Monte Carlo method to obtain the dynamic reliable capacity of the wind turbine within the current rolling time window.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images