A wind power output sequence differentiation reduction method, system and device

Abstract

The present application relates to a kind of wind power output sequence differentiation reduction method, system and equipment, wherein the method comprises: obtaining original wind power output time sequence and standardization;The original wind power output time sequence after standardization is divided into event peak area, trend peak area, periodic peak area and background area;Different division interval is set with corresponding differentiation reduction strategy, and the reduction parameter combination of the reduction parameter in differentiation reduction strategy is extracted;Based on the dynamic penalty term of KL divergence and total utilization hour deviation of original wind power output time sequence, optimization objective function is constructed;Based on the preset optimization algorithm, optimization objective function is iteratively optimized, and the optimal reduction parameter combination is obtained;Based on the optimal reduction parameter combination, original wind power output sequence is executed differentiation reduction strategy, and the final reduced wind power output sequence is obtained.

Description

Technical Field

[0001] This invention relates to the field of wind power output sequence generation technology, and mainly to a method, system and equipment for differential reduction of wind power output sequence. Background Technology

[0002] In recent years, with the rapid growth of installed capacity of new energy sources such as wind power and photovoltaics, their inherent intermittent and fluctuating characteristics have placed higher demands on the accuracy of power system production simulation. In the dispatch and operation analysis of power systems with a high proportion of new energy sources, it is often necessary to preprocess the output sequences of wind power and photovoltaics to simulate scenarios such as wind curtailment, solar curtailment, or setting utilization hour constraints. However, existing reduction methods directly multiply by a fixed absorption rate or reduction coefficient (i.e., uniform reduction); but this reduction method uniformly reduces the peak output, completely eliminating the scenario of full-power generation, resulting in overly optimistic production simulation results that cannot truly reflect the peak-shaving pressure of the system during periods of high wind power output; at the same time, the shape of the output probability distribution is distorted, affecting subsequent statistical analysis based on distribution characteristics.

[0003] Chinese invention patent application CN119538801A discloses a method and system for wind power output reduction based on time-series point-by-point correction. This technical solution includes acquiring turbine data and SCADA data for each turbine in an existing wind farm to determine the actual power output curve of the wind farm; determining the reduction coefficient at each moment based on the actual power output curve and the annual theoretical power output curve; grouping the reduction coefficients to determine the reduction coefficient corresponding to each wind speed segment; performing function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model and distribution model; acquiring the target annual theoretical power output curve of the target wind farm; and correcting the target annual theoretical power output curve based on the reduction probability model, distribution model, and wind speed. However, the above technical solution mainly relies on grouping and reducing based on wind speed segments. The above technical solutions fail to consider the temporal structure characteristics of the power output sequence itself, such as the different requirements for reduction fidelity for peak regions with different forms, such as event-driven peaks, seasonal trend peaks, and daily cycle peaks caused by weather events. At the same time, the above technical solutions rely on historical actual operation data for statistical modeling, which is difficult to apply to wind farms in the planning stage or scenarios lacking high-quality SCADA data. In addition, the above technical solutions use a uniform reduction logic for all power outputs at the same time and wind speed, and only fit the reduction using a single wind speed function. There is no hierarchical differentiated processing mechanism, and no distinction is made between seasonal trend peaks, daily cycle peaks, and flat background areas, resulting in weak morphological fidelity.

[0004] Therefore, there is an urgent need for a wind power output sequence reduction method that can simultaneously achieve precise control of utilization hours and preservation of output timing characteristics. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a wind power output sequence differential reduction method, system, and equipment.

[0006] The technical solution of the present invention is as follows: On the one hand, this invention proposes a wind power output sequence differential reduction method, the method comprising: Obtain the original wind power output time series and normalize it; The normalized original wind power output time series is divided into event peak region, trend peak region, periodic peak region and background region; Differentiated reduction strategies are set for different division intervals, and reduction parameters in the differentiated reduction strategies are extracted to form a combination of reduction parameters; An optimization objective function is constructed based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; the optimization objective function is iteratively optimized based on a preset optimization algorithm to obtain the optimal combination of reduction parameters; A differentiated reduction strategy is applied to the original wind power output sequence based on the optimal reduction parameter combination to obtain the final reduced wind power output sequence.

[0007] Preferably, the specific steps of the per-unit standardization are as follows: Obtain the installed capacity of the wind farm; Divide the original wind power output time series by the installed capacity of the wind farm to obtain the normalized original wind power output time series.

[0008] Preferably, the original wind power output time series is decomposed into trend term, seasonal term and residual term using the time series decomposition method; Set the amplitude and time thresholds for the trend, seasonal, and residual items respectively; The sequence intervals in the residual term that satisfy both the residual amplitude threshold and the residual time threshold are divided into event peak regions; The sequence intervals in the trend term that satisfy both the trend amplitude threshold and the trend time threshold are divided into trend peak regions. The sequence intervals in the seasonal term that satisfy both the seasonal amplitude threshold and the seasonal time threshold are divided into periodic peak regions. The original wind power output time series, excluding the event peak area, trend peak area, and cycle peak area, is divided into the background area.

[0009] Preferably, a strategy of completely preserving the event peak region is adopted.

[0010] Preferably, a nonlinear reduction strategy is adopted for the trend peak area, specifically by using a power function to reduce the trend peak area to obtain the wind power output sequence of the reduced trend peak area.

[0011] Preferably, a probability reduction strategy is adopted for the periodic peak region, specifically: Set the retention probability; Generate random numbers within a preset range for all periodic peak regions; If the random number generated in the current period peak area is less than the retention probability, the time series data of the current period peak area is completely retained; if the random number generated in the current period peak area is greater than the retention probability, the time series data of the current period peak area is multiplied by the period peak area reduction factor to obtain the reduced period peak area wind power output sequence.

[0012] Preferably, a uniform coefficient reduction strategy is adopted for the background area, specifically: Multiply the background area reduction factor by the wind power output sequence of the background area to obtain the reduced background area wind power output sequence.

[0013] Preferably, the optimization objective function is as follows: The original wind power output sequence is reduced based on the current reduction parameter combination to obtain the reduced wind power output sequence, and the probability distribution of the reduced wind power output sequence is calculated. The KL divergence between the probability distribution of the original wind power output sequence and the probability distribution of the reduced wind power output sequence is used as the first optimization objective. Calculate the absolute deviation between the annual total utilization hours of the reduced wind power output sequence and the target annual total utilization hours; use the product of the absolute deviation and the dynamic penalty term as the second optimization objective; The optimization objective function is to minimize the sum of the first and second optimization objectives. The objective function is iteratively optimized based on a pre-defined optimization algorithm to obtain the optimal combination of reduction parameters; The dynamic penalty item is specifically as follows: If the absolute deviation is less than or equal to the preset acceptable threshold, the penalty term dynamic adjustment parameter is zero; If the absolute deviation is greater than a preset acceptable threshold, the penalty term is dynamically adjusted to a preset penalty factor.

[0014] On the other hand, the present invention also provides a wind power output sequence differential reduction system, the system comprising: The data acquisition module acquires the raw wind power output time series and standardizes it. The interval division module divides the standardized original wind power output time series into event peak region, trend peak region, periodic peak region and background region; The strategy setting module sets corresponding differentiated reduction strategies for different division intervals and extracts the reduction parameters from the differentiated reduction strategies to form a combination of reduction parameters; The parameter optimization module constructs an optimization objective function based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; it iteratively optimizes the optimization objective function based on a preset optimization algorithm to obtain the optimal combination of reduction parameters. The final reduction module applies a differentiated reduction strategy to the original wind power output sequence based on the optimal reduction parameter combination to obtain the final reduced wind power output sequence.

[0015] In another aspect, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the present invention.

[0016] The present invention has the following beneficial effects: 1. This invention divides the original power output sequence into four types of intervals—event peak region, trend peak region, periodic peak region, and background region—through time-series location pattern decomposition, comprehensively exploring the physical characteristics and fluctuation patterns of wind power output at different time scales, and improving the refinement level and analysis depth of sequence preprocessing; 2. This invention constructs differentiated reduction strategies for different intervals. Event peaks are completely preserved to maintain the integrity of extreme scenarios. Trend peaks are reduced using output-dependent nonlinear reduction to protect the peak shape. Periodic peaks are reduced using probability sampling to preserve periodicity. Background areas are reduced using a uniform coefficient to undertake the main power reduction task. This invention considers the physical meaning and shape protection requirements of various intervals, enhancing the scientificity and rationality of the reduction method. 3. This invention adaptively optimizes the reduction parameters, which can quickly converge to the optimal parameter combination. This improves the adaptability and robustness of the reduction method to different wind farm characteristics and different target hour settings. It can quickly respond to diverse constraints in production simulation calculations, thereby enhancing the flexibility of the reduction method and its adaptability to complex operating environments. Attached Figure Description

[0017] Figure 1 This is a detailed flowchart of an embodiment of the present invention; Figure 2 This is a per-unit original wind power output sequence diagram according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the trend terms after time-series decomposition according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the seasonal terms after time-series decomposition according to an embodiment of the present invention; Figure 5 These are the residual terms after time-series decomposition in this embodiment of the invention; Figure 6 This is a diagram showing the identification results of the event peak region, seasonal peak region, periodic peak region, and background region after reduction according to an embodiment of the present invention. Figure 7 This is a diagram showing the reduced wind power output sequence according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0020] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0022] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0023] Example 1: See Figure 1 This invention provides a wind power output sequence differential reduction method, the method comprising: S1. Obtain the original wind power output time series and perform per-unit normalization; the specific steps of the per-unit normalization are as follows: Obtain the installed capacity of the wind farm; Dividing the original wind power output time series by the installed capacity of the wind farm yields the normalized original wind power output time series, calculated as follows: ; In the formula, This represents the original wind power output time series after standardization; This indicates the installed capacity of the wind farm; This represents the original wind power output time series; S2. Divide the normalized original wind power output time series into event peak region, trend peak region, periodic peak region and background region; The original wind power output time series was decomposed into trend, seasonal and residual terms using the time series decomposition method. Set the amplitude and time thresholds for the trend, seasonal, and residual items respectively; The sequence intervals in the residual term that satisfy both the residual magnitude threshold and the residual time threshold are divided into event peak regions, where a single event peak region satisfies the following conditions: ; In the formula, This represents the smallest residual term within the sequence interval of a single event peak region; Indicates the residual amplitude threshold; Indicates the residual time threshold; Indicates the length of a single event peak sequence interval; The sequence intervals in the trend term that satisfy both the trend amplitude threshold and the trend time threshold are divided into trend peak regions, where a single trend peak region satisfies the following conditions: ; In the formula, It represents the smallest trend term within a single trend peak sequence interval; Indicates the threshold value for trend magnitude; Indicates the time threshold for the trend; Indicates the length of a single trend peak sequence interval; The sequence intervals in the seasonal term that satisfy both the seasonal amplitude threshold and the seasonal time threshold are divided into periodic peak regions, where a single periodic peak region satisfies the following conditions: ; In the formula, It represents the smallest trend term within a single periodic peak region sequence interval; Indicates the periodic amplitude threshold; Indicates the periodic time threshold; Indicates the length of a single periodic peak region sequence interval; The original wind power output time series, excluding the event peak area, trend peak area, and cycle peak area, is divided into the background area. S3. Set corresponding differentiated reduction strategies for different division intervals. Specifically, the event peak area adopts a complete retention strategy, the trend peak area adopts a non-linear reduction strategy, the periodic peak area adopts a probability reduction strategy, and the background area adopts a uniform coefficient reduction strategy. The reduction parameters in the differentiated reduction strategy are extracted to form a reduction parameter combination, including the trend peak area reduction coefficient, the trend peak area power exponent, the periodic peak area reduction coefficient, the periodic peak area retention probability, and the background area reduction coefficient; S31. The event peak area adopts a complete retention strategy, that is, all data within the event peak area is retained; The event peak area is triggered by sudden events such as extreme weather. It is characterized by its extreme nature, suddenness, and scarcity throughout the year. It is a necessary boundary condition for power system safety verification. Reducing it would cause the production simulation to be unable to identify the system's operational risks under extreme scenarios. Therefore, a strategy of full retention is adopted. S32. A non-linear reduction strategy is adopted for the trend peak area; The trend peak area is the main contributor to the annual electricity output. Its peak height is a key indicator of the wind farm's power generation capacity. The rising and falling sections have little impact on peak regulation, so non-linear reduction is adopted. The high output points at the peak are retained in a high proportion to maintain the shape, while the low output points bear more reduction to reduce the electricity output, while maintaining a continuous and smooth climbing process. The trend peak area is reduced by a power function to obtain the wind power output sequence of the trend peak area. When the wind power output is larger, the minimum reduction coefficient of the trend peak area tends to 1. When the wind power output is smaller, the minimum reduction coefficient of the trend peak area tends to the set minimum reduction coefficient. The calculation method for the reduced trend peak area wind power output sequence is as follows: ; In the formula, This represents the wind power output sequence in the peak trend area after reduction. Indicates the reduction factor for the trend peak area; Indicates the trend peak area; Indicates the power index of the trend peak area; S33. A probability reduction strategy is adopted for the periodic peak region; Periodic peak areas exhibit regular and recurring fluctuations. Their statistical periodicity is crucial for subsequent energy storage configuration and intraday scheduling. However, individual periodic peak areas are similar in form and can be substituted for each other. Therefore, probability sampling reduction is adopted to retain complete peak area individuals with a certain probability, thereby reducing electricity consumption while maintaining the statistical characteristics of intraday fluctuations. Set the retention probability; Generate random numbers within a preset range for all periodic peak regions; If the random number generated in the current period peak area is less than the retention probability, the time series data of the current period peak area is fully retained; if the random number generated in the current period peak area is greater than the retention probability, the time series data of the current period peak area is multiplied by the period peak area reduction factor to obtain the reduced period peak area wind power output sequence. The calculation method is as follows: ; In the formula, This represents the wind power output sequence during the peak period after reduction. Indicates the probability of retention; Indicates the periodic peak region; Indicates the periodic peak reduction factor; S34. A uniform coefficient reduction strategy is adopted for the background area; The background area fluctuates gently and does not carry key information such as extreme scenarios, trends, or periods. Therefore, a uniform coefficient reduction is adopted to undertake the main task of power reduction. Multiply the background area reduction factor by the wind power output sequence of the background area to obtain the reduced wind power output sequence of the background area. The calculation method is as follows: ; In the formula, This represents the wind power output sequence in the background area after reduction; Indicates the background area reduction factor; Indicates the background area; S4. Construct an optimization objective function based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; S41. Based on the current reduction parameter combination, reduce the original wind power output sequence to obtain the reduced wind power output sequence, and calculate the probability distribution of the reduced wind power output sequence. In this embodiment, the probability distribution of wind power output can be calculated using discrete histograms and kernel density estimation. The discrete histogram is specifically created by dividing the power output range into several equally spaced boxes, and statistically analyzing the proportion of wind power output data falling into each box to obtain a probability distribution. The kernel density estimation specifically involves fitting a continuous probability density function from the wind power output data using a Gaussian kernel smoothing method to obtain a probability distribution; The KL divergence between the probability distribution of the original wind power output sequence and the probability distribution of the reduced wind power output sequence is used as the first optimization objective. Calculate the absolute deviation between the annual total utilization hours of the reduced wind power output sequence and the target annual total utilization hours; use the product of the absolute deviation and the dynamic penalty term as the second optimization objective; The objective function is to minimize the sum of the first and second optimization objectives, and it is calculated as follows: ; In the formula, This represents the objective function to be optimized. Describes the minimum value function; This represents the probability distribution of the original wind power output sequence; This represents the probability distribution of the reduced wind power output sequence; Indicates a dynamic penalty item; This represents the total number of utilization hours for the target year; This represents the total annual utilization hours of the wind power output sequence after reduction. The objective function is iteratively optimized based on a pre-defined optimization algorithm to obtain the optimal combination of reduction parameters; S52, the dynamic penalty term is specifically as follows: If the absolute deviation is less than or equal to the preset acceptable threshold, the penalty term dynamic adjustment parameter is zero; If the absolute deviation is greater than a preset acceptable threshold, the penalty term is dynamically adjusted to a preset penalty factor, specifically expressed as follows: ; In the formula, This represents the preset penalty factor; This indicates the preset acceptable threshold. S5. Based on the preset optimization algorithm, iteratively optimize the objective function to obtain the optimal combination of reduction parameters; In this embodiment, the particle swarm optimization algorithm is used to solve the optimization objective function; S6. Based on the optimal reduction parameter combination, a differentiated reduction strategy is applied to the original wind power output sequence to obtain the final reduced wind power output sequence. S7. Compare the effects of the present invention and the traditional reduction method from the dimensions of KL divergence, target utilization hours (taken as 2094.4009), and high output scene retention rate. The traditional reduction method is to uniformly multiply the original output sequence by the same reduction coefficient. The comparison of the reduction results is shown in Table 1. Table 1 Comparison of Reduction Results

[0024] In Table 1, N represents the per-unit value of the reduced wind power output time series. Based on the above results, it can be seen that, using the method proposed in this invention, compared with the traditional reduction method, the KL divergence is close to 0, which means that it can be regarded as the same distribution as the original output series. At the same time, using the method proposed in this invention, the utilization hours of the reduced output series can also be accurately controlled to the target value. However, by comparing the retention rate of high-output scenarios, it can be seen that this invention has a better retention effect on high-output scenarios compared with the traditional reduction method. In a specific embodiment, taking the annual raw wind power output data of a wind farm as an example, the raw output data is normalized to per-unit, such as... Figure 2 As shown; Standardized data is decomposed into trend, seasonal, and residual terms, such as Figure 3 , Figure 4 and Figure 5 As shown; Set the amplitude threshold and time threshold for each decomposition item, for example: The trend amplitude threshold is set at 0.4314, which is the 80th percentile of the trend items in descending order; the trend time threshold is set at 6. The seasonal amplitude threshold was set to 1.5 times the seasonal series standard deviation of 0.0555, and the seasonal time threshold was set to 4. The residual amplitude threshold is set to 3 times the sum of the residual term sequence standard deviation and the residual term mean, which is 0.6103. The residual time threshold is set to 2. The obtained event peak region, seasonal peak region, periodic peak region, and background region, such as Figure 6 As shown; The optimal combination of reduction parameters obtained after solving the objective function is as follows: the trend peak region reduction coefficient is 0.7161, the trend peak region power exponent is 3, the periodic peak region reduction coefficient is 0.7159, the periodic peak region retention probability is 61%, and the background region reduction coefficient is 0.7152. The optimal combination of reduction parameters is used to perform the final reduction on different partitioned intervals, such as... Figure 7 As shown.

[0025] Example 2: This embodiment provides a wind power output sequence differential reduction system, the system comprising: The data acquisition module acquires the raw wind power output time series and standardizes it. The interval division module divides the standardized original wind power output time series into event peak region, trend peak region, periodic peak region and background region; The strategy setting module sets corresponding differentiated reduction strategies for different division intervals and extracts the reduction parameters from the differentiated reduction strategies to form a combination of reduction parameters; The parameter optimization module constructs an optimization objective function based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; it iteratively optimizes the optimization objective function based on a preset optimization algorithm to obtain the optimal combination of reduction parameters. The final reduction module applies a differentiated reduction strategy to the original wind power output sequence based on the optimal reduction parameter combination to obtain the final reduced wind power output sequence.

[0026] Example 3: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a wind power output sequence differential reduction method as described in any one of Embodiment 1.

[0027] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0028] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0029] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0030] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0031] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for differentiated reduction of wind power output sequence, characterized in that, The method includes: Obtain the original wind power output time series and normalize it; The normalized original wind power output time series is divided into event peak region, trend peak region, periodic peak region and background region; Differentiated reduction strategies are set for different division intervals, and reduction parameters in the differentiated reduction strategies are extracted to form a combination of reduction parameters; An optimization objective function is constructed based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; the optimization objective function is iteratively optimized based on a preset optimization algorithm to obtain the optimal combination of reduction parameters; A differentiated reduction strategy is applied to the original wind power output sequence based on the optimal reduction parameter combination to obtain the final reduced wind power output sequence.

2. The wind power output sequence differential reduction method according to claim 1, characterized in that, The specific steps of the per-unit standardization are as follows: Obtain the installed capacity of the wind farm; Divide the original wind power output time series by the installed capacity of the wind farm to obtain the normalized original wind power output time series.

3. The wind power output sequence differential reduction method according to claim 1, characterized in that, The original wind power output time series was decomposed into trend, seasonal and residual terms using the time series decomposition method. Set the amplitude and time thresholds for the trend, seasonal, and residual items respectively; The sequence intervals in the residual term that satisfy both the residual amplitude threshold and the residual time threshold are divided into event peak regions; The sequence intervals in the trend term that satisfy both the trend amplitude threshold and the trend time threshold are divided into trend peak regions. The sequence intervals in the seasonal term that satisfy both the seasonal amplitude threshold and the seasonal time threshold are divided into periodic peak regions. The original wind power output time series, excluding the event peak area, trend peak area, and cycle peak area, is divided into the background area.

4. The wind power output sequence differential reduction method according to claim 1, characterized in that, A strategy of complete preservation is adopted for the event peak area.

5. The wind power output sequence differential reduction method according to claim 1, characterized in that, A nonlinear reduction strategy is adopted for the trend peak area, specifically by using a power function to reduce the trend peak area, resulting in the wind power output sequence of the reduced trend peak area.

6. The wind power output sequence differential reduction method according to claim 1, characterized in that, A probability reduction strategy is adopted for the periodic peak region, specifically: Set the retention probability; Generate random numbers within a preset range for all periodic peak regions; If the random number generated in the current period peak area is less than the retention probability, the time series data of the current period peak area is completely retained; if the random number generated in the current period peak area is greater than the retention probability, the time series data of the current period peak area is multiplied by the period peak area reduction factor to obtain the reduced period peak area wind power output sequence.

7. The wind power output sequence differential reduction method according to claim 1, characterized in that, A uniform coefficient reduction strategy is adopted for the background area, specifically: Multiply the background area reduction factor by the wind power output sequence of the background area to obtain the reduced background area wind power output sequence.

8. The wind power output sequence differential reduction method according to claim 1, characterized in that, The optimization objective function is specifically as follows: The original wind power output sequence is reduced based on the current reduction parameter combination to obtain the reduced wind power output sequence, and the probability distribution of the reduced wind power output sequence is calculated. The KL divergence between the probability distribution of the original wind power output sequence and the probability distribution of the reduced wind power output sequence is used as the first optimization objective. Calculate the absolute deviation between the annual total utilization hours of the reduced wind power output sequence and the target annual total utilization hours; use the product of the absolute deviation and the dynamic penalty term as the second optimization objective; The optimization objective function is to minimize the sum of the first and second optimization objectives. The objective function is iteratively optimized based on a pre-defined optimization algorithm to obtain the optimal combination of reduction parameters; The dynamic penalty item is specifically as follows: If the absolute deviation is less than or equal to the preset acceptable threshold, the penalty term dynamic adjustment parameter is zero; If the absolute deviation is greater than a preset acceptable threshold, the penalty term is dynamically adjusted to a preset penalty factor.

9. A wind power output sequence differential reduction system, characterized in that, The system includes: The data acquisition module acquires the raw wind power output time series and standardizes it. The interval division module divides the standardized original wind power output time series into event peak region, trend peak region, periodic peak region and background region; The strategy setting module sets corresponding differentiated reduction strategies for different division intervals and extracts the reduction parameters from the differentiated reduction strategies to form a combination of reduction parameters; The parameter optimization module constructs an optimization objective function based on the KL divergence of the original wind power output time series and the dynamic penalty term of the total utilization hours deviation; it iteratively optimizes the optimization objective function based on a preset optimization algorithm to obtain the optimal combination of reduction parameters. The final reduction module applies a differentiated reduction strategy to the original wind power output sequence based on the optimal reduction parameter combination to obtain the final reduced wind power output sequence.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.

Images