Solar power generation adaptive delivery method and system based on prediction model

Abstract

This invention discloses a predictive model-based adaptive solar power transmission method and system, relating to the field of data processing technology. The method includes: acquiring a transmission period to be managed, acquiring historical power transmission data sequences, and acquiring the average power transmission change rate; selecting a fluctuation threshold based on the transmission period to be managed, and acquiring the over-limit coverage rate; if the over-limit coverage rate does not exceed a preset threshold, acquiring a first target transmission data sequence for the transmission period to be managed based on the predictive model and the average power transmission change rate; if the over-limit coverage rate exceeds the preset threshold, acquiring a correction ratio based on the historical power transmission data sequence, and acquiring a corrected power transmission change rate based on the average power transmission change rate and the correction ratio, and acquiring a second target transmission data sequence for the transmission period to be managed based on the predictive model and the corrected power transmission change rate. This invention has the advantages of high prediction efficiency, reliable prediction, and good transmission management effect.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to an adaptive solar power transmission method and system based on a predictive model. Background Technology

[0002] In solar power generation systems, the significant intermittency, volatility, and uncontrollability of solar energy resources cause the power output to fluctuate frequently with weather conditions (such as cloud cover and changes in solar radiation intensity) and environmental factors (such as temperature and seasonal changes). This instability makes the power generation and transmission process uncontrollable and unable to achieve adaptive adjustment, posing a severe challenge to power transmission management.

[0003] Specifically, existing power transmission methods employ fixed threshold control, which cannot adapt to seasonal and weather pattern changes. Furthermore, they fail to adequately consider the inherent stability defects of historical data sequences. Especially when historical data fluctuates frequently and significantly, directly applying the average rate amplifies prediction errors because the average rate only reflects the overall trend and ignores the impact of local abrupt changes (such as sudden increases or decreases). This causes the prediction results to deviate from actual power generation capacity, further leading to a mismatch between transmission plans and actual power generation, resulting in energy waste, grid frequency oscillations, power imbalances, and even equipment overload risks. Simultaneously, existing methods lack a quantitative assessment mechanism for data volatility, cannot distinguish between stable data states, and lack adaptive adjustment strategies to cope with unstable scenarios. Consequently, they cannot provide reliable transmission sequences during periods of high volatility, ultimately reducing the overall system efficiency, renewable energy utilization rate, and economic benefits. Summary of the Invention

[0004] The technical problem of this invention can be summarized as how to adaptively identify stable data states and intelligently adjust power generation and transmission strategies to ensure the generation of a relatively reliable power generation and transmission sequence under any fluctuating conditions. To address this technical problem, this invention provides a solar power adaptive transmission method and system based on a predictive model.

[0005] An adaptive solar power transmission method based on a predictive model includes: acquiring a transmission period to be managed, acquiring a historical power transmission data sequence for a preset period prior to the transmission period to be managed, and acquiring an average power transmission change rate based on the historical power transmission data sequence; selecting a fluctuation threshold based on the transmission period to be managed, acquiring the proportion of the historical power transmission data sequence that exceeds the fluctuation threshold and using it as the over-limit coverage rate; if the over-limit coverage rate does not exceed the preset threshold, acquiring a first target transmission data sequence for the transmission period to be managed based on the predictive model and the average power transmission change rate; if the over-limit coverage rate exceeds the preset threshold, acquiring a correction ratio based on the historical power transmission data sequence, acquiring a corrected power transmission change rate based on the average power transmission change rate and the correction ratio, and acquiring a second target transmission data sequence for the transmission period to be managed based on the predictive model and the corrected power transmission change rate.

[0006] Optionally, obtaining the average rate of change of power generation and transmission based on the historical power generation and transmission data sequence includes: obtaining the change in power generation and transmission at the beginning and end of the historical power generation and transmission data sequence and the historical transmission duration; and obtaining the average rate of change of power generation and transmission based on the ratio of the change in power generation and transmission at the beginning and end of the historical transmission duration.

[0007] Optionally, obtaining the first target transmission data sequence for the transmission period to be managed based on the prediction model and the average rate of change of power generation and transmission includes: obtaining multiple key time points for the transmission period to be managed based on the prediction model; using the power generation and transmission data at the end of the historical power generation and transmission data sequence as the starting power generation and transmission data for the transmission period to be managed, and obtaining the predicted power generation and transmission data corresponding to multiple key time points based on the prediction model, the average rate of change of power generation and transmission, and the starting power generation and transmission data; and forming the first pre-transmission data sequence for the transmission period to be managed based on the predicted power generation and transmission data corresponding to multiple key time points.

[0008] Optionally, selecting the fluctuation threshold based on the delivery period to be managed includes: obtaining multiple consecutive numerical intervals, each numerical interval corresponding to a different fluctuation threshold, wherein the larger the numerical interval, the smaller the fluctuation threshold; selecting the numerical interval to which the delivery period to be managed belongs based on the size of the delivery period to be managed, and obtaining the fluctuation threshold corresponding to that numerical interval.

[0009] Optionally, obtaining the proportion of data exceeding the fluctuation threshold in the historical power generation and transmission data sequence and using it as the over-limit coverage rate includes: obtaining the difference between each adjacent power generation and transmission data in the historical power generation and transmission data sequence; obtaining the number of differences exceeding the fluctuation threshold and using it as the over-limit quantity; and obtaining the proportion of the over-limit quantity to the number of differences in the historical power generation and transmission data sequence and using it as the over-limit coverage rate.

[0010] Optionally, obtaining the correction ratio based on the historical power generation and transmission data sequence includes: obtaining the total trend variables with upward and downward trends respectively based on the historical power generation and transmission data sequence, obtaining the total trend variable with the same trend as the average power generation and transmission rate of change as the target variable, and obtaining the total trend variable with the opposite trend to the average power generation and transmission rate of change as the reference variable; and obtaining the correction ratio based on the target variable and the reference variable.

[0011] Optionally, the correction ratio obtained from the target variable and the reference variable is expressed as follows: ;in, To correct the ratio, For the target variable, Used as a reference variable.

[0012] A predictive model-based adaptive solar power transmission system is also provided. The system includes: a first data processing module for acquiring the transmission period to be managed, acquiring historical power transmission data sequences for a preset time period prior to the transmission period to be managed, and acquiring the average power transmission change rate based on the historical power transmission data sequences; a second data processing module for selecting a fluctuation threshold based on the transmission period to be managed, acquiring the proportion of the historical power transmission data sequences exceeding the fluctuation threshold as the over-limit coverage rate; a first predictive transmission module for acquiring a first target transmission data sequence for the transmission period to be managed based on the predictive model and the average power transmission change rate when the over-limit coverage rate does not exceed the preset threshold; and a second predictive transmission module for acquiring a correction ratio based on the historical power transmission data sequences when the over-limit coverage rate exceeds the preset threshold, acquiring a corrected power transmission change rate based on the average power transmission change rate and the correction ratio, and acquiring a second target transmission data sequence for the transmission period to be managed based on the predictive model and the corrected power transmission change rate.

[0013] Optionally, the first predictive transmission module is further configured to: obtain multiple key time points for the transmission period to be managed based on the prediction model; use the power generation transmission data at the end of the historical power generation transmission data sequence as the starting power generation transmission data for the transmission period to be managed, and obtain the predicted power generation transmission data corresponding to multiple key time points based on the prediction model, the average power generation transmission change rate and the starting power generation transmission data; and form a first pre-transmission data sequence for the transmission period to be managed based on the predicted power generation transmission data corresponding to multiple key time points.

[0014] Optionally, the second predictive transmission module is further configured to: obtain the total trend variables with upward and downward trends respectively based on the historical power generation and transmission data sequence, and obtain the total trend variable with the same trend as the average power generation and transmission rate of change as the target variable, and obtain the total trend variable with the opposite trend to the average power generation and transmission rate of change as the reference variable; and obtain the correction ratio based on the target variable and the reference variable.

[0015] The beneficial effects of this invention are reflected in: In the entire adaptive transmission method for solar power generation based on the prediction model, firstly, the average rate of change calculated based on the first and last values ​​of the historical sequence quickly captures the overall trend characteristics, laying the foundation for initial prediction; secondly, according to the design of intelligent matching wave dynamic thresholds for the management period, the defects of fixed thresholds are overcome, achieving a dynamic balance between short-term noise tolerance and long-term sensitivity; by calculating the over-limit coverage rate of the difference between adjacent data points, the volatility intensity of historical data is accurately quantified, providing an objective standard for determining the stable state. Thus, in stable scenarios, linear extrapolation prediction is directly used to ensure efficiency, while in volatile scenarios, in-depth trend conflict analysis is initiated—separating and quantifying the opposing strength of the combined force of the same trend and the combined force of the opposite fluctuation, thereby generating a proportional correction to ensure that the average rate more accurately reflects the true trend: when the trend is dominant, the impact of sudden interference is positively weakened; when the fluctuation is dominant, the optimistic prediction bias is negatively suppressed, and finally, a smooth transmission sequence after the fluctuation noise is eliminated is output. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a schematic diagram of the steps in one implementation of the solar power adaptive transmission method based on a prediction model according to the present invention. Figure 2 This is a schematic diagram of a portion of steps S1 in the solar power adaptive transmission method based on a prediction model of the present invention; Figure 3 This is a schematic diagram of part of step S3 in the solar power adaptive transmission method based on a prediction model of the present invention; Figure 4 This is a schematic diagram of a portion of step S2 in the solar power adaptive transmission method based on a prediction model of the present invention; Figure 5 This is a schematic diagram of another part of step S2 in the solar power adaptive transmission method based on a prediction model of the present invention; Figure 6 This is a schematic diagram of part of step S4 in the adaptive transmission method for solar power generation based on a prediction model of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] like Figure 1 As shown, an adaptive solar power transmission method based on a predictive model is provided. In one embodiment, the method includes: S1. Obtain the transmission period to be managed, and obtain the historical power generation and transmission data sequence of the preset time period before the transmission period to be managed, and obtain the average power generation and transmission change rate based on the historical power generation and transmission data sequence. S2. Select the fluctuation threshold according to the transmission period to be managed, obtain the proportion of the historical power generation and transmission data sequence that exceeds the fluctuation threshold and use it as the over-limit coverage rate; S3. If the over-limit coverage rate does not exceed the preset threshold, the first target transmission data sequence for the transmission period to be managed is obtained based on the prediction model and the average power generation and transmission change rate. S4. If the over-limit coverage rate exceeds the preset threshold, the correction ratio is obtained based on the historical power generation and transmission data sequence, and the corrected power generation and transmission change rate is obtained based on the average power generation and transmission change rate and the correction ratio. The second target transmission data sequence for the transmission period to be managed is obtained based on the prediction model and the corrected power generation and transmission change rate.

[0022] In this embodiment, it should be noted that in S1, the target time period requiring transmission management is first determined as the object of subsequent prediction and control. To this end, a historical power generation and transmission data sequence of a preset length preceding the target time period is acquired. This sequence contains continuous records of past power generation changes, reflecting the behavior patterns of actual power output under the influence of weather and environmental factors. Based on this historical sequence, the average rate of change in power generation and transmission is calculated. This rate represents the overall trend of power generation change over the entire time period, and its core is a simplified representation by dividing the difference between the first and last values ​​of the sequence by the historical duration. This method avoids tedious point-to-point calculations but implicitly treats historical data as linearly changing, thus providing a basic input for the prediction model and supporting subsequent decisions on whether to directly apply this rate.

[0023] Furthermore, to illustrate with an example, suppose on a cloudy spring day, solar power generation needs to manage the transmission period from 2:00 PM to 4:00 PM. Extract historical data sequences preceding this target period, such as power generation changes over the previous 5 hours, with the sequence starting at 9:00 AM and ending at 1:00 PM. First, evaluate the difference between the first value (9:00 AM data) and the last value (1:00 PM data), then divide by the duration of these 5 hours to obtain the average rate. This simplifies the frequent fluctuations caused by cloud cover within the sequence into an overall upward or downward trend value. While this rate value ignores internal abrupt changes, it provides an initial reference benchmark for subsequent adaptive assessment of the steady state.

[0024] In S2, firstly, fluctuation thresholds are dynamically adapted to address the characteristics of different management periods: multiple consecutive time intervals are preset (e.g., less than 1 hour, 2-3 hours, and more than 3 hours), and a differentiated fluctuation judgment threshold is assigned to each interval. The design logic follows the core principle that the longer the management period, the smaller the threshold. This is because longer management periods are more sensitive to fluctuations (e.g., a full-day delivery plan in summer), and subtle changes can accumulate into significant deviations; therefore, a smaller threshold is needed to capture subtle fluctuations. Conversely, shorter periods (e.g., a 2-hour afternoon) allow for greater fluctuation tolerance, thus using a larger threshold. By matching the actual delivery period to be managed to the corresponding interval, the appropriate fluctuation threshold is automatically selected, thereby solving the problem that fixed thresholds cannot adapt to seasonal, weather, or time-of-day changes.

[0025] Furthermore, based on this dynamic threshold, the historical series is subjected to a quantitative assessment of volatility: the difference in power generation changes between adjacent time points in the series (such as the increase or decrease in power generation per minute or quarter hour) is calculated successively, and the percentage of times the difference exceeds the volatility threshold is counted to form the over-limit coverage rate. This over-limit coverage rate directly reflects the inherent volatility intensity of the historical data—for example, in a data series on a cloudy, flickering morning, cloud movement causes frequent abrupt changes in power generation between adjacent minutes, with a large number of differences exceeding the threshold, resulting in a significant increase in coverage; while in a clear, stable afternoon series, most differences are within the threshold, and the coverage rate is lower. This mechanism can objectively distinguish between stable states (low coverage) and volatile states (high coverage), providing a quantitative basis for subsequent adaptive decision-making.

[0026] In S3, when the over-limit coverage rate does not exceed a preset threshold (indicating that the historical power generation and transmission data sequence is in a stable state), prediction is made directly based on the average power generation and transmission change rate. First, the split duration is defined through the prediction model, and then the transmission period to be managed is split into multiple key time nodes with interval split durations (e.g., if the split duration is defined as 30 minutes, a 3-hour period can be divided into 6 key points every 30 minutes). These nodes represent the output target of the prediction model, and their distribution logic depends on the prediction model's determination of the total duration of the historical power generation and transmission data sequence and the total duration of the transmission period to be managed: for example, for a period of stable afternoon solar radiation decay, the key points may decrease, and the split duration will increase; for example, the split duration may not exceed the total duration of the historical power generation and transmission data sequence.

[0027] Furthermore, a linear prediction sequence is constructed using the average rate: the power generation data at the end of the historical sequence is used as the starting baseline value for the transmission period to be managed. Using this as the origin, the theoretical power generation at each key time point is calculated in conjunction with the average rate. The calculation method is as follows: for each key point, the average rate is multiplied by the time span from that point to the starting time, and the starting power generation data is superimposed. Finally, all the predicted values ​​of key points are connected in chronological order to form a smooth and continuous first target transmission data sequence. For example, in a stable sunny spring scenario, if the historical sequence shows a steady decrease in hourly power generation, the sequence generated by the prediction model will show a uniform decrease, with its slope strictly matching the average rate, ensuring that the transmission plan matches the power generation capacity and avoiding energy dispatch deviations. In summary, relying only on general trend extrapolation simplifies computational complexity and ensures the generation of highly reliable transmission schemes in low-fluctuation scenarios, thereby improving grid stability and energy utilization.

[0028] In S4, when the over-coverage rate exceeds a preset threshold (indicating that historical data is in an unstable and fluctuating state), the conflicting trends of the historical sequence are first analyzed in depth: based on the difference in power generation changes between adjacent time points, the resultant force of the trend in the same direction as the average rate (called the target variable) and the resultant force of the fluctuation in the opposite direction (called the reference variable) are calculated respectively. For example, under severe convective weather, a historical sequence at a certain time period may simultaneously include a sudden increase in power generation caused by cloud dissipation (positive trend) and a sudden decrease caused by cloud obstruction (reverse fluctuation). All positive and negative differences will be accumulated to form two quantitative results, thereby reflecting the strength of the conflict between the trend and the fluctuation.

[0029] Furthermore, a dynamic correction ratio is generated based on the relative strength of the two forces: if the combined trend force represented by the target variable is dominant (e.g., overall power generation still shows an upward trend when sunny skies turn cloudy), the correction ratio is a positive adjustment, weakening the impact of sudden fluctuations on the average rate; conversely, if the combined fluctuation force represented by the reference variable is stronger (e.g., continuous rain causes a precipitous drop), the correction ratio is a negative adjustment, significantly suppressing the optimistic prediction tendency of the average rate. Finally, the correction ratio is multiplied by the average power generation and transmission change rate to form a correction rate that eliminates fluctuation interference. Based on this average power generation and transmission change rate, the same prediction model as S3 is used to construct a power generation sequence at key time points (starting from historical end data, with the time span as the horizontal axis and the correction rate as the slope, generating linear prediction values), forming the second target transmission sequence.

[0030] For example, during the windy and unpredictable days at the transition from summer to autumn, even if the overall historical data shows a slow upward trend, the frequent gusts causing minute-level fluctuations in power generation can negatively amplify the correction ratio. The resulting second-sequence slope is gentler than the original average rate, avoiding grid power imbalances caused by blind transmission. This mechanism ensures that even in highly volatile scenarios, it can automatically weigh the trade-off between trends and disturbances, outputting feasible solutions with manageable risks.

[0031] In summary, the entire adaptive transmission method for solar power generation based on the prediction model firstly captures the overall trend characteristics by calculating the average rate of change based on the first and last values ​​of the historical sequence, laying the foundation for initial prediction. Secondly, the design of intelligent matching wave dynamic thresholds for the management period overcomes the shortcomings of fixed thresholds, achieving a dynamic balance between short-term noise tolerance and long-term sensitivity. By calculating the over-limit coverage rate of the difference between adjacent data points, the intensity of fluctuations in historical data is accurately quantified, providing an objective standard for determining the stable state. Thus, in stable scenarios, linear extrapolation prediction is directly used to ensure efficiency, while in volatile scenarios, in-depth trend conflict analysis is initiated—separating and quantifying the opposing strength of the combined force of the same trend and the combined force of the opposite fluctuation, thereby generating a proportional correction to ensure that the average rate more accurately reflects the true trend: when the trend is dominant, the impact of sudden interference is positively weakened; when fluctuations dominate, optimistic prediction bias is negatively suppressed, ultimately outputting a smooth transmission sequence after fluctuation noise is eliminated.

[0032] In one implementation, obtaining the average rate of change in power generation and transmission based on historical power generation and transmission data sequences in S1 includes: S11. Obtain the changes in power generation and transmission at the beginning and end of the historical power generation and transmission data sequence and the historical transmission duration; S12. Obtain the average rate of change in power generation and transmission based on the ratio of the change in power generation and transmission at the beginning and end of the transmission period to the historical transmission duration.

[0033] In this embodiment, it should be noted that in S11, the historical power generation and transmission data sequence is first extracted. By accurately locating the key values ​​at the start and end of the sequence, the algebraic difference between the two is calculated. This value directly reflects the net change in power generation capacity within a preset time period (which may be a positive increase or a negative decrease). At the same time, the total span of continuous historical transmission time covered by the sequence is strictly statistically analyzed. This duration is used as the benchmark denominator for rate calculation to ensure that the results have a uniform time dimension comparability (such as hourly rate or minute rate).

[0034] In S12, the changes in power generation and transmission at the beginning and end of S11 are divided by the historical transmission duration to obtain a single average rate of change in power generation and transmission. This operation essentially compresses all complex fluctuations in the historical sequence (including sudden drops caused by cloud cover or sudden increases caused by clear skies) into a linear expression—for example, in a historical sequence from a winter afternoon, even if multiple instances of snowmelt cause sawtooth fluctuations in power generation, the calculation result will still abstract it into a single, continuously rising or falling slope. This simplified calculation based on the beginning and end value framework significantly reduces computational complexity while providing a quantifiable and comparable trend benchmark for subsequent steady-state determination, avoiding the high computational load and real-time bottlenecks caused by point-by-point analysis.

[0035] In one implementation, S3, obtaining the first target transmission data sequence for the transmission period to be managed based on the prediction model and the average rate of change of power generation and transmission, includes: S31. Obtain multiple key time points for the transportation period to be managed based on the prediction model; S32. Take the power generation and transmission data at the end of the historical power generation and transmission data sequence as the starting power generation and transmission data of the transmission period to be managed, and obtain the predicted power generation and transmission data corresponding to multiple key time points based on the prediction model, the average power generation and transmission change rate and the starting power generation and transmission data. S33. Based on the predicted power generation and transmission data corresponding to multiple key time points, form the first pre-transmission data sequence for the transmission period to be managed.

[0036] In this embodiment, it should be noted that in S31, a refined time grid is used to divide the transmission period to be managed through a prediction model. Based on the duration characteristics of historical power generation and transmission data sequences and the span of the transmission period to be managed, the distribution locations of key time nodes are determined: for example, during the noon period when sunlight patterns are strong, the node intervals can be appropriately widened to simplify calculations; while during sunset when weather changes are frequent, the nodes are automatically densified to capture nonlinear changes. These nodes constitute the predicted time coordinate framework, and their distribution logic ensures that it reflects typical power generation patterns (such as seasonal solar radiation curves) while also adapting to management needs of different durations (such as the difference between short-term regulation and long-term planning).

[0037] In S32, the latest power generation data at the end of the historical sequence is used as the dynamic prediction baseline, which is then used as the starting power generation for the period to be managed. Based on this baseline value and the average power generation transmission change rate generated in S1, a linear trend extrapolation calculation is performed for each key time node: the average rate is multiplied by the time span of the node from the baseline, and the baseline power generation is superimposed to obtain the theoretical power generation prediction value for that node. For example, in a stable scenario with clear autumn skies, if the historical data at the end of the sequence shows that power generation is at a high level and the average rate is slowly decreasing, the predicted value will decrease uniformly over time, forming a linear evolution path that strictly follows the overall trend.

[0038] In S33, the predicted values ​​of all key nodes are connected sequentially along the time axis to construct a continuous and smooth target transmission sequence. This sequence fills the transition values ​​between nodes with linear interpolation, forming a smooth curve without jumps; furthermore, the overall slope completely inherits the linear characteristics of the average rate; and further still, the sequence's starting value is forcibly aligned with the latest power generation status. For example, in a scenario of continuous low sunlight in winter, the generated sequence represents a transmission plan that starts from the current power generation and gradually decreases according to historical trends, ensuring that under stable conditions, power generation capacity is strictly synchronized with the grid dispatch plan.

[0039] In one implementation, selecting the fluctuation threshold based on the managed delivery period in S2 includes: S21. Obtain multiple consecutive numerical intervals, each corresponding to a different fluctuation threshold. The larger the numerical interval, the smaller the fluctuation threshold. S22. Select the corresponding numerical range based on the size of the transport period to be managed, and obtain the fluctuation threshold corresponding to that numerical range.

[0040] In this embodiment, it should be noted that in S21, a preset dynamic threshold mapping rule is established: a set of continuous and non-overlapping management period length intervals (such as 0-1 hour, 1-3 hours, and more than 3 hours) are constructed, and a differentiated fluctuation judgment threshold is configured for each interval. The core principle is that the larger the upper limit of the interval length, the smaller the corresponding threshold. This is because longer management periods have a lower tolerance for fluctuations—minor changes may significantly affect the overall transportation stability through cumulative effects (such as an 8-hour cross-day transportation plan), so a smaller threshold needs to be set to capture minor fluctuations; conversely, for short periods (such as a 30-minute peak-shaving period), the threshold restriction can be relaxed to avoid over-sensitivity triggering the correction mechanism.

[0041] In S22, the actual transmission period to be managed is compared with a preset interval database to determine its interval category and automatically extract the corresponding fluctuation threshold. For example, a 4-hour seasonal transmission and distribution plan will be matched with a long-term interval of more than 3 hours and a smaller fluctuation threshold will be used; while the 1-hour peak control at noon will be classified as a short-term interval of 0-1 hours and a larger fluctuation threshold will be used. This mechanism enables dynamic adaptation to the sensitivity differences of different seasons, weather patterns and grid load scenarios.

[0042] In one implementation, obtaining the percentage of historical power generation and transmission data sequences exceeding the fluctuation threshold and using it as the over-limit coverage rate in S2 includes: S23. Obtain the difference between each adjacent power generation and transmission data in the historical power generation and transmission data sequence; S24. Obtain the number of differences exceeding the fluctuation threshold and use it as the number of over-limit values. Obtain the proportion of the number of over-limit values ​​to the number of differences in the historical power generation and transmission data sequence and use it as the over-limit coverage rate.

[0043] In this embodiment, it should be noted that in S23, the difference in power generation changes between adjacent time points in the calculation sequence (such as the change amplitude between data points per second, per minute, or every 5 minutes) is calculated pairwise to form a complete fluctuation trajectory sequence. This sequence essentially reflects the instantaneous change intensity of power generation capacity within a historical period. For example, under cloudy weather conditions, the difference between adjacent minutes may exhibit a sawtooth-like, violent oscillation, while under sunny weather conditions, it shows a smooth, gradual change. This operation generates a set of basic fluctuation quantification data, providing direct input for subsequent threshold determination.

[0044] In S24, the number of differences exceeding the threshold selected in S22 in the fluctuation trajectory sequence is counted, and the proportion of this number in the total number of differences (i.e., the over-limit coverage rate) is calculated. This indicator directly quantifies the density of abrupt change points within historical data (e.g., 70% of the differences exceed the limit during a period of severe convective weather); this indicator reflects the intensity of the fluctuation (e.g., a high coverage state lasting for 30 minutes). For example, during a sandstorm, if minute-level differences are detected to continuously exceed the threshold and the coverage rate exceeds the preset threshold, it can be determined as an extremely unstable state, providing a decision-making basis for adaptive correction.

[0045] In one implementation, obtaining the correction ratio based on the historical power generation and transmission data sequence in S4 includes: S41. Based on the historical power generation and transmission data sequence, obtain the total trend variables with upward and downward trends respectively, and obtain the total trend variable with the same trend as the average power generation and transmission rate of change as the target variable, and obtain the total trend variable with the opposite trend to the average power generation and transmission rate of change as the reference variable. S42. Obtain the correction ratio based on the target variable and the reference variable.

[0046] In this embodiment, it should be noted that in S41, based on the difference in power generation changes between adjacent time points (such as the increase or decrease in power generation from a certain time to the next time), all differences are decomposed according to direction into an upward overall trend variable (the sum of all positive differences) and a downward overall trend variable (the sum of the absolute values ​​of all negative differences). Subsequently, using the average rate direction generated in S1 as a reference: if the average rate is positive (overall upward trend), the upward trend resultant force is defined as the target variable supporting the trend, and the downward trend resultant force serves as a reference variable for reverse interference; conversely, if the average rate is negative (overall downward trend), the downward trend resultant force becomes the target variable, and the upward trend resultant force becomes a reference variable. For example, during periods of frequent summer thunderstorms, the minute-level surge in power generation caused by intense sunlight and the precipitous drop caused by sudden rainstorms create a counterforce; by classifying and accumulating, the total force of both sides can be clearly quantified.

[0047] In S42, a correction ratio is generated based on the conflicting energy ratio between the target variable and the reference variable. When the target variable is dominant (e.g., the total positive fluctuation is larger in an overall upward trend), the relative proportion by which the target variable exceeds the reference variable is calculated as a positive correction value, amplifying the reliability of the original trend. Conversely, when the reference variable is stronger (e.g., a sudden hailstorm causes the total negative fluctuation to exceed the target variable), the proportion by which the reference variable exceeds the target variable is calculated and taken as a negative value, suppressing the prediction bias of the original trend. For example, on the morning of a typhoon's passage, although the average rate shows a slow increase, the sudden drop in instantaneous power generation caused by frequent squall lines causes a surge in the reference variable. Based on this, a negative correction ratio is generated, making the final corrected prediction trend closer to the actual fluctuation risk. This conflict quantification mechanism ensures an intelligent balance between trend dominance and the intensity of sudden disturbances in high-noise scenarios.

[0048] It should also be noted that the corrected power generation and transmission rate of change is obtained by multiplying the correction ratio by the average power generation and transmission rate of change. Then, a similar step can be used in S3 to obtain the first target transmission data sequence for the transmission period to be managed, which is to obtain the second target transmission data sequence for the transmission period to be managed, based on the prediction model and the average power generation and transmission rate of change.

[0049] In one implementation, the correction ratio obtained in S42 based on the target variable and the reference variable is expressed as follows: ;in, To correct the ratio, For the target variable, Used as a reference variable.

[0050] In this embodiment, it should be noted that the entire expression essentially constructs a dynamic trend reliability assessment, which addresses the insensitivity of the average rate to local abrupt changes in solar power generation fluctuation scenarios. Specifically, when Time (i.e., trend-driven), expression The calculation refers to the net advantage percentage of the overall trend variable after removing interference; if the total fluctuations in the historical series are in the same direction as the average rate... Significantly greater than the total reverse fluctuation This indicates that despite local mutations, the overall trend remains relatively reliable. In this case, the correction ratio is positive, meaning only a mild reduction in the average rate (molecular) is needed. It is the net trend force, divided by Normalization to a proportional representation helps preserve the trend direction while reflecting the weakening effect of sudden disturbances. For scenarios like clear skies turning cloudy (overall rise but occasional cloud cover), it avoids over-correction due to a few sharp drops, thus preventing the loss of the main trend and improving forecast stability.

[0051] Furthermore, when Time (i.e., fluctuation-dominated): Expression This calculates the relative strength of the counter-trend fluctuation exceeding the trend. If the total counter-trend fluctuation... Exceeding or equaling the combined force of trends This indicates that the sudden disturbance has severely distorted the overall trend. At this point, the correction ratio is negative. (molecular It is the net interference intensity, divided by (Normalization) generates a strong suppression coefficient, which directly reverses the rate of change of average power generation and transmission, causing the subsequent predicted trend to change from rising to falling, or from falling to rising. At the same time, the magnitude of the absolute value of the rate of change weakened by the strong suppression coefficient reflects the influence of the other trend.

[0052] Furthermore, regardless of whether the target or reference variable is dominant, its own total amount is used as the denominator. or Ensure the correction ratio always remains within the range of [-1, 1]. When hour, Located in (0, 1], The larger or The smaller the value, the closer the correction ratio is to 1, avoiding over-correction of stable trends that are the same as the rate of change in average power generation and transmission. hour, The stronger the interference, the more likely it is to occur when the range is [-1, 0). The larger or The smaller the value, the closer the negative ratio is to -1, ensuring that the corrected rate of change in power generation and transmission is directly opposite to the trend of the original average rate of change in power generation and transmission, and that the trend is stable, thus avoiding over-correction of the value.

[0053] A predictive model-based adaptive solar power transmission system is also provided, comprising: The first data processing module is used to obtain the transmission period to be managed, obtain the historical power generation and transmission data sequence of the preset time period before the transmission period to be managed, and obtain the average power generation and transmission change rate based on the historical power generation and transmission data sequence. The second data processing module is used to select the fluctuation threshold according to the transmission period to be managed, obtain the proportion of the historical power generation and transmission data sequence that exceeds the fluctuation threshold and use it as the over-limit coverage rate. The first predictive transmission module is used to obtain the first target transmission data sequence for the transmission period to be managed based on the prediction model and the average rate of change of power generation and transmission when the over-limit coverage rate does not exceed a preset threshold. The second predictive transmission module is used to obtain a correction ratio based on historical power generation and transmission data sequences when the over-limit coverage rate exceeds a preset threshold, obtain a corrected power generation and transmission change rate based on the average power generation and transmission change rate and the correction ratio, and obtain a second target transmission data sequence for the transmission period to be managed based on the predictive model and the corrected power generation and transmission change rate.

[0054] In one embodiment, the first predictive transmission module is further configured to: obtain multiple key time points of the transmission period to be managed based on the prediction model; take the power generation transmission data at the end of the historical power generation transmission data sequence as the starting power generation transmission data of the transmission period to be managed, and obtain the predicted power generation transmission data corresponding to multiple key time points based on the prediction model, the average power generation transmission change rate and the starting power generation transmission data; and form a first pre-transmission data sequence of the transmission period to be managed based on the predicted power generation transmission data corresponding to multiple key time points.

[0055] In one embodiment, the second predictive transmission module is further configured to: obtain total trend variables with upward and downward trends respectively based on historical power generation and transmission data sequences, obtain a total trend variable with the same trend as the average power generation and transmission rate of change as a target variable, and obtain a total trend variable with the opposite trend to the average power generation and transmission rate of change as a reference variable; and obtain a correction ratio based on the target variable and the reference variable.

[0056] In this embodiment, it should be noted that the specific method of performing the operation of the above-mentioned solar power adaptive transmission system based on the prediction model has been described in detail in the embodiments of the solar power adaptive transmission method based on the prediction model, and will not be elaborated here.

[0057] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0058] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0059] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. 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 or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A solar power generation adaptive transmission method based on a predictive model, characterized in that, include: Obtain the transmission period to be managed, and obtain the historical power generation and transmission data sequence of the preset time period before the transmission period to be managed, and obtain the average power generation and transmission change rate based on the historical power generation and transmission data sequence; Select a fluctuation threshold based on the transmission period to be managed, obtain the percentage of historical power generation and transmission data sequences that exceed the fluctuation threshold, and use it as the over-limit coverage rate; If the over-limit coverage rate does not exceed the preset threshold, the first target transmission data sequence for the transmission period to be managed is obtained based on the prediction model and the average rate of change of power generation and transmission. If the over-limit coverage rate exceeds the preset threshold, the total trend variables with upward and downward trends are obtained respectively based on the historical power generation and transmission data sequence. The total trend variable with the same trend as the average power generation and transmission rate of change is obtained as the target variable, and the total trend variable with the opposite trend to the average power generation and transmission rate of change is obtained as the reference variable. The correction ratio is obtained based on the target variable and the reference variable; The corrected power generation and transmission rate is obtained based on the average power generation and transmission rate of change and the correction ratio. The second target transmission data sequence for the transmission period to be managed is obtained based on the prediction model and the corrected power generation and transmission rate of change.

2. The method of claim 1, wherein, The step of obtaining the average rate of change in power generation and transmission based on historical power generation and transmission data sequences includes: Obtain the changes in power generation and transmission at the beginning and end of the historical power generation and transmission data sequence, as well as the historical transmission duration; The average rate of change in power generation and transmission is obtained by comparing the change in the first and last power generation and transmission amounts with the historical transmission duration.

3. The method of claim 1, wherein, The first target transmission data sequence for the managed transmission period, obtained based on the prediction model and the average rate of change in power generation and transmission, includes: Multiple key time points for the delivery period to be managed are obtained based on the prediction model; The power generation and transmission data at the end of the historical power generation and transmission data sequence is used as the starting power generation and transmission data for the transmission period to be managed. Based on the prediction model, the average rate of change of power generation and transmission, and the starting power generation and transmission data, the predicted power generation and transmission data corresponding to multiple key time points are obtained. The first pre-transmission data sequence for the transmission period to be managed is formed based on the predicted power generation and transmission data corresponding to multiple key time points.

4. The method of claim 1, wherein, The selection of fluctuation thresholds based on the transport period to be managed includes: Obtain multiple consecutive numerical intervals, each corresponding to a different fluctuation threshold. The larger the numerical value, the smaller the fluctuation threshold. Select the corresponding numerical range based on the size of the transport period to be managed, and obtain the fluctuation threshold corresponding to that numerical range.

5. The method of claim 1, wherein, The acquisition of the percentage of historical power generation and transmission data sequences exceeding the fluctuation threshold, and its determination as the over-limit coverage rate, includes: Obtain the difference between each adjacent power generation and transmission data in the historical power generation and transmission data sequence; The number of differences exceeding the fluctuation threshold is obtained as the number of over-limit values, and the proportion of the number of over-limit values ​​to the number of differences in the historical power generation and transmission data sequence is obtained as the over-limit coverage rate.

6. The method of claim 1, wherein, The correction ratio obtained based on the target variable and the reference variable is expressed as follows: ;in, To correct the ratio, For the target variable, Used as a reference variable.

7. A solar power generation adaptive delivery system based on a predictive model, characterized by, The system includes: The first data processing module is used to obtain the transmission period to be managed, obtain the historical power generation and transmission data sequence of the preset time period before the transmission period to be managed, and obtain the average power generation and transmission change rate based on the historical power generation and transmission data sequence. The second data processing module is used to select the fluctuation threshold according to the transmission period to be managed, obtain the proportion of the historical power generation and transmission data sequence that exceeds the fluctuation threshold and use it as the over-limit coverage rate. The first predictive transmission module is used to obtain the first target transmission data sequence for the transmission period to be managed based on the prediction model and the average rate of change of power generation and transmission when the over-limit coverage rate does not exceed a preset threshold. The second predictive transmission module is used to obtain a correction ratio based on historical power generation and transmission data sequences when the over-limit coverage rate exceeds a preset threshold, obtain a corrected power generation and transmission change rate based on the average power generation and transmission change rate and the correction ratio, and obtain a second target transmission data sequence for the transmission period to be managed based on the predictive model and the corrected power generation and transmission change rate. The second predictive transmission module is further configured to: obtain the total trend variables of upward and downward trends respectively based on the historical power generation and transmission data sequence, and obtain the total trend variable with the same trend as the average power generation and transmission rate of change as the target variable, and obtain the total trend variable with the opposite trend to the average power generation and transmission rate of change as the reference variable; and obtain the correction ratio based on the target variable and the reference variable.

8. The prediction model based solar power adaptive delivery system of claim 7, wherein, The first predictive delivery module is also used for: Multiple key time points for the delivery period to be managed are obtained based on the prediction model; The power generation and transmission data at the end of the historical power generation and transmission data sequence is used as the starting power generation and transmission data for the transmission period to be managed. Based on the prediction model, the average rate of change of power generation and transmission, and the starting power generation and transmission data, the predicted power generation and transmission data corresponding to multiple key time points are obtained. The first pre-transmission data sequence for the transmission period to be managed is formed based on the predicted power generation and transmission data corresponding to multiple key time points.

Images