A multi-energy fusion power generation scheduling method and system based on power gap prediction

Abstract

The application discloses a kind of based on power gap prediction's multi-energy fusion power generation scheduling method and system, comprising: based on the environmental meteorological data and historical load data collected in future preset time period Power gap curve is calculated;Based on power gap curve extraction power gap amplitude and corresponding gap duration;Gap duration is compared with preset time threshold, obtain comparison result, the time threshold is used to define the time limit of short power gap;According to comparison result, determine the target response subject matched with gap duration, and generate dynamic operation strategy based on target response subject.The application compared with prior art, can significantly improve the power gap prediction accuracy of multi-energy system, the matching degree of response strategy and gap duration and equipment operation economy, overcome the technical shortcomings of large power gap prediction deviation, response strategy and gap duration mismatch and equipment frequent start-stop loss in traditional scheme.

Description

Technical Field

[0001] This application relates to the field of microgrid energy management technology, and more specifically, to a multi-energy integrated power generation dispatching method and system based on power gap prediction. Background Technology

[0002] With the transformation of the global energy structure and the advancement of the "dual carbon" goal, multi-energy integrated microgrids, composed of renewable energy sources such as wind and solar power and traditional energy sources, have become an important carrier for achieving efficient utilization of distributed energy. In the actual operation of microgrids, accurately maintaining the power balance between sources and loads is the core of ensuring system stability, which requires the system to have high-precision prediction capabilities for future power gaps and corresponding flexible scheduling capabilities. Modern microgrid application scenarios require energy management systems not only to cope with long-term trend fluctuations caused by macro-weather changes, but also to have the ability to capture instantaneous power surges caused by micro-environmental factors such as cloud cover and wind turbine wakes.

[0003] While some existing technologies attempt to predict and schedule power gaps through historical load regression or single macro weather forecasts, they still suffer from a lack of coordination between the accuracy of micro-environmental perception and multi-source response decision-making mechanisms. This results in existing multi-energy forecasting systems exhibiting large power gap prediction deviations, mismatches between response strategies and gap durations, and high losses due to frequent equipment start-ups and shutdowns under complex weather change scenarios.

[0004] Therefore, how to provide a multi-energy integrated power generation dispatching method and system based on power gap prediction, which can significantly improve the power gap prediction accuracy, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems, and overcome the technical shortcomings of traditional schemes such as large power gap prediction deviation, mismatch between response strategy and gap duration, and high losses from frequent equipment start-up and shutdown, has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a multi-energy integrated power generation dispatch method based on power gap prediction, which can significantly improve the accuracy of power gap prediction, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems. It overcomes the technical shortcomings of traditional schemes, such as large power gap prediction deviation, mismatch between response strategy and gap duration, and high losses from frequent equipment start-up and shutdown.

[0006] The first technical solution provided in this application is as follows: This application provides a multi-energy integrated power generation scheduling method based on power gap prediction, comprising: S1 collecting environmental meteorological data and historical load data, and calculating a predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data; S2 extracting the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; S3 comparing the gap duration with a preset time threshold to obtain a comparison result, wherein the time threshold is used to define the time limit of the short-term power gap; S4 determining a target response subject matching the gap duration based on the comparison result, and generating a dynamic operation strategy based on the target response subject; wherein the target response subject includes: fuel cell equipment and internal combustion engine equipment.

[0007] Further, in a preferred embodiment of the present invention, the step of calculating the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data includes: Collect satellite remote sensing images and weather forecast data, and predict the power generation trend of renewable energy power generation equipment in a future preset time period based on the satellite remote sensing images and weather forecast data; Predict the load baseline trend within the preset future time period based on the historical load data; Calculate the difference sequence between the load demand baseline trend and the power generation trend, and construct the difference sequence as a power baseline gap curve; Real-time operating data of the renewable energy power generation equipment is collected, spatiotemporal correlation analysis is performed, and the power reference gap curve is corrected based on the spatiotemporal correlation analysis results to obtain the predicted power gap curve. The renewable energy power generation equipment includes photovoltaic sub-arrays and wind turbine generator sets.

[0008] Furthermore, in a preferred embodiment of the present invention, the step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: Real-time monitoring of the instantaneous output power of each photovoltaic sub-array and marking the corresponding timestamp; Identify the source photovoltaic subarray and associated photovoltaic subarrays, wherein the source photovoltaic subarray refers to the photovoltaic subarray with the first instantaneous power change, and the associated photovoltaic subarray refers to the photovoltaic subarray with the second instantaneous power change adjacent to the source photovoltaic subarray; Based on the geographical distance and time difference between the source photovoltaic subarray and the associated photovoltaic subarray, the cloud movement speed and cloud movement direction are calculated. A cloud shadow trajectory prediction model is constructed based on the cloud movement speed and direction. Based on the cloud shadow trajectory prediction model, the cloud shadow power drop time of each photovoltaic sub-array on the cloud shadow movement path is calculated. The power reference gap curve is corrected based on the time of the cloud shadow power drop to obtain the predicted power gap curve.

[0009] Furthermore, in a preferred embodiment of the present invention, the step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: Real-time monitoring of the output power and speed change rate of wind turbine generators located upwind; When a power drop is detected in the wind turbine at the upwind location, the wake propagation delay time is calculated based on the current real-time wind speed and wind direction data. The wake power drop time of each wind turbine generator set is calculated based on the wake propagation delay time. The power reference gap curve is corrected based on the wake power drop time to obtain the predicted power gap curve.

[0010] Further, in a preferred embodiment of the present invention, the step of determining a target response subject matching the duration of the gap based on the comparison result includes: The duration of the gap is compared with a preset first time threshold and a preset second time threshold, wherein the second time threshold is greater than the first time threshold. When the duration of the power deficit is less than or equal to the first time threshold, the fuel cell is activated to compensate for the power deficit amplitude, while the internal combustion engine remains off. When the duration of the gap is greater than or equal to the second time threshold, the internal combustion engine is started to perform preheating. After the internal combustion engine has completed preheating, the fuel cell is controlled to shut down. When the duration of the power gap is greater than the first time threshold and less than the second time threshold, the estimated power generation cost of the fuel cell and the internal combustion engine is calculated based on the power gap amplitude and the duration of the gap, and the target response subject is determined.

[0011] Further, in a preferred embodiment of the present invention, the step of calculating the estimated power generation cost of the fuel cell and the internal combustion engine based on the power deficit magnitude and deficit duration includes: Collect the methanol fuel unit price and historical fuel conversion efficiency of the fuel cell, and calculate the first fuel cost consumed during the duration of the power gap. The fuel unit price, oil-to-electricity conversion efficiency and fixed loss cost per start of the internal combustion engine equipment are obtained, and the second comprehensive cost of the internal combustion engine equipment during the duration of the gap is calculated. Compare the first fuel cost with the second comprehensive cost, and determine the target response subject based on the comparison results.

[0012] Furthermore, in a preferred embodiment of the present invention, the step of starting the internal combustion engine to perform preheating, and controlling the fuel cell to shut down after the internal combustion engine has completed preheating, specifically involves: Obtain the residual temperature of the internal combustion engine equipment; The residual temperature is compared with a preset first temperature threshold and a second temperature threshold, wherein the first temperature threshold is greater than the second temperature threshold. Based on the comparison results, the supply status of the associated heat of the fuel cell is controlled. If the residual temperature is greater than the first temperature threshold, the internal combustion engine is determined to be in hot standby mode, the supply of heat generated by the fuel cell is stopped, and the fuel cell is controlled to execute a shutdown command. If the residual temperature is between the first temperature threshold and the second temperature threshold, the internal combustion engine is determined to be in a warm-up state, and the associated heat is controlled to be supplied intermittently. If the residual temperature is less than the second temperature threshold, the internal combustion engine is determined to be in a cold state, and the associated heat is controlled to be continuously supplied.

[0013] Furthermore, in a preferred embodiment of the present invention, the method further includes feedback correction of the prediction error of the predicted power gap curve, including: Collect actual power deficit data during the execution of the dynamic operation strategy; Calculate the prediction deviation between the actual power gap data and the predicted power gap curve at the corresponding time. A correction coefficient is generated based on the predicted deviation value, and the power gap amplitude and gap duration are adjusted according to the correction coefficient.

[0014] This application provides a second technical solution as follows: This application also provides a multi-energy integrated power generation dispatch system based on power gap prediction, the system comprising: The power gap prediction module collects environmental meteorological data and historical load data, and calculates the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data. The gap feature extraction module extracts the power gap amplitude and the corresponding gap duration based on the predicted power gap curve. The duration threshold determination module compares the duration of the gap with a preset time threshold and obtains the comparison result. The time threshold is used to define the time limit of the short-term power gap. The operation strategy generation module determines the target response subject that matches the duration of the gap based on the comparison results, and generates a dynamic operation strategy based on the target response subject. The target response entities include fuel cell devices and internal combustion engine devices.

[0015] Furthermore, in a preferred embodiment of the present invention, the power gap prediction module includes: The power generation trend prediction unit collects satellite remote sensing images and weather forecast data, and predicts the power generation trend of renewable energy power generation equipment within a preset time period based on the satellite remote sensing images and weather forecast data. The load baseline prediction unit predicts the load baseline trend within the future preset time period based on the historical load data. The benchmark gap construction unit calculates the difference sequence between the load demand benchmark trend and the power generation trend, and constructs the difference sequence into a power benchmark gap curve; The gap curve correction unit collects real-time operating data of the renewable energy power generation equipment, performs spatiotemporal correlation analysis, corrects the power benchmark gap curve based on the spatiotemporal correlation analysis results, and obtains the predicted power gap curve.

[0016] This invention provides a multi-energy integrated power generation scheduling method based on power gap prediction, comprising: S1 collecting environmental meteorological data and historical load data, and calculating a predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data; S2 extracting the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; S3 comparing the gap duration with a preset time threshold to obtain a comparison result, wherein the time threshold is used to define the time limit of the short-term power gap; S4 determining a target response subject matching the gap duration based on the comparison result, and generating a dynamic operation strategy based on the target response subject; wherein the target response subject includes: fuel cell equipment and internal combustion engine equipment. This invention first collects environmental meteorological data and historical load data to calculate the predicted power gap curve for a preset future time period. Utilizing multi-source data, it achieves quantitative prediction of the future supply-demand difference, overcoming the large prediction deviation caused by a single data source in existing technologies. Based on this predicted power gap curve, it extracts the power gap amplitude and corresponding gap duration, comparing the gap duration with a preset time threshold. This time threshold serves as the time boundary for defining short-term power gaps, thus constructing a precise identification mechanism for gaps at different time scales. Based on the comparison results, it determines the target response subject matching the gap duration. Using the gap duration characteristics, it diverts and schedules fuel cell or internal combustion engine equipment with different response characteristics, effectively avoiding equipment malfunctions caused by mismatch between strategy and duration. Finally, it generates a dynamic operation strategy based on the target response subject, ensuring power balance while eliminating high losses caused by frequent equipment start-stops. Therefore, this invention significantly improves the power gap prediction accuracy, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems, overcoming the technical shortcomings of traditional solutions such as large power gap prediction deviations, mismatch between response strategies and gap duration, and high losses from frequent equipment start-stops. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the steps of the multi-energy integrated power generation scheduling method based on power gap prediction provided in this embodiment of the invention; Figure 2 A schematic diagram of the process for predicting the power gap curve provided in an embodiment of the present invention; Figure 3A schematic diagram of the process for predicting the power gap based on photovoltaic correction provided in an embodiment of the present invention; Figure 4 A schematic diagram of the preheating coordinated control provided in an embodiment of the present invention; Figure 5 The structural block diagram of a multi-energy integrated power generation dispatch system based on power gap prediction provided in an embodiment of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly set on the other component; when a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to the other component.

[0021] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "first", "second", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" or "several" means two or more, unless otherwise explicitly specified.

[0023] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which this application can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size should still fall within the scope of the technical content disclosed in this application, provided that they do not affect the effects and purposes that this application can produce.

[0024] like Figures 1 to 5 As shown in the embodiments of this application, a multi-energy integrated power generation dispatching method based on power gap prediction can significantly improve the power gap prediction accuracy, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems. It overcomes the technical shortcomings of traditional schemes, such as large power gap prediction deviation, mismatch between response strategy and gap duration, and high losses from frequent equipment start-up and shutdown.

[0025] This invention provides a multi-energy integrated power generation scheduling method based on power gap prediction, specifically including: S1 collecting environmental meteorological data and historical load data, and calculating the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data; S2 extracting the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; S3 comparing the gap duration with a preset time threshold to obtain the comparison result, the time threshold being used to define the time limit of the short-term power gap; S4 determining the target response subject matching the gap duration based on the comparison result, and generating a dynamic operation strategy based on the target response subject; wherein the target response subject includes: fuel cell equipment and internal combustion engine equipment. This invention first collects environmental meteorological data and historical load data to calculate the predicted power gap curve for a preset future time period. Utilizing multi-source data, it achieves quantitative prediction of the future supply-demand difference, overcoming the large prediction deviation caused by a single data source in existing technologies. Based on this predicted power gap curve, it extracts the power gap amplitude and corresponding gap duration, comparing the gap duration with a preset time threshold. This time threshold serves as the time boundary for defining short-term power gaps, thus constructing a precise identification mechanism for gaps at different time scales. Based on the comparison results, it determines the target response subject matching the gap duration. Using the gap duration characteristics, it diverts and schedules fuel cell or internal combustion engine equipment with different response characteristics, effectively avoiding equipment malfunctions caused by mismatch between strategy and duration. Finally, it generates a dynamic operation strategy based on the target response subject, ensuring power balance while eliminating high losses caused by frequent equipment start-stops. Therefore, this invention significantly improves the power gap prediction accuracy, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems, overcoming the technical shortcomings of traditional solutions such as large power gap prediction deviations, mismatch between response strategies and gap duration, and high losses from frequent equipment start-stops.

[0026] The following describes in detail the steps of a multi-energy integrated power generation dispatch method based on power gap prediction, using specific embodiments.

[0027] Specifically, such as Figure 2 As shown, in a specific embodiment of the present invention, the steps for calculating the predicted power gap curve for a future preset time period based on environmental meteorological data and historical load data include: S11 collecting satellite remote sensing images and weather forecast data, and predicting the power generation trend of renewable energy power generation equipment for a future preset time period based on the satellite remote sensing images and weather forecast data; S12 predicting the load baseline trend for a future preset time period based on historical load data; S13 calculating the difference sequence between the load demand baseline trend and the power generation trend, and constructing the difference sequence into a power baseline gap curve; S14 collecting real-time operating data of renewable energy power generation equipment, performing spatiotemporal correlation analysis, correcting the power baseline gap curve based on the spatiotemporal correlation analysis results, and obtaining the predicted power gap curve, wherein the renewable energy power generation equipment includes photovoltaic sub-arrays and wind turbine generators.

[0028] In a specific embodiment of the present invention, during the stage of constructing the power reference gap curve, the system first accesses a wide-area meteorological data source, collecting high-resolution satellite remote sensing images and numerical weather forecast data covering the area where the target microgrid is located, such as visible light or infrared cloud images updated at regular intervals. Based on the satellite remote sensing images, the distribution and movement trends of large-scale cloud clusters can be identified. Combined with meteorological parameters such as irradiance, wind speed, temperature, and air pressure provided by weather forecasts, a multi-source fusion deep learning prediction model is constructed to deduce the theoretical power generation trend of photovoltaic and wind power generation equipment in the future preset time period. As a preferred embodiment, the prediction model adopts a deep convolutional long short-term memory network architecture: First, the satellite images are processed using a convolutional neural network layer. The first convolutional layer contains 32 3×3 convolutional kernels, combined with a 2×2 max pooling layer. The model first extracts primary features of cloud texture from a layer, and then uses a second convolutional layer containing 64 3×3 convolutional kernels to further extract deeper semantic features such as cloud coverage and flatten them into spatial feature vectors. Next, a long short-term memory network is used to process the meteorological sequence. The first layer contains 128 hidden units and includes a dropout layer with a dropout rate of 0.2 to prevent overfitting. The second layer contains 64 hidden units and outputs a temporal context feature vector. Finally, the spatial and temporal feature vectors are concatenated in a fusion layer. After feature fusion, the model outputs a linearly activated output layer that directly outputs the predicted power sequence for a predetermined time period. Furthermore, the model is pre-trained using historical measured power data as labels, and the loss function is root mean square error with an added L2 regularization term to constrain the weights. The optimizer used is Adam, with an initial learning rate set to 0.001. It is combined with the ReduceLROnPlateau strategy. To prevent overfitting, an early stopping mechanism is set: when the validation set RMSE does not decrease within 20 consecutive epochs, the training is automatically terminated and the optimal weight parameters are saved, thereby achieving high-precision fitting of the photovoltaic and wind power trends. In this embodiment, the preset time period can be set to the next 4 hours. At the same time, the system retrieves load data from the same historical period, such as the past week or similar weather days in history, and predicts the changes in electricity demand on the user side during this time period using the time series extrapolation method, forming a load demand baseline trend. Based on this, the system aligns the load demand baseline trend with the power generation trend on the time axis and performs a difference calculation, that is, subtracting the predicted power generation value from the predicted load value at each moment, thereby constructing a preliminary power baseline gap curve. Although this baseline curve can reflect the general trend of supply and demand, it is often difficult to capture local sudden weather disturbances. Therefore, this embodiment further introduces a spatiotemporal correlation analysis correction mechanism: the system collects the current actual operating data of renewable energy power generation equipment in real time and compares the real-time data with the baseline predicted value at the same monitored time.Specifically, the system utilizes the geographical spatial distribution of renewable energy power generation nodes to capture the time delay of power surge signals propagating between different device nodes. When a significant deviation between the actual power and the benchmark value is detected at an upstream node, such as an upwind wind turbine or a photovoltaic array first blocked by clouds, the system calculates the time lag of the deviation being transmitted to the downstream node based on the current fluid velocity. This allows for the early amplitude correction of the power benchmark curve of the downstream node at future moments. This method transforms isolated single-point monitoring data into transitive spatiotemporal vector information, thereby obtaining a complete predicted power gap curve containing details of micro-fluctuations.

[0029] Specifically, such as Figure 3 As shown, in a specific embodiment of the present invention, the step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: S141 real-time monitoring of the instantaneous output power of each photovoltaic subarray and marking the corresponding timestamp; S142 identifying the source photovoltaic subarray and associated photovoltaic subarrays, where the source photovoltaic subarray represents the first photovoltaic subarray with a sudden change in instantaneous power, and the associated photovoltaic subarray represents the second photovoltaic subarray with a sudden change in instantaneous power adjacent to the source photovoltaic subarray; S143 calculating the cloud movement speed and cloud movement direction based on the geographical distance and time difference between the source photovoltaic subarray and the associated photovoltaic subarray; S144 constructing a cloud shadow trajectory prediction model based on the cloud movement speed and cloud movement direction, and calculating the cloud shadow power drop time of each photovoltaic subarray on the cloud shadow movement path based on the cloud shadow trajectory prediction model; S145 correcting the power reference gap curve based on the cloud shadow power drop time to obtain the predicted power gap curve.

[0030] In an embodiment of the invention, the system captures microscopic power fluctuations caused by cloud shadow movement by acquiring the instantaneous output power of each photovoltaic subarray at high frequency (e.g., sub-second or second-level) and recording precise timestamps. During this process, when a cloud passes over the photovoltaic power station, the system first identifies the source photovoltaic subarray by monitoring the power change rate. This source photovoltaic subarray is defined as the first unit in the entire array to experience a significant power decrease, specifically determined by its power change gradient exceeding a preset negative threshold. This negative threshold is typically configured based on the rated capacity of the photovoltaic inverter and historical natural fluctuation noise benchmarks under clear weather conditions, aiming to distinguish between severe power drops caused by cloud edge shading and changes in sunlight angle. The slow decay caused by changes; as a specific preferred example, if the rated power of a single photovoltaic subarray is 50kW, the negative threshold can be set to a power drop of 10%-15% of the rated power per second. That is, when the instantaneous power drop of the subarray exceeds 5kW-7.5kW within 1 second, the system determines that a valid cloud shading event has occurred. Once the source photovoltaic subarray is identified, the system immediately scans its physical neighborhood to identify associated photovoltaic subarrays, that is, subarrays that are spatially adjacent to the source photovoltaic subarray and temporally follow it with a similar large power drop. After obtaining the identities of these two key arrays, the system uses pre-stored geographic information system data... By retrieving the latitude and longitude coordinates of both, the physical straight-line distance between them is calculated. Combined with the time stamp difference between the two power surges, the moving speed of the cloud shadow is deduced by calculating the ratio of physical distance to time difference. Simultaneously, the direction of cloud movement is determined based on the spatial vector pointing from the source to the associated array. Based on these two vector parameters, the system constructs a cloud shadow trajectory prediction model. This model virtually extends the calculated movement vector on the map of the photovoltaic power station, thereby covering other photovoltaic sub-arrays located downstream of the movement path. Based on the physical distance between this extended trajectory and each downstream sub-array, the system calculates the target time when each photovoltaic sub-array on the cloud shadow's movement path will experience a power drop. Finally, the system analyzes these dispersed data points with specific time... The predicted drop events marked by time intervals are mapped onto the time axis of the originally smooth power reference gap curve. The power values ​​at the corresponding moments are subtracted or the waveform is reconstructed, thereby correcting the macroscopic reference curve into a predicted power gap curve that can reflect the details of real-time cloud cover. This significantly improves the granularity and accuracy of power gap prediction. Tests have verified that, under clear, cloudless conditions, the prediction accuracy of the power gap is 3.2% ± 0.5%, while the traditional method's prediction accuracy is 5.8% ± 0.8%, a reduction of 44.8% in error rate. Under cloudy conditions with cloud shadows, the prediction accuracy of the power gap is 6.5% ± 1.2%, while the traditional method's prediction accuracy is 15.3% ± 2.1%, a reduction of 57.5% in error rate.

[0031] Specifically, in a specific embodiment of the present invention, the step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: real-time monitoring of the output power and speed change rate of wind turbine generators located upwind; when a power drop is detected in a wind turbine generator located upwind, calculating the wake propagation delay time based on the current real-time wind speed and wind direction data; calculating the wake power drop time of each wind turbine generator based on the wake propagation delay time; and correcting the power reference gap curve based on the wake power drop time to obtain the predicted power gap curve.

[0032] In this embodiment of the invention, the system first dynamically defines the topology of the turbines under the current wind direction based on the wind direction data from the wind farm's real-time weather station. The turbine located at the forefront of the wind is marked as the upwind turbine, and its output power and speed change rate are monitored in real-time at the millisecond level. When a significant power drop is detected in the upwind turbine, such as a power decrease of more than 20% within 5 seconds due to a sudden drop in wind speed or power cut-off protection, the system immediately initiates a wake propagation calculation program. This program does not assume that the downstream turbine will drop simultaneously; instead, it calculates the time difference for the low-energy air mass to reach the downstream turbine based on the current real-time wind speed and direction data. As a preferred example, assuming the physical straight-line distance between the upwind turbine and the downstream turbine directly behind it is 600m, and the current real-time average wind speed is 12m / s, the system calculates the wake propagation delay time as 50s by calculating the ratio of physical distance to wind speed. This means that the downstream turbine is expected to be affected 50 seconds after the upwind turbine drops. Based on this calculation, the system calculates the wake power drop time for each affected downstream wind turbine and accurately maps this drop event onto the future time axis of the power reference gap curve. In practice, the system pushes the current time on the reference curve back 50 seconds and adds an estimated power reduction, thereby correcting the originally smooth prediction curve into a predicted power gap curve that accurately reflects the cascading characteristics of the wind farm's power drop time series. This ensures that the dispatch system can anticipate and prepare for downstream power gaps that have not yet occurred. Tests have verified that under strong wind wake conditions, the prediction accuracy of the power gap is 7.8% ± 1.5%, while the prediction accuracy of the traditional scheme is 18.7% ± 2.5%, resulting in a 58.3% reduction in error rate.

[0033] More specifically, in this embodiment of the invention, for concurrent scenarios where photovoltaic correction and wind turbine correction may be triggered within the same time window: the system starts two parallel computing threads in the background to simultaneously calculate the photovoltaic power drop caused by cloud cover and the downstream wind turbine power drop caused by the wake effect of the upstream wind turbine. At this time, the two threads do not interfere with each other and output their respective power loss sequences on the time axis. Since there is a physical difference between the cloud shadow movement speed and the wake propagation speed, the time when the power drop occurs may be inconsistent. The system uses the time axis of the power reference gap curve as a unified reference system, aligns the timestamps of the calculated photovoltaic power drop value and the wind turbine power drop value, and performs linear superposition, so that the two independently calculated power drop values ​​are applied to the power reference gap curve simultaneously to correct the predicted power gap curve.

[0034] Specifically, in this embodiment of the invention, the step of determining the target response subject matching the duration of the gap based on the comparison results includes: comparing the duration of the gap with a preset first time threshold and a preset second time threshold, wherein the second time threshold is greater than the first time threshold; when the duration of the gap is less than or equal to the first time threshold, starting the fuel cell to compensate for the power gap amplitude, while the internal combustion engine remains off; when the duration of the gap is greater than or equal to the second time threshold, starting the internal combustion engine to perform preheating, and controlling the fuel cell to shut down after the internal combustion engine completes preheating; when the duration of the gap is greater than the first time threshold and less than the second time threshold, calculating the estimated power generation cost of the fuel cell and the internal combustion engine based on the power gap amplitude and the duration of the gap, and determining the target response subject. After testing and verification, under the implementation of this scheme, the average number of daily starts of the internal combustion engine is 2.3 ± 0.6 times, while the average number of daily starts of the internal combustion engine in the traditional scheme is 8.7 ± 1.2 times, a reduction of 73.6% in the number of starts.

[0035] In a specific embodiment of the present invention, regarding the determination of the target response subject, the system first classifies and defines different power generation devices based on their physical response characteristics. Fuel cell devices are defined as rapid compensation sources with millisecond-level dynamic response capabilities and no cold-start mechanical wear. Internal combustion engine devices are defined as baseload compensation sources with large thermal inertia, requiring a preheating process, and high steady-state operating efficiency. In this embodiment, the fuel cell is a methanol fuel cell, and the internal combustion engine is a diesel generator set. Based on these characteristic differences, the system establishes a time-based hierarchical response mechanism, presets a first time threshold as the economical time limit for a fuel cell to independently bear the load, and presets a time limit greater than this threshold. The first time threshold and the second time threshold serve as the minimum time threshold for starting the internal combustion engine. As a specific numerical setting example: For the first time threshold, the setting is based on the ratio of the fixed cost of a single start-up of the internal combustion engine to the per-minute operating cost difference between the fuel cell and the internal combustion engine. That is, the first time threshold is the payback period for starting the internal combustion engine; starting the engine before this time is not cost-effective. For the second time threshold, the setting is based on the sum of the physical preheating time of the internal combustion engine and the minimum steady-state operating time required to prevent carbon buildup. Based on this, the system compares the duration of the power gap extracted from the predicted power gap curve with these two key thresholds. When the duration of the power gap is less than or equal to the first time threshold... When the power shortage lasts only 10 minutes, the system considers that starting the internal combustion engine at this time would result in severe cold start wear and require shutdown before reaching the optimal thermal efficiency zone. Therefore, it determines this to be a short-term power fluctuation and directly generates a command to start the fast-responding fuel cell device to immediately compensate for the power shortage amplitude, while simultaneously forcing the internal combustion engine to remain off. When the duration of the power shortage is greater than or equal to the second time threshold, for example, when the duration reaches 60 minutes, the system recognizes that the shortage is sufficient to cover the preheating cost of the internal combustion engine and can leverage its cost advantage for long-term operation. It then generates a command to start the internal combustion engine and execute the preheating procedure. Furthermore, to ensure power supply continuity, the system... During the preheating period of the internal combustion engine, the system does not immediately shut down other power sources. Instead, it continuously monitors the cylinder temperature and output voltage of the internal combustion engine until it detects that the internal combustion engine has completed preheating and successfully connected to the grid. Only then does it issue a command to control the fuel cell equipment to shut down, completing a smooth transition from the fast source to the baseload source. When the duration of the power gap is between the first and second time thresholds, for example, if the gap lasts for 25 minutes, the system enters the economic optimization mode. Based on the current power gap amplitude and the duration of the gap, it calculates the expected cost of continuous operation of the fuel cell equipment and the comprehensive expected cost of starting up the internal combustion engine equipment, respectively. Based on the principle of lowest cost, it selects the unique target response subject between the two.

[0036] Specifically, in a specific embodiment of the present invention, the steps are based on the power gap amplitude and the gap duration to calculate the estimated power generation cost of the fuel cell and internal combustion engine equipment, including: collecting the methanol fuel unit price and historical fuel conversion efficiency of the fuel cell, calculating the first fuel cost consumed during the gap duration; obtaining the fuel unit price, oil-to-electricity conversion efficiency and fixed loss cost per start-up of the internal combustion engine equipment, calculating the second comprehensive cost of the internal combustion engine equipment during the gap duration; comparing the first fuel cost and the second comprehensive cost, and determining the target response subject based on the comparison result.

[0037] In this embodiment of the invention, the core of the system's cost calculation lies in distinguishing between linear operating costs and comprehensive costs including start-up losses. For fuel cell equipment, based on the collected methanol fuel price and historical fuel conversion efficiency, the system first multiplies the power deficit amplitude and deficit duration to obtain the total energy demand. Then, based on the conversion efficiency, the total energy demand is converted into the required fuel quantity and multiplied by the fuel price, thereby obtaining the first fuel cost that accumulates linearly with operating time. Simultaneously, for internal combustion engine equipment, the system not only calculates the variable cost of fuel consumption but also introduces the fixed cost of a single start-up as a decision correction term. Based on the basic fuel expenditure calculated using the fuel price and fuel-electric conversion efficiency, the system... The system forcibly superimposes a preset fixed cost for a single start-up, which includes the additional fuel consumption during the preheating phase and the depreciation of mechanical components caused by cold starts, thus deriving a second comprehensive cost. Finally, the system compares the first fuel cost (purely variable attribute) with the second comprehensive cost (including fixed penalties) and directly selects the equipment with the lower value as the target response subject. This comparison logic, which monetizes start-up losses, effectively reveals the hidden cost disadvantages of internal combustion engine units under non-long-term operating conditions, preventing the erroneous start-up of large-inertia equipment during uneconomical time spans due to focusing only on fuel price differences. This ensures that the scheduling decisions of multi-energy systems always conform to the economic principle of minimizing total operating costs.

[0038] Specifically, in an embodiment of the present invention, the step of starting the internal combustion engine to perform preheating, and controlling the fuel cell to shut down after the internal combustion engine has completed preheating, includes: acquiring the residual temperature of the internal combustion engine; comparing the residual temperature with a preset first temperature threshold and a second temperature threshold, wherein the first temperature threshold is greater than the second temperature threshold; determining the supply state of the associated heat of the fuel cell based on the comparison result; if the residual temperature is greater than the first temperature threshold, determining that the internal combustion engine is in a hot standby state, stopping the supply of associated heat of the fuel cell, and controlling the fuel cell to execute a shutdown command; if the residual temperature is between the first temperature threshold and the second temperature threshold, determining that the internal combustion engine is in a warm-up state, and controlling the associated heat to be supplied intermittently; if the residual temperature is less than the second temperature threshold, determining that the internal combustion engine is in a cold state, and controlling the associated heat to be supplied continuously.

[0039] Among them, such as Figure 4As shown in this embodiment of the invention, when the system detects a cloud approaching and determines that a preheating program needs to be initiated, it does not immediately start heat exchange at full speed. Instead, it first reads the current residual temperature of the internal combustion engine and compares it with a preset thermal state threshold. The first temperature threshold is set to 60 degrees Celsius, based on the lower limit temperature of the internal combustion engine's optimal thermal balance condition. The second temperature threshold is set to 40 degrees Celsius, based on the critical viscosity temperature of the lubricating oil for establishing an effective hydrodynamic oil film. If the current residual temperature is higher than the first temperature threshold, the internal combustion engine is determined to be in a hot standby state. At this time, the engine body clearance meets the design tolerance and no external intervention is required. The system controls the circulating pump to remain in a stopped state to cut off the heat transfer path and prevent backflow of internal combustion engine temperature due to pipeline heat dissipation. If the residual temperature is between the first temperature threshold and the second temperature threshold, the system determines that the internal combustion engine is in a hot standby state. Between the temperature threshold, the internal combustion engine is determined to be in a warm-up state. The system controls the circulation pump to operate in a low-flow intermittent pulse mode, for example, setting a cycle of running for 30 seconds and stopping for 90 seconds, to slowly transfer the heat generated by the fuel cell to the internal combustion engine to prevent its temperature from dropping rapidly. If the residual temperature is lower than the second temperature threshold, the internal combustion engine is determined to be in a cold-up state. The system controls the circulation pump to operate in full-speed mode, forcibly pumping high-temperature coolant into the internal combustion engine block to achieve rapid warm-up. It can concretize the abstract thermal energy coordination into the adaptive adjustment of the circulation medium flow rate, making full use of the thermal inertia of the physical system and avoiding the waste of pump auxiliary energy and heat backflow caused by blind heat exchange in cloudy weather. Compared with the traditional one-size-fits-all preheating logic, it significantly reduces the energy consumption of the auxiliary system and reduces the thermal stress fatigue of the internal combustion engine components. By introducing this precise heat transfer control mechanism based on system thermal inertia perception, the system concretizes the abstract thermal energy coordination into adaptive adjustment of the circulating medium flow rate, effectively solving the problems of pump auxiliary energy waste and heat backflow caused by blind heat exchange in intermittent shading scenarios. Through theoretical deduction and experimental testing, this method can reduce the auxiliary power consumption of the thermal management system by about 40%, while significantly reducing the thermal stress fatigue caused by the frequent rapid heating and cooling cycles of internal combustion engine components, and significantly improving the system's operating economy and equipment life in complex weather environments.

[0040] Specifically, in a specific embodiment of the present invention, the method further includes feedback correction of the prediction error of the predicted power gap curve, including: collecting actual power gap data during the execution of the dynamic operation strategy; calculating the prediction deviation between the actual power gap data and the predicted power gap curve at the corresponding time; generating a correction coefficient based on the prediction deviation value; and adjusting the power gap amplitude and gap duration according to the correction coefficient.

[0041] In a specific embodiment of the present invention, during the execution cycle of the dynamic operation strategy, the actual power gap data occurring at the load end is collected in real time using a high-frequency sampling sensor or state observer. For example, with a sampling interval of 10 to 50 milliseconds, the difference between the actual load demand power and the current power supply capacity at the current moment is recorded as the actual sample point. The collected actual power gap data is compared with the predicted data of the previously generated predicted power gap curve at the same time node to calculate the prediction deviation value at the current moment. This prediction deviation value can be reflected as the difference between the actual value and the predicted value, or as the ratio between the two. Based on the calculated prediction deviation value, a corresponding correction coefficient is generated by looking up a table or by using a proportional-integral algorithm. For example, when multiple consecutive When the actual power gap at the sampling point is greater than the predicted value and the deviation exceeds 5%, a gain correction coefficient greater than 1 is generated; or when the actual gap duration shows a trend of extension, a duration extension coefficient is generated. Based on the generated correction coefficient, the power gap amplitude and gap duration of the unexecuted portion in the predicted power gap curve are dynamically adjusted. Specifically, the gap amplitude in the subsequent prediction time window is multiplied by the gain correction coefficient to raise the prediction benchmark, or the estimated gap end time is updated based on the duration extension coefficient. This can eliminate static prediction errors caused by environmental changes or model aging in real time, making the system's response to power fluctuations more consistent with actual operating conditions, and preventing battery over-discharge due to insufficient prediction amplitude or power response lag due to insufficient prediction duration.

[0042] This invention also provides a multi-energy integrated power generation dispatch system based on power gap prediction. The system includes: a power gap prediction module, which collects environmental meteorological data and historical load data, and calculates a predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data; a gap feature extraction module, which extracts the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; a duration threshold determination module, which compares the gap duration with a preset time threshold to obtain a comparison result, wherein the time threshold is used to define the time limit of the short-term power gap; and an operation strategy generation module, which determines the target response subject that matches the gap duration based on the comparison result, and generates a dynamic operation strategy based on the target response subject; wherein the target response subject includes: fuel cell equipment and internal combustion engine equipment.

[0043] Specifically, such as Figure 5 As shown, in a specific embodiment of the present invention, the power gap prediction module includes: The power generation trend prediction unit collects satellite remote sensing images and weather forecast data, and predicts the power generation trend of renewable energy power generation equipment in the future within a preset time period based on the satellite remote sensing images and weather forecast data; The load baseline forecasting unit predicts the load baseline trend within a preset time period based on historical load data. The benchmark gap construction unit calculates the difference sequence between the load demand benchmark trend and the power generation trend, and constructs the difference sequence into a power benchmark gap curve; The gap curve correction unit collects real-time operating data of renewable energy power generation equipment, performs spatiotemporal correlation analysis, corrects the power benchmark gap curve based on the spatiotemporal correlation analysis results, and obtains the predicted power gap curve.

[0044] As described above, the present invention relates to a multi-energy integrated power generation scheduling method based on power gap prediction, comprising: S1 collecting environmental meteorological data and historical load data, and calculating a predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data; S2 extracting the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; S3 comparing the gap duration with a preset time threshold to obtain a comparison result, wherein the time threshold is used to define the time limit of the short-term power gap; S4 determining a target response subject matching the gap duration based on the comparison result, and generating a dynamic operation strategy based on the target response subject; wherein the target response subject includes: fuel cell equipment and internal combustion engine equipment. This invention first collects environmental meteorological data and historical load data to calculate the predicted power gap curve for a preset future time period. Utilizing multi-source data, it achieves quantitative prediction of the future supply-demand difference, overcoming the large prediction deviation caused by a single data source in existing technologies. Based on this predicted power gap curve, it extracts the power gap amplitude and corresponding gap duration, comparing the gap duration with a preset time threshold. This time threshold serves as the time boundary for defining short-term power gaps, thus constructing a precise identification mechanism for gaps at different time scales. Based on the comparison results, it determines the target response subject matching the gap duration. Using the gap duration characteristics, it diverts and schedules fuel cell or internal combustion engine equipment with different response characteristics, effectively avoiding equipment malfunctions caused by mismatch between strategy and duration. Finally, it generates a dynamic operation strategy based on the target response subject, ensuring power balance while eliminating high losses caused by frequent equipment start-stops. Therefore, this invention significantly improves the power gap prediction accuracy, the matching degree between response strategy and gap duration, and the economic efficiency of equipment operation in multi-energy systems, overcoming the technical shortcomings of traditional solutions such as large power gap prediction deviations, mismatch between response strategies and gap duration, and high losses from frequent equipment start-stops.

[0045] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-energy integrated power generation dispatch method based on power gap prediction, characterized in that, The method includes: S1 collects environmental meteorological data and historical load data, and calculates the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data. S2 extracts the power gap amplitude and the corresponding gap duration based on the predicted power gap curve; S3 compares the duration of the gap with a preset time threshold to obtain a comparison result. The time threshold is used to define the time limit of the short-term power gap. S4 determines the target response subject that matches the duration of the gap based on the comparison results, and generates a dynamic operation strategy based on the target response subject; The target response entities include: fuel cell equipment and internal combustion engine equipment; Based on the comparison results, target response subjects matching the duration of the gap are determined, including: The duration of the gap is compared with a preset first time threshold and a preset second time threshold, wherein the second time threshold is greater than the first time threshold. When the duration of the power deficit is less than or equal to the first time threshold, the fuel cell is activated to compensate for the power deficit amplitude, while the internal combustion engine remains off. When the duration of the gap is greater than or equal to the second time threshold, the internal combustion engine is started to perform preheating. After the internal combustion engine has completed preheating, the fuel cell is controlled to shut down. When the duration of the power gap is greater than the first time threshold and less than the second time threshold, the estimated power generation cost of the fuel cell and the internal combustion engine is calculated based on the power gap amplitude and the duration of the gap, and the target response subject is determined.

2. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 1, characterized in that, The step involves calculating the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data, including: Collect satellite remote sensing images and weather forecast data, and predict the power generation trend of renewable energy power generation equipment in a future preset time period based on the satellite remote sensing images and weather forecast data; Predict the load baseline trend within the preset future time period based on the historical load data; Calculate the difference sequence between the load demand baseline trend and the power generation trend, and construct the difference sequence as a power baseline gap curve; Real-time operating data of the renewable energy power generation equipment is collected, spatiotemporal correlation analysis is performed, and the power reference gap curve is corrected based on the spatiotemporal correlation analysis results to obtain the predicted power gap curve. The renewable energy power generation equipment includes photovoltaic sub-arrays and wind turbine generator sets.

3. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 2, characterized in that, The step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: Real-time monitoring of the instantaneous output power of each photovoltaic sub-array and marking the corresponding timestamp; Identify the source photovoltaic subarray and associated photovoltaic subarrays, wherein the source photovoltaic subarray refers to the photovoltaic subarray with the first instantaneous power change, and the associated photovoltaic subarray refers to the photovoltaic subarray with the second instantaneous power change adjacent to the source photovoltaic subarray; Based on the geographical distance and time difference between the source photovoltaic subarray and the associated photovoltaic subarray, the cloud movement speed and cloud movement direction are calculated. A cloud shadow trajectory prediction model is constructed based on the cloud movement speed and direction. Based on the cloud shadow trajectory prediction model, the cloud shadow power drop time of each photovoltaic sub-array on the cloud shadow movement path is calculated. The power reference gap curve is corrected based on the time of the cloud shadow power drop to obtain the predicted power gap curve.

4. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 2, characterized in that, The step of correcting the power reference gap curve based on the spatiotemporal correlation analysis results includes: Real-time monitoring of the output power and speed change rate of wind turbine generators located upwind; When a power drop is detected in the wind turbine at the upwind location, the wake propagation delay time is calculated based on the current real-time wind speed and wind direction data. The wake power drop time of each wind turbine generator set is calculated based on the wake propagation delay time. The power reference gap curve is corrected based on the wake power drop time to obtain the predicted power gap curve.

5. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 1, characterized in that, The step, based on the power deficit magnitude and deficit duration, calculates the estimated power generation cost of the fuel cell and the internal combustion engine, including: Collect the methanol fuel unit price and historical fuel conversion efficiency of the fuel cell, and calculate the first fuel cost consumed during the duration of the power gap. The fuel unit price, oil-to-electricity conversion efficiency and fixed loss cost per start of the internal combustion engine equipment are obtained, and the second comprehensive cost of the internal combustion engine equipment during the duration of the gap is calculated. Compare the first fuel cost with the second comprehensive cost, and determine the target response subject based on the comparison results.

6. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 1, characterized in that, The steps involve determining the power deficit amplitude and duration, and calculating the estimated shutdown of the fuel cell and internal combustion engine. Specifically: Obtain the residual temperature of the internal combustion engine equipment; The residual temperature is compared with a preset first temperature threshold and a second temperature threshold, wherein the first temperature threshold is greater than the second temperature threshold. Based on the comparison results, the supply status of the associated heat of the fuel cell is controlled. If the residual temperature is greater than the first temperature threshold, the internal combustion engine is determined to be in hot standby mode, the supply of heat generated by the fuel cell is stopped, and the fuel cell is controlled to execute a shutdown command. If the residual temperature is between the first temperature threshold and the second temperature threshold, the internal combustion engine is determined to be in a warm-up state, and the associated heat is controlled to be supplied intermittently. If the residual temperature is less than the second temperature threshold, the internal combustion engine is determined to be in a cold state, and the associated heat is controlled to be continuously supplied.

7. The multi-energy integrated power generation dispatch method based on power gap prediction according to claim 1, characterized in that, It also includes, as well as, the prediction error feedback correction for the predicted power gap curve, including: Collect actual power deficit data during the execution of the dynamic operation strategy; Calculate the prediction deviation between the actual power gap data and the predicted power gap curve at the corresponding time. A correction coefficient is generated based on the predicted deviation value, and the power gap amplitude and gap duration are adjusted according to the correction coefficient.

8. A multi-energy integrated power generation dispatch system based on power gap prediction, characterized in that, The system comprising the multi-energy integrated power generation dispatch method based on power gap prediction as described in any one of claims 1 to 7, wherein the system includes: The power gap prediction module collects environmental meteorological data and historical load data, and calculates the predicted power gap curve for a future preset time period based on the environmental meteorological data and historical load data. The gap feature extraction module extracts the power gap amplitude and the corresponding gap duration based on the predicted power gap curve. The duration threshold determination module compares the duration of the gap with a preset time threshold and obtains the comparison result. The time threshold is used to define the time limit of the short-term power gap. The operation strategy generation module determines the target response subject that matches the duration of the gap based on the comparison results, and generates a dynamic operation strategy based on the target response subject. The target response entities include fuel cell devices and internal combustion engine devices.

9. The multi-energy integrated power generation dispatch system based on power gap prediction according to claim 8, characterized in that, The power gap prediction module includes: The power generation trend prediction unit collects satellite remote sensing images and weather forecast data, and predicts the power generation trend of renewable energy power generation equipment within a preset time period based on the satellite remote sensing images and weather forecast data. The load baseline prediction unit predicts the load baseline trend within the future preset time period based on the historical load data. The benchmark gap construction unit calculates the difference sequence between the load demand benchmark trend and the power generation trend, and constructs the difference sequence into a power benchmark gap curve; The gap curve correction unit collects real-time operating data of the renewable energy power generation equipment, performs spatiotemporal correlation analysis, corrects the power benchmark gap curve based on the spatiotemporal correlation analysis results, and obtains the predicted power gap curve.

Images