A source-side preprocessing method, system, equipment, and medium for microgrid planning in highway service areas.
By acquiring meteorological data from highway service areas, the available output per unit capacity of photovoltaic and wind power is calculated, forming a joint source-side scenario and performing state transitions and resampling. This solves the problem that existing technologies cannot effectively preserve the physical mechanisms and temporal risks of photovoltaic and wind power, and enables direct calling of capacity configuration models, thereby improving the accuracy and reliability of planning.
Patent Information
- Application Number
- CN202611211225.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies lack a source-side preprocessing method for planning microgrids in highway service areas, which cannot effectively preserve the physical mechanisms and temporal risks of photovoltaic and wind power, and cannot directly output unit capacity scenarios, probability weights, and extreme risk labels for capacity configuration models to call.
By acquiring meteorological data from highway service areas over several days, the available output per unit capacity of photovoltaic and wind power is calculated, forming photovoltaic and wind power output scenarios. These scenarios are then linked to form a joint source-side scenario. State transitions and resampling are performed to generate standardized source-side data packets that can be directly input into the capacity configuration model.
It enables the preservation of the physical mechanisms and temporal risks of photovoltaic and wind power in the planning of microgrids in highway service areas, outputs unit capacity scenarios, probability weights and extreme risk labels, which can be directly called by the capacity configuration model, thus improving the accuracy and reliability of planning.
Smart Images

Figure CN123121830A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power grid planning, and in particular to a source-side preprocessing method, system, equipment, and medium for microgrid planning in highway service areas. Background Technology
[0002] Highway service area microgrids typically require the simultaneous configuration of photovoltaic, wind power, energy storage, charging facilities, distribution transformers, and external grid interfaces. With the increasing charging demands of new energy vehicles, service area microgrid planning no longer simply involves comparing annual power generation or annual investment costs. Instead, it necessitates identifying the impact of renewable energy source output on energy storage capacity, grid connection capacity, and operational risks under intraday, seasonal, weather-related, and extreme low-output events.
[0003] Existing planning for microgrids in highway service areas involves capacity planning, smart capacity expansion, probabilistic modeling of renewable energy output, and generation of combined wind and solar power scenarios. However, these technologies typically handle capacity optimization, operation scheduling, probability distribution fitting, and combined scenario generation separately, lacking a source-side preprocessing method for planning microgrids in highway service areas that preserves the physical mechanisms and temporal risks of photovoltaic and wind power while outputting unit capacity scenarios, probability weights, and extreme risk labels that can be directly used by the capacity configuration model. Summary of the Invention
[0004] This invention aims to address at least one of the technical problems existing in the prior art. To this end, this invention proposes a source-side preprocessing method for microgrid planning in highway service areas, which can retain the physical mechanisms and temporal risks of photovoltaic and wind power, and output unit capacity scenarios, probability weights, and extreme risk labels that can be directly called by the capacity configuration model.
[0005] The present invention also proposes a system, device and medium having the above-mentioned source-side preprocessing method for microgrid planning in highway service areas.
[0006] A source-side preprocessing method for microgrid planning in highway service areas according to a first aspect embodiment of the present invention includes: Obtain multiple days of meteorological data for the configured highway service areas; Based on the meteorological data, calculate the available power output per unit capacity of photovoltaic power and the available power output per unit capacity of wind power respectively. Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, each day is divided into preset photovoltaic output events and wind output events, respectively. By connecting the separately obtained light output scenarios and wind output scenarios, a joint source-side scenario is formed; The state transition and resampling of the joint source-side scenario are performed to obtain the input of the new energy output capacity configuration model.
[0007] According to an embodiment of the present invention, a source-side preprocessing method for microgrid planning in highway service areas has at least the following beneficial effects: The present invention obtains wind power output and photovoltaic power output from meteorological data of the planned highway service area, thus obtaining daily wind power output scenarios (i.e., wind power output scenarios) and daily photovoltaic power output scenarios (i.e., photovoltaic power output scenarios), and then combines them to obtain wind-solar combined scenarios, i.e., multi-day combined source-side scenarios. Finally, after transfer and resampling, the combined source-side scenarios within one year are obtained and directly input into the new energy output capacity; the physical mechanisms and temporal risks of photovoltaic and wind power are preserved, and unit capacity scenarios, probability weights, and extreme risk labels that can be directly called by the capacity configuration model are output.
[0008] According to some embodiments of the present invention, the optical output scene and wind output scene obtained by the connection respectively form a joint source-side scene, including: For the same time period and the same weather type, the solar power output scenarios and wind power output scenarios are respectively edge-distributed; among them, the weather types corresponding to the solar power output scenarios and wind power output scenarios include extreme weather; Connect the edge distributions corresponding to the light output scenario and the wind output scenario respectively.
[0009] According to some embodiments of the present invention, it further includes: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, calculate the changes in available output at adjacent time points. The event where the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold and the change in available output at the adjacent time is less than the preset minimum change is defined as a continuous low output event. A short-term ramp-up event is defined as any moment when the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold, and the change in available output at the adjacent moment is greater than the preset maximum change value.
[0010] According to some embodiments of the present invention, after performing state transition and resampling on the joint source-side scenario, the method further includes: If the events corresponding to the formed joint source-side scenario include short-term ramp events and continuous low-output events, then the joint source-side scenario is identified. Clustering and filtering are performed on unidentified joint source-side scenarios; The selected joint source-side scenarios are weighted by their sampling frequency to obtain standardized source-side data packets for use by the capacity configuration model; the sampling frequency is obtained by the proportion of a certain joint source-side scenario in all joint source-side scenarios.
[0011] A source-side preprocessing system for microgrid planning in highway service areas according to a second aspect of the present invention includes: The data acquisition module is used to acquire meteorological data for multiple days from the configured highway service areas; The power output modeling module calculates the available power output per unit capacity of photovoltaic power and the available power output per unit capacity of wind power based on the meteorological data. The three-layer time-series probabilistic modeling module is used to divide each day into preset solar power output events and wind power output events according to the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power. The wind and solar joint sampling module is used to connect the separately obtained solar power output scenarios and wind power output scenarios to form a joint source-side scenario; The source-side scenario library and planning interface module are used to perform state transitions and resampling on the joint source-side scenarios to obtain the input of the new energy output capacity configuration model.
[0012] According to some embodiments of the present invention, the optical output scene and wind output scene obtained by the connection respectively form a joint source-side scene, including: For the same time period and the same weather type, the solar power output scenarios and wind power output scenarios are respectively edge-distributed; among them, the weather types corresponding to the solar power output scenarios and wind power output scenarios include extreme weather; Connect the edge distributions corresponding to the light output scenario and the wind output scenario respectively.
[0013] According to some embodiments of the present invention, it further includes: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, calculate the changes in available output at adjacent time points. The event where the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold and the change in available output at the adjacent time is less than the preset minimum change is defined as a continuous low output event. A short-term ramp-up event is defined as any moment when the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold, and the change in available output at the adjacent moment is greater than the preset maximum change value.
[0014] According to some embodiments of the present invention, after performing state transition and resampling on the joint source-side scenario, the method further includes: If the events corresponding to the formed joint source-side scenario include short-term ramp events and continuous low-output events, then the joint source-side scenario is identified. Clustering and filtering are performed on unidentified joint source-side scenarios; The selected joint source-side scenarios are weighted by their sampling frequency to obtain standardized source-side data packets for use by the capacity configuration model; the sampling frequency is obtained by the proportion of a certain joint source-side scenario in all joint source-side scenarios.
[0015] An electronic device according to a third aspect of the present invention includes: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method as described in any one of the first aspects.
[0016] According to a fourth aspect of the present invention, a storage medium stores computer-executable instructions for performing the method as described in any one of the first aspects.
[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description
[0018] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.
[0019] Figure 1 This is a flowchart of a source-side preprocessing method for microgrid planning in highway service areas, provided by an embodiment of the present invention. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0021] It should be understood that in the description of the embodiments of the present invention, "multiple" (or "amounts") means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.
[0022] like Figure 1As shown, this embodiment of the invention provides a source-side preprocessing method for microgrid planning in highway service areas, including: Step S100: Obtain multiple days of meteorological data for the configured highway service areas; Step S200: Calculate the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power based on meteorological data. Step S300: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, the day is divided into preset photovoltaic output events and wind output events respectively. Step S400: Connect the obtained light output scene and wind output scene respectively to form a joint source-side scene; Step S500: Perform state transition and resampling on the joint source-side scenario to obtain the input of the new energy output capacity configuration model.
[0023] It should be noted that the joint source-side scenario is actually the scenario description parameters for each day. After state transition and resampling of the joint source-side scenario, the scenario description parameters for each day of the year are obtained and then input into the capacity configuration model.
[0024] This invention obtains wind power output and photovoltaic output from meteorological data of planned highway service areas, thus obtaining daily wind power output scenarios (wind power scenario) and daily photovoltaic output scenarios (solar power scenario). These are then combined to obtain a combined wind and solar scenario, i.e., a multi-day combined source-side scenario. Finally, after transfer and resampling, a combined source-side scenario within one year is obtained and directly input into the new energy output capacity. This invention preserves the physical mechanisms and temporal risks of photovoltaic and wind power, and outputs unit capacity scenarios, probability weights, and extreme risk labels that can be directly called by the capacity configuration model.
[0025] It should be noted that this invention is applicable to the planning of integrated microgrids for power generation, grid, load, and energy storage in highway service areas, transportation hubs, park charging stations, and similar scenarios. These scenarios generally have the following characteristics: the load is affected by vehicle arrivals, charging demands, and basic energy consumption in the service area; the source-side resources are mainly distributed photovoltaic power on rooftops, carports, or surrounding developable spaces, and distributed wind power can be configured according to site conditions; the planning stage typically includes geographical location, preliminary equipment parameters, and public meteorological data, but lacks long-term on-site solar and wind measurement and operational monitoring data.
[0026] This invention does not directly provide the optimal photovoltaic or wind power capacity during the source-side modeling stage. Instead, it represents the source-side renewable energy as a unit capacity available output scenario. The subsequent capacity configuration model can convert the unit capacity scenario into actual output based on decision variables, thereby solving for photovoltaic capacity, wind power capacity, energy storage capacity, grid-connected capacity, and DC bus capacity in the same optimization model.
[0027] In one embodiment, step S100, acquiring multi-day meteorological data of the configured highway service area includes: reading the service area's geographical coordinates, time range, rooftop or carport photovoltaic installation parameters, wind turbine candidate parameters, public meteorological data, reanalysis wind field data, and optional field correction data, and recording the data source, year, time step, and applicable boundaries.
[0028] In one embodiment, step S200, based on meteorological data, calculates the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, including: The usable output per unit capacity of photovoltaic (PV) systems is calculated based on solar geometry, tilt surface irradiance, module temperature, system losses, and conversion efficiency. For nighttime and periods with insufficient solar altitude angle, Take a deterministic zero value; for the effective daytime power generation period, retain correctable parameters such as irradiance disturbance, temperature reduction and shading loss. To calculate the usable output per unit capacity of wind power, the reference wind speed is corrected to the hub height, and then combined with air density, wind direction range, power curve, site reduction, and equipment availability to obtain the usable output per unit capacity of wind power. The output is zero below the cut-in wind speed and above the cut-out wind speed. The output is mapped according to the power curve between the cut-in and rated wind speeds, and is treated as not exceeding the unit capacity boundary near the rated wind speed. By season, month, time of day, weather type, and wind direction zone and Establish conditional marginal distributions and statistically analyze capacity factor, effective utilization hours, low output probability, power duration curve, P10 / P50 / P90 quantiles, ramp rate, autocorrelation, and duration of continuous low output.
[0029] It should be noted that the conditional marginal distribution is the probability distribution of available output per unit capacity of photovoltaic and wind power established after dividing the data according to multi-dimensional conditions such as season, month, intraday time period, weather type and wind direction interval.
[0030] Let the condition be denoted as ,in Indicates month or season. Indicates a time period within an hour or day. Indicates weather type, This indicates the wind direction range. For the photovoltaic side, the wind direction range can be omitted as needed; when the sample size is insufficient, the month can be backed up to the season, the hour to the intraday time period, and the detailed weather pattern to a similar weather pattern in sequence. For those that meet the conditions Time set Let the sample size be... The empirical marginal distributions for photovoltaic and wind power are as follows: Besides empirical distributions, kernel density, Beta distributions, or mixed distributions including zero-output mass can also be used for fitting. If parametric distribution fitting is used, the shape, scale, and other parameters of the distribution under that condition can be directly obtained; if empirical distributions are used, the above statistics are obtained through the empirical distribution function. The capacity factor and effective utilization hours per unit capacity can be calculated as follows: in, The effective utilization hours corresponding to a single unit capacity curve. The effective utilization hours are calculated using a multi-scenario probability weighting. The probability of low output is... P10, P50, and P90 are calculated using the generalized inverse function of the distribution: The power duration curve is obtained by... The output samples within are obtained by arranging them in descending order. The first-order autocorrelation coefficient can be expressed as: The duration of continuous low output is identified. The population statistic is obtained by taking a continuous interval of 1. Population statistics under different conditions can be calculated according to conditional weights. Weighted summation.
[0031] In one embodiment, step S400 involves connecting the obtained light output scene and wind output scene to form a joint source-side scene, including: For the same time period and the same weather type, the solar power output scenarios and wind power output scenarios are respectively edge-distributed; among them, the weather types corresponding to the solar power output scenarios and wind power output scenarios include extreme weather; Connect the edge distributions corresponding to the light output scenario and the wind output scenario respectively.
[0032] In one embodiment, edge distributions are formed for the solar power output scene and the wind power output scene corresponding to the same time period and the same weather type, respectively; connecting the edge distributions corresponding to the solar power output scene and the wind power output scene respectively includes: Under conditions Taking "summer, 12:00 to 14:00, cloudy, southerly wind zone" as an example, data was obtained from the same timestamp. Group of landscape paired samples Establish edge distribution for photovoltaic and wind power respectively. and .
[0033] For the Performing a rank transformation on paired samples yields pseudo-observations: according to Calculate Kendall's rank correlation coefficient When using a Gaussian Copula, its linear correlation parameter can be taken as: condition The combined distribution of wind and light can be represented as follows: Among them, Gaussian Copula is: During sampling, relevant two-dimensional standard normal variables are first generated. Then, the wind and solar power output samples are obtained through the generalized inverse function of the marginal distribution: There are many discrete points between 0 and the rated value in wind and solar power output. Ordinary continuous inverse functions will lose point probabilities. The generalized inverse distribution function is used to handle marginal distributions with discrete probabilities. Using the generalized inverse distribution function can preserve the discrete probability quality of zero solar power output at night, zero power output under low wind conditions, and the area near the rated power output. When the number of conditional samples is small, two alternative methods can be used: one is to maintain the same timestamp. First, pairwise resampling directly forms an empirical joint distribution; second, photovoltaic and wind power are divided into several output intervals to form conditional probability tables. in, As a condition Simultaneously falling into the photovoltaic zone and wind power area The number of samples. Sampling is done by first... By extracting interval combinations and then extracting wind and solar output within the corresponding intervals, the above methods can avoid the disruption of the correlation between wind and solar power under the same weather conditions caused by independent sampling.
[0034] Specifically, based on daily irradiance, daily power generation, precipitation, temperature, intraday P95 ramp rate, wind speed, wind direction, average daily wind power output, and low-output events, the samples are divided into several weather types or source-side states, and a state transition matrix is established. Or daily block resampling index; under the same season, time period, and weather conditions, based on the wind-solar related structure and The combination forms a joint output sample. Given the photovoltaic capacity percentage... lWhen the combined unit output is, it can be expressed as If the capacity configuration model directly uses capacity variables, then retain... and Two columns, not fixed in advance l .
[0035] In one embodiment, step S500 involves performing state transition and resampling on the joint source-side scenario to obtain the input of the new energy output capacity configuration model. Specifically, this includes: using a time step of 1 hour, first performing state transition and resampling on historical days... Features such as total daily irradiance, daily photovoltaic power generation, average daily wind power output, precipitation, daily maximum temperature, absolute P95 ramp rate, and duration of continuous low power output are extracted, and the day is marked as the combined source-side state. For example, sunny days with high light and low wind, cloudy days with fluctuating low wind, overcast and rainy days with low light and medium to high wind, and wind and solar power synchronizing with low output.
[0036] For months Internal state Transition to state Number of times Statistical analysis was performed; after smoothing, the monthly state transition probability is: in, The total number of states on the joint source side. To prevent a smoothing coefficient with zero probability under small sample conditions, it can be set to 1, for example. When the effective sample size for a certain month is insufficient, the condition level can be rolled back to the corresponding season to increase the statistical sample size. Generate the first When sampling for the first planning year, the initial state is first extracted according to the empirical frequency of each state in the target month; subsequently, for the second planning year... Generate uniform random numbers daily. And extract the state for the next day according to the cumulative probability of the monthly state transition matrix: In historical daily block collection In the middle, a random number related to the month is selected. ,state The same complete 24-hour wind-solar paired solar power grid will allow the available output of the photovoltaic / wind power unit capacity within that grid. and Synchronously write to the sample path. For continuous states such as low output, the duration in days can be extracted from the empirical duration distribution of that state. During this duration, daily blocks in the same state are continuously sampled, and then sampling is resumed according to the state transition matrix. This process is repeated until a planning year path of 365 days or 8760 hours is formed. By repeating this process, multiple Monte Carlo planning year samples can be obtained. For the For each sample, the probability of low output during wind-solar synchronization can be expressed as: in, The set valid verification period. This is an indicator function; it takes a value of 1 when the condition within the parentheses is true, and a value of 0 otherwise. The photovoltaic side can be limited to the effective sunshine period to avoid increasing the probability of low output due to deterministic zero output at night. Let the length of each segment where the aforementioned low-output indicator variable continuously takes the value 1 be... The longest continuous low-output duration is: The absolute gradeability of the P95 with combined unit capacity output is: If the wind and solar capacity ratio is not preset in the source-side preprocessing stage, then each will be retained. and Two columns, calculated by the capacity configuration model given the capacity variables. The initial sample weights can be taken as follows: After scene reduction, the weight of the representative sample is the sum of the weights of all the original samples it represents.
[0037] In one embodiment, it further includes: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, calculate the changes in available output at adjacent time points. The event where the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold and the change in available output at the adjacent time is less than the preset minimum change is defined as a continuous low output event. A short-term ramp-up event is defined as any moment when the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold, and the change in available output at the adjacent moment is greater than the preset maximum change value.
[0038] It should be noted that both the scenarios and events are essentially datasets. Solar power output events and wind power output events are combined to form a joint source-side scenario; the weather patterns corresponding to continuous low output events and short-term ramp-up events may be extreme weather; the output changes of solar and wind power at adjacent moments are calculated separately. This forms uphill, downhill, absolute uphill, and high quantile uphill distributions; it also identifies distributions below a threshold. i The continuous low-output events form a duration distribution. L This step is used to avoid generating unrealistic jumps through independent sampling and to preserve the risk of continuous low output.
[0039] Specifically, the above-mentioned uphill, downhill, absolute uphill, and high-quantile uphill distributions include: For any unit capacity output sequence The ramp rate of adjacent time steps is defined as: Accordingly, the uphill and downhill amplitudes and the absolute slope are defined as follows: The empirical distributions of the three types of climbing samples can be uniformly represented as: The absolute climbing rate of the high quantile is: In one embodiment, after performing state transition and resampling on the joint source-side scene, the method further includes: If the events corresponding to the formed joint source-side scenario include short-term ramp events and continuous low-output events, then the joint source-side scenario is identified. Clustering and filtering are performed on unidentified joint source-side scenarios; The selected joint source-side scenarios are weighted by their sampling frequency to obtain standardized source-side data packets for use by the capacity configuration model; the sampling frequency is obtained by the proportion of a certain joint source-side scenario in all joint source-side scenarios.
[0040] Specifically, multiple planning year samples are generated based on the state transition matrix, duration distribution, and daily block resampling rules. The daily block length can be set to 24 hours or adjusted to multiple time steps according to the time resolution, so that the samples simultaneously retain intraday morphology, hourly slope, and continuous weather processes.
[0041] Furthermore, under the same season, time period, and weather conditions, based on the wind-sun related structure, and The combination forms a joint output sample. Given the photovoltaic capacity percentage... When the combined unit output is, it can be expressed as If the capacity configuration model directly uses capacity variables, then retain... and Two columns, not fixed in advance .
[0042] In one embodiment, multiple planning year samples are generated based on the state transition matrix, duration distribution, and daily block resampling rules, including: set up For the first Path number The combined source-side state of the day. First, generate random numbers. The state for the next day is extracted from the state transition matrix. Then, the index of a day block is randomly selected from the set of historical day blocks with the same state in the same month. The time step within the day Write the selected scenic spot pairings into the entire sample path: Among them, hourly data If a continuous low-output state is encountered, start by analyzing the duration distribution. The system extracts consecutive days and maintains the corresponding state or continuously extracts daily blocks in the corresponding state during this duration. Daily blocks use a 1-hour time resolution. H d =24, concatenating 365 daily blocks yields an 8760h Monte Carlo planning year path; change the random number seed and repeat. This yields multiple planning year samples.
[0043] It is important to emphasize that the solar and wind power curves in the daily data block are extracted in pairs according to the original timestamp, rather than extracted independently. Therefore, it is possible to simultaneously maintain the intraday pattern, hourly ramp-up, weather process continuity, and wind-solar correlation.
[0044] In one specific embodiment, to generate the first s Taking the Monte Carlo sample path as an example, the initial joint source-side state is set for the first day. This represents a cloudy, moderately sunny condition, corresponding to the month. m 1=6, generate random numbers Query the state transition matrix to obtain the state for day 2. =1 (High output on sunny days, month) m 2=6).
[0045] against d =1, in the historical daily block set of June, state 2 Randomly select daily block indexes according to a uniform distribution. =12, retrieve the original paired time sequence of this historical day block: =[0,0.12,0.35,...,0] , =[0.42,0.38,...,0.45] , h =1,2,...,24 Write the entire 24-hour landscape sequence into the first... s Path number d =1 day.
[0046] Similarly, d =2 in the set Extracting the index =27, read the paired scenery time sequence and concatenate it to the path.
[0047] Repeat this state transition-daily block extraction-assembly process for 365 days to obtain a complete 8760h Monte Carlo year sample; change the random seed and repeat. N s Multiple sets of planning year samples can be obtained in one step. This process realizes the joint embedding of state transition, duration constraints, and wind and solar time series coupling into Monte Carlo sampling.
[0048] It should be noted that clustering screening refers to the reduction of clusters using K-medoids, hierarchical clustering, Wasserstein distance, fastforward selection, or other representative scenario screening methods, and calculating the scenario probability based on sample frequency or calibrated weights. Risk samples with mandatory retention flags are not deleted due to cluster center distance or low sample frequency, but are used separately in the expected cost and safety verification items; standardized source-side data packages are provided for use by the capacity configuration model; the photovoltaic and wind power data packages contain unit capacity curves, and the capacity configuration model uses capacity variables... , It is converted into actual output, and at the same time, the scenario probability and risk label are read to calculate the expected indicators and verify the safety boundary.
[0049] The "expected cost item" and "security verification item" are two uses of standardized source-side data in the subsequent capacity configuration model, and are not two new outputs added in the source-side preprocessing stage. The source-side data packets provide distinguishing information to the subsequent model through "scenario probability weights" and "risk / mandatory retention identifiers".
[0050] The expected cost term is a probability-weighted objective calculated based on scenario probabilities, and can be expressed as: in, For decision variables including photovoltaic, wind power, energy storage, grid connection, and bus capacity, For the annualized cost of investment, This may include electricity purchase costs, wind and solar curtailment costs, energy storage losses or attenuation costs, operation and maintenance costs, and costs of penalties for power outages. If a risk sample has a calibrable probability of occurrence, its original or calibrated probability will be used in the calculation of the expected value.
[0051] Safety verification items are hard constraints or boundary checks applied to each of the risk scenarios in the set of risk scenarios, and can be uniformly represented as: This type of constraint does not change with scenario probability. Smaller, more relaxed or removed requirements may include: power balance at various times, upper and lower limits of energy storage state of charge, upper limit of grid connection power, upper limit of allowable power shortage for important loads, continuous support capability of energy storage during periods of continuous low output, and power regulation capability during periods of high ramp-up.
[0052] Specifically, the scenario probability is calculated based on sample frequency or calibration weights, including: assuming that it has been generated. The Monte Carlo method is used to plan annual paths. First, a standardized feature vector is generated for each path: By combining the complete sustained wind and solar power curves, the combined distance between any two paths can be defined: in, The first-order Wasserstein distance. to The distance weights are normalized.
[0053] In the K-medoids implementation, samples with mandatory retention flags are first grouped into a risk set. These samples will not be included in the deletion of ordinary samples; for the remaining samples, selection will be made. An initial set of centers is used, and each non-risk sample is assigned to the nearest center. Then, within each cluster, the true sample that minimizes the sum of weighted distances within the cluster is selected as the new center. The new center of each cluster satisfies: Repeat the "assign-update" process until the center no longer changes. (Representative scenario) The probability weights are: Risk samples retain their original or calibrated weights. For example, 20 typical representative paths can be selected from 1000 planning year paths; separately identified risk paths are retained separately from the 20 typical representative paths, and this number is not included in the target number of typical scenarios; the above number is only an example and can be adjusted according to the scale of the calculation.
[0054] Specific processing implementation example: Assuming the original Monte Carlo sample contains 1000 routes for each planning year, five extreme risk routes are identified and marked for mandatory retention, thus being included in the risk sample set. S risk These 5 samples do not participate in the K-medoids clustering iteration.
[0055] K-medoids clustering was performed on the remaining 995 ordinary samples, with a target number of clusters set. K=20, resulting in 20 typical representative scenarios, each with its own probability weight. P k .
[0056] It should also be noted that the standardized source-side data includes the following fields: When the capacity configuration model calls this interface, the actual output can be calculated according to the following relationship: in, s Number the scene. t For time indexing, and These are decision variables in the capacity configuration model; this invention does not require them to be determined in advance during the source-side scenario generation stage. and .
[0057] Taking the microgrid planning of a service area on a southern highway as an example. During the planning phase, the service area's latitude and longitude, the area suitable for photovoltaic installation, the photovoltaic tilt angle, the candidate wind turbine hub height, and the general power curve were known. Hourly solar irradiance, temperature, precipitation, and reanalysis wind speed and direction data were also available. The system was implemented according to the following steps.
[0058] First, we read hourly meteorological samples for 8760 hours a year to calculate variables such as total horizontal irradiance, direct normal irradiance, diffuse irradiance, temperature and precipitation on the photovoltaic side, and calculate the wind speed and wind direction at the reference height on the wind power side.
[0059] Second, based on the service area coordinates and photovoltaic installation parameters, calculate solar geometry, tilt surface irradiance, module temperature, and system losses to form a usable output curve for 1kWp of photovoltaic unit capacity. Specifically, first calculate the solar zenith angle, solar altitude angle, and solar azimuth angle from latitude, longitude, date, and time; then calculate the incident angle from the module tilt angle, module azimuth angle, and solar azimuth angle; subsequently, convert the total horizontal irradiance, direct normal irradiance, and diffuse irradiance into module planar irradiance, and superimpose factors such as shading, dust, module mismatch, cable loss, equipment availability, module temperature, and inverter or DC conversion efficiency to finally obtain hourly output. In the example, the annual power generation per unit capacity can fall within the range of approximately 900 to 1200 kWh / kWp, with the specific value varying depending on meteorological year, shading, tilt angle, temperature coefficient, and system losses.
[0060] Third, based on wind speed, wind direction, hub height, wind shear, and turbine power curves, a usable output curve for a 1kW wind power unit capacity is generated. Specifically, the reference height wind speed is first corrected to the candidate hub height using a power law or logarithmic law, and then corrected for air density based on temperature and air pressure. Subsequently, site reductions caused by road ancillary buildings, green belts, billboards, slopes, and wakes are superimposed on the wind direction interval, and the equivalent hub height wind speed is input into the manufacturer's power curve or a piecewise function of cut-in, rated, and cut-out wind speeds. Finally, the curve is normalized according to site reductions and equipment availability and truncated to 0 to 1 to obtain hourly outputs. If wind resources are average, wind power can be included in the planning model as an extended power source or a sensitivity variable, rather than being the default benchmark power source.
[0061] Fourth, based on daily sample statistics of photovoltaic daily power generation, daily total horizontal irradiance, precipitation, daily maximum temperature, and daily P95 ramp rate, photovoltaic systems are classified into types such as sunny high irradiance, cloudy fluctuating irradiance, cloudy and rainy low irradiance, high temperature and high irradiance, and continuous low photovoltaic risk.
[0062] Fifth, based on the unit capacity output after wind speed is converted through the power curve, wind power is classified into types such as zero output in low wind, power generation in weak wind, stable output in medium wind, high output in high wind, and power generation limited or shut down in strong wind.
[0063] Sixth, the system combines daily solar weather patterns with daily average wind power output to form joint source-side scenarios. For example, the system can generate joint scenarios such as sunny high light and low wind, sunny high light and medium wind, cloudy with fluctuating low wind, overcast and rainy with low light and medium to high wind, simultaneous wind and solar power with low output, high temperature with high light and low wind, and extreme continuous low output. For simultaneous wind and solar power with low output and extreme continuous low output scenarios, the system writes a mandatory reservation flag.
[0064] Seventh, several planning year samples are generated by using daily state transition and daily block resampling in the same month and state. Each sample includes indicators such as photovoltaic unit capacity output, wind power unit capacity output, scenario weight, low output probability, longest continuous low output duration, and P95 ramp rate.
[0065] Eighth, the source-side data packets are passed to the capacity configuration model. The capacity configuration model does not directly use a single average curve, but instead solves for photovoltaic capacity, wind power capacity, energy storage capacity, grid-connected capacity, and DC bus capacity under multiple source-side scenarios based on scenario probabilities and risk labels.
[0066] This invention also provides a source-side preprocessing system for microgrid planning in highway service areas, comprising: The data acquisition module is used to acquire meteorological data for multiple days from the configured highway service areas; The power output modeling module calculates the available power output per unit capacity of photovoltaic power and the available power output per unit capacity of wind power based on meteorological data. The three-layer time-series probabilistic modeling module is used to divide each day into preset solar power output events and wind power output events according to the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power. The wind and solar joint sampling module is used to connect the separately obtained solar power output scenarios and wind power output scenarios to form a joint source-side scenario; The source-side scenario library and planning interface module are used to perform state transitions and resampling on the joint source-side scenarios to obtain the input of the new energy output capacity configuration model.
[0067] It should be noted that the system provided in this embodiment is used to execute any of the above methods.
[0068] In one specific embodiment, the source-side preprocessing system for planning microgrids in highway service areas includes: The data acquisition module receives service area coordinates, altitude, available rooftop or carport area, photovoltaic tilt and azimuth angles, wind turbine hub height, wind turbine power curves, system loss coefficients, equipment availability, public meteorological data, reanalysis wind field data, and optional on-site photometric and wind measurement data. Data from different sources are recorded with different data source labels to prevent public meteorological data, reanalysis wind field data, and on-site measured data from being lumped together as the same observation sample. The power output modeling module includes a photovoltaic (PV) physical power output modeling module. This module calculates the solar altitude angle, solar azimuth angle, and module planar irradiance based on service area coordinates and time index. It then combines this with total horizontal irradiance, direct normal irradiance, diffuse irradiance, surface albedo, module temperature model, system losses, inverter efficiency, or DC conversion efficiency to obtain the usable power output per unit capacity of PV. Nighttime and periods with insufficient solar altitude angles are treated as deterministic zero power output, while effective daytime power generation periods are then entered into probabilistic modeling. The wind power physical power output modeling module calculates the usable power output per unit capacity of wind power based on reference height wind speed, wind direction, hub height, wind shear index, air density, terrain and obstacle reduction, equipment availability, and turbine power curves. Wind speed is defined by cut-in wind speed, rated wind speed, and cut-out wind speed, forming a zero-output point mass, a continuous output range, and a probabilistic accumulation near the rated output. The unit capacity normalization module converts the actual available output of photovoltaic (PV) and wind power into unit capacity variables. Let the available output per unit capacity of PV be... The available output per unit capacity of wind power is Both values range from 0 to 1 or the corresponding unit capacity power value. This module ensures that subsequent capacity configuration models can be passed. , Calculate the actual output, where and To optimize the capacity decision variables in the model; The three-layer temporal probability modeling module models photovoltaic and wind power as three types of probabilistic objects: marginal distribution of sub-conditions, distribution of conditional changes between adjacent moments, and distribution of continuous low output duration. The marginal distribution of sub-conditions describes the amount of power that may be generated at the time of order placement in a certain season, time period, weather type, or wind direction range; the distribution of conditional changes between adjacent moments constrains the range of change from the previous moment to the next moment; the distribution of continuous low output duration describes whether low irradiance, low wind, or low output during wind-solar synchronization will continuously span multiple hours or multiple intraday cycles. The weather-type and state transition module aggregates features such as daily irradiance, daily power generation, precipitation, temperature, intraday fluctuations, average wind speed, wind direction, and low-output events into weather-type or source-side states, such as sunny with high irradiance, cloudy with fluctuations, overcast with low irradiance, high temperature with high irradiance, low wind, medium to high wind, and wind-solar synchronous low output. This module establishes state transition probabilities, duration distributions, or daily block resampling indexes to maintain the continuity of weather processes. The wind-solar joint sampling module extracts variables from both the photovoltaic and wind power sides under given seasonal, time-of-day, and weather conditions, and establishes the wind-solar joint relationship through correlation coefficients, Copula, empirical joint distributions, or conditional probability tables. Sample generation prioritizes meteorological variables, which are then converted into unit capacity output using a physical model. In cases of insufficient data or rapid planning analysis, sampling can also be performed directly from the calibrated empirical distribution of unit capacity output. The scenario reduction and risk retention module clusters or selects representative scenarios from a large number of Monte Carlo samples and calculates the probability of typical scenarios based on sample weights. Unlike methods that only retain representative samples near the mean, this module sets mandatory retention flags for continuous low photovoltaic power, sustained low wind power, low output during wind and solar synchrotron power generation, extreme continuous low output power, and short-term large ramp-up scenarios before reduction. This ensures that low-frequency, high-impact scenarios can still be used as risk constraints or safety verification scenarios in subsequent capacity configuration models.
[0069] The source-side scenario library and planning interface module outputs a standard data package, including a time index, available output per unit capacity for photovoltaic (PV) power, available output per unit capacity for wind power, scenario number, scenario probability weight, season or month label, weather type or status label, typical or extreme scenario identifier, capacity factor, P95 absolute ramp rate, longest continuous low output duration, probability of low output during wind-solar synchronization, and data source description. The capacity configuration model can directly read this data package and automatically determine the PV and wind power capacities during the optimization process.
[0070] This invention also provides an electronic device, which includes, but is not limited to: Memory, used to store programs; The processor is used to execute programs stored in memory. When the processor executes programs stored in memory, it is used to execute the aforementioned source-side preprocessing method for microgrid planning in highway service areas.
[0071] The processor and memory can be connected via a bus or other means.
[0072] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the method described in the embodiments of the present invention. The processor implements the above method by running the non-transitory software program and instructions stored in the memory.
[0073] The memory may include a program storage area and a data storage area, wherein the program storage area may store the operating system and application programs required for at least one function; the data storage area may store data for executing the methods described above. Furthermore, the memory may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0074] The non-transitory software program and instructions required to implement the above terminal selection method are stored in memory and are executed by one or more processors.
[0075] This invention also provides a storage medium storing computer-executable instructions for performing the above-described methods.
[0076] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors.
[0077] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0078] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.
[0079] This document describes embodiments of the invention, including preferred embodiments known to the inventors for carrying out the invention. Variations of these embodiments will become apparent to those skilled in the art upon reading the foregoing description. The inventors encourage those skilled in the art to adopt such variations as appropriate, and the inventors intend to practice embodiments of the invention in ways other than those specifically described herein. Therefore, the scope of the invention includes all modifications and equivalents of the subject matter set forth in the appended claims, as permitted by applicable law. Furthermore, the scope of the invention covers any combination of the foregoing elements in all possible variations thereof, unless otherwise indicated herein or otherwise clearly contradicted by the context.
Claims
1. A source-side preprocessing method for microgrid planning in highway service areas, characterized in that, include: Obtain multiple days of meteorological data for the configured highway service areas; Based on the meteorological data, calculate the available power output per unit capacity of photovoltaic power and the available power output per unit capacity of wind power respectively. Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, each day is divided into preset photovoltaic output events and wind output events, respectively. By connecting the separately obtained light output scenarios and wind output scenarios, a joint source-side scenario is formed; The state transition and resampling of the joint source-side scenario are performed to obtain the input of the new energy output capacity configuration model.
2. The source-side preprocessing method for microgrid planning in highway service areas according to claim 1, characterized in that, The light output scene and wind output scene obtained by the connection respectively form a joint source-side scene, including: For the same time period and the same weather type, the solar power output scenarios and wind power output scenarios are respectively edge-distributed; among them, the weather types corresponding to the solar power output scenarios and wind power output scenarios include extreme weather; Connect the edge distributions corresponding to the light output scenario and the wind output scenario respectively.
3. The source-side preprocessing method for microgrid planning in highway service areas according to claim 1, characterized in that, Also includes: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, calculate the changes in available output at adjacent time points. The event where the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold and the change in available output at the adjacent time is less than the preset minimum change is defined as a continuous low output event. A short-term ramp-up event is defined as any moment when the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold, and the change in available output at the adjacent moment is greater than the preset maximum change value.
4. The source-side preprocessing method for microgrid planning in highway service areas according to claim 3, characterized in that, After performing state transition and resampling on the joint source-side scenario, the method further includes: If the events corresponding to the formed joint source-side scenario include short-term ramp events and continuous low-output events, then the joint source-side scenario is identified. Clustering and filtering are performed on unidentified joint source-side scenarios; The selected joint source-side scenarios are weighted by their sampling frequency to obtain standardized source-side data packets for use by the capacity configuration model; the sampling frequency is obtained by the proportion of a certain joint source-side scenario in all joint source-side scenarios.
5. A source-side preprocessing system for microgrid planning in highway service areas, characterized in that, include: The data acquisition module is used to acquire meteorological data for multiple days from the configured highway service areas; The power output modeling module calculates the available power output per unit capacity of photovoltaic power and the available power output per unit capacity of wind power based on the meteorological data. The three-layer time-series probabilistic modeling module is used to divide each day into preset solar power output events and wind power output events according to the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power. The wind and solar joint sampling module is used to connect the separately obtained solar power output scenarios and wind power output scenarios to form a joint source-side scenario; The source-side scenario library and planning interface module are used to perform state transitions and resampling on the joint source-side scenarios to obtain the input of the new energy output capacity configuration model.
6. The source-side preprocessing system for microgrid planning in highway service areas according to claim 5, characterized in that, The light output scene and wind output scene obtained by the connection respectively form a joint source-side scene, including: For the same time period and the same weather type, the solar power output scenarios and wind power output scenarios are respectively edge-distributed; among them, the weather types corresponding to the solar power output scenarios and wind power output scenarios include extreme weather; Connect the edge distributions corresponding to the light output scenario and the wind output scenario respectively.
7. A source-side preprocessing system for microgrid planning in highway service areas according to claim 5, characterized in that, Also includes: Based on the available output per unit capacity of photovoltaic power and the available output per unit capacity of wind power, calculate the changes in available output at adjacent time points. The event where the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold and the change in available output at the adjacent time is less than the preset minimum change is defined as a continuous low output event. A short-term ramp-up event is defined as any moment when the available output per unit capacity of photovoltaic power or the available output per unit capacity of wind power is lower than the corresponding threshold, and the change in available output at the adjacent moment is greater than the preset maximum change value.
8. A source-side preprocessing system for microgrid planning in highway service areas according to claim 7, characterized in that, After performing state transition and resampling on the joint source-side scenario, the method further includes: If the events corresponding to the formed joint source-side scenario include short-term ramp events and continuous low-output events, then the joint source-side scenario is identified. Clustering and filtering are performed on unidentified joint source-side scenarios; The selected joint source-side scenarios are weighted by their sampling frequency to obtain standardized source-side data packets for use by the capacity configuration model; the sampling frequency is obtained by the proportion of a certain joint source-side scenario in all joint source-side scenarios.
9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method as described in any one of claims 1 to 4.
10. A storage medium, characterized in that, The device stores computer-executable instructions for performing the method as described in any one of claims 1 to 4.
