Space-time complementary wind-solar grid-connected inversion control method and system

Abstract

The invention discloses a space-time complementary wind-solar grid-connected inversion control method and system, which is characterized in that an output rule is pre-judged by constructing a space-time prediction model, output weights are dynamically distributed in combination with complementary characteristic data and a power optimization algorithm, short-time power buffer requirements are accurately calculated, and the time-space complementary wind-solar grid-connected inversion control is realized. And when the demand exceeds the threshold value, the output is regulated and controlled through the energy storage module and the decoupling control strategy, so that the power stability is ensured. Meanwhile, a multi-level conversion structure and a harmonic suppression device are adopted to optimize a current signal, it is ensured that the output current and the power grid have the same frequency and phase, the resource allocation efficiency is improved through cross-regional scheduling optimization, and finally the remarkable improvement of the grid-connected conversion efficiency and the stable guarantee of the current quality are achieved. According to the method, the problems of volatility and stability in the renewable energy grid connection process are effectively solved, and reliable technical support is provided for large-scale renewable energy access to a power grid.

Description

Technical Field

[0001] This invention relates to the field of wind and solar power grid connection technology, and in particular discloses a spatiotemporally complementary wind and solar grid-connected inverter control method and system. Background Technology

[0002] Against the backdrop of energy transition, renewable energy grid connection technology has become a crucial pillar for promoting green development. Wind and solar energy, as core components of clean energy, have irreplaceable value in ensuring energy security and reducing carbon emissions when integrated into the grid on a large scale. However, how to stably and efficiently integrate these energy sources into the grid remains a key focus for the industry.

[0003] Current technological solutions often struggle to address the inherent instability and uneven regional distribution of wind and solar power when dealing with grid integration issues. Many methods lack a comprehensive consideration of resource characteristics across time and space when addressing energy output fluctuations, leading to challenges to grid stability in the face of sudden output changes or regional resource disparities. This limitation frequently results in inefficiencies or poor power quality in grid-connected systems during actual operation.

[0004] A deeper technical challenge lies in achieving coordinated control of wind and solar power output across both temporal and spatial dimensions. Temporally, wind and solar power output is significantly affected by day-night cycles and seasonal variations, exhibiting marked periodic differences. Spatially, wind speeds and sunlight conditions vary greatly across different regions, and the uneven distribution of resources further exacerbates the difficulty of control. Due to the failure to effectively balance these two dimensions, the system often cannot respond precisely to complex scenarios, thus affecting the overall grid connection performance. Specifically, in real-world business scenarios, a region may experience excess photovoltaic power generation during the day when sunlight is abundant, while wind power output may be underutilized during nighttime when winds are strong, leading to energy waste. Simultaneously, the grid must withstand the pressure from sudden power fluctuations.

[0005] How to smooth out the difference in power output between day and night in the time dimension and rationally allocate regional resources in the spatial dimension to avoid energy waste and grid impact has become a key problem that this study urgently needs to solve. Summary of the Invention

[0006] This invention provides a spatiotemporally complementary wind-solar grid-connected inverter control method and system, aiming to solve at least one of the defects existing in the prior art.

[0007] One aspect of the present invention relates to a spatiotemporally complementary wind-solar grid-connected inverter control method, comprising the following steps: Step S100: Obtain real-time data from regional wind speed data and light intensity data through a data acquisition device, and construct a spatiotemporal prediction model by combining it with a historical meteorological database. The spatiotemporal prediction model is based on time series analysis to determine the power output pattern in order to predict the fluctuation characteristics of renewable energy. Step S200: Based on the power output pattern determined by the spatiotemporal prediction model, obtain complementary characteristic data from renewable energy resources, dynamically allocate power output weights using a power optimization algorithm, and obtain short-term power buffer demand. The complementary characteristic data reflects the energy complementary power generation mode, and the short-term power buffer demand represents the buffer capacity required for power fluctuations. Step S300: Determine whether the short-term power buffer demand exceeds the threshold. If it does, obtain backup energy from the energy storage module, regulate the energy output through a decoupling control strategy, and obtain the regulated output data. The energy storage module includes a battery system. Step S400: For the output data after regulation, a multi-level conversion structure is used to obtain the current signal from the inverter layer, and the harmonics are suppressed by the harmonic suppression device to determine that the output current is in phase and frequency with the grid. The inverter layer is responsible for DC to AC conversion. Step S500: Obtain total harmonic distortion (THD) data from the current signal after harmonic suppression. If the THD data is below the threshold, it is incorporated into cross-regional scheduling optimization to obtain spatial resource allocation to improve grid connection conversion efficiency. The THD data measures current quality, and cross-regional scheduling optimization involves resource allocation algorithms.

[0008] Further, step S100 includes: Step S110: Obtain wind speed data and light intensity data from the target area through a data acquisition device, and use a pre-established sensor network to collect wind speed data and light intensity data at regular intervals to obtain a preliminary environmental monitoring dataset. Step S120: Based on the preliminary environmental monitoring dataset, integrate the historical meteorological database to construct a spatiotemporal prediction model.

[0009] Further, step S200 includes: Step S210: Based on the power output pattern determined by the spatiotemporal prediction model, obtain complementary characteristic data from the target renewable energy resources, classify them according to different power generation modes, and obtain the classified resource characteristic set. Step S220: Based on the resource characteristic set, use a power optimization tool to dynamically adjust the output weight. If the dynamically adjusted output weight exceeds the preset threshold, reduce the corresponding ratio to obtain the weight allocation scheme. Step S230: According to the weight allocation scheme, use the power balance calculation tool to monitor fluctuations in real time, obtain short-term power fluctuation data, and determine the initial buffer demand range. Step S240: For short-term power fluctuation data within the initial buffer demand range, a data integration platform is used for secondary calibration to determine whether the balance requirements are met, and the short-term power buffer demand value is obtained.

[0010] Further, step S300 includes: Step S310: Determine whether the short-term power buffer demand exceeds the threshold. If the short-term power buffer demand exceeds the preset threshold range, trigger the call command and obtain the energy scheduling signal. Step S320: Obtain backup energy from the energy storage module according to the energy dispatch signal, use the distribution tool to perform stratified processing of the backup energy, and determine the stratified distribution scheme; Step S330: For the stratified allocation scheme, use control tools to dynamically adjust the energy output and obtain the output data after regulation.

[0011] Further, step S400 includes: Step S410: Based on the output data after regulation, a preliminary current signal is extracted from the inverter layer using a multi-level conversion structure. The preliminary current signal is then preliminarily processed to obtain the processed basic current data. Step S420: If there are interference fluctuations in the basic current data, the basic current data is filtered by a harmonic suppression device to determine the filtered stable current data. Step S430: Use a frequency detection tool to compare and analyze the frequency characteristics of the stable current data to obtain adjusted current data that is consistent with the grid frequency. Step S440: Use a phase calibration tool to perform phase comparison and correction on the adjusted current data, determine whether it has reached the state of being in phase with the power grid, and obtain the final output current signal.

[0012] Further, step S500 includes: Step S510: Extract total harmonic distortion rate data from the current signal after harmonic suppression using signal analysis tools to obtain a standardized distortion rate index. Step S520: If the standardized distortion rate index is lower than the preset threshold, then the data comparison tool is used to perform condition verification to determine the available distortion rate reference value. Step S530: Use a data fusion tool to integrate the available distortion rate reference value with cross-regional scheduling information to obtain the fused scheduling input data; Step S540: The fused scheduling input data is processed for spatial resource allocation using a resource allocation algorithm tool to obtain an optimized regional resource distribution scheme; Step S550: For the optimized regional resource distribution scheme, continuously monitor the changes in the distortion rate index to ensure the matching degree with cross-regional scheduling information.

[0013] Another aspect of the present invention relates to a spatiotemporally complementary wind-solar grid-connected inverter control system for executing the above-described spatiotemporally complementary wind-solar grid-connected inverter control method, comprising: The spatiotemporal prediction model construction module is used to acquire real-time information from regional wind speed data and light intensity data through data acquisition devices, and to construct a spatiotemporal prediction model by combining it with historical meteorological database. The spatiotemporal prediction model is based on time series analysis method to determine the power output pattern in order to predict the fluctuation characteristics of renewable energy. The short-term power buffer demand acquisition module is used to obtain complementary characteristic data from renewable energy resources based on the output pattern determined by the spatiotemporal prediction model, and dynamically allocate output weights using a power optimization algorithm to obtain the short-term power buffer demand. The complementary characteristic data reflects the energy complementary power generation mode, and the short-term power buffer demand represents the buffer capacity required for power fluctuations. The output data acquisition module is used to determine whether the short-term power buffer demand exceeds the threshold. If it does, it obtains backup energy from the energy storage module, regulates the energy output through a decoupling control strategy, and obtains the regulated output data. The energy storage module includes a battery system. The current signal acquisition module is used to acquire the current signal from the inverter layer based on the regulated output data using a multi-level conversion structure. Harmonics are suppressed by a harmonic suppression device to ensure that the output current is in phase and frequency with the power grid. The inverter layer is responsible for DC to AC conversion. The spatial resource allocation acquisition module is used to obtain total harmonic distortion (THD) data from the current signal after harmonic suppression. If the THD data is lower than the threshold, it is incorporated into cross-regional scheduling optimization to obtain spatial resource allocation to improve grid connection conversion efficiency. The THD data measures the current quality, and the cross-regional scheduling optimization involves resource allocation algorithms.

[0014] The beneficial effects achieved by this invention are as follows: This invention discloses a spatiotemporally complementary wind-solar grid-connected inverter control method and system. Addressing the challenges of high power output volatility, high power buffering requirements, and unstable grid-connected current quality in renewable energy scenarios, it constructs a spatiotemporal prediction model to predict power output patterns. Combining complementary characteristic data and power optimization algorithms, it dynamically allocates power output weights, accurately calculates short-term power buffering requirements, and regulates output through energy storage modules and decoupling control strategies when demand exceeds thresholds, ensuring power stability. Simultaneously, it employs a multi-level conversion structure and harmonic suppression devices to optimize the current signal, ensuring that the output current is in phase and frequency with the grid. Furthermore, it improves resource allocation efficiency through cross-regional scheduling optimization, ultimately achieving a significant improvement in grid-connected conversion efficiency and stable current quality. This invention effectively solves the volatility and stability problems in the renewable energy grid connection process, providing reliable technical support for large-scale renewable energy grid integration. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating an embodiment of a spatiotemporally complementary wind-solar grid-connected inverter control method of the present invention. Figure 2 This is a functional block diagram of an embodiment of a spatiotemporally complementary wind-solar grid-connected inverter control system of the present invention.

[0016] Explanation of icon numbers: 10. Spatiotemporal prediction model construction module; 20. Short-term power buffer demand acquisition module; 30. Output data acquisition module; 40. Current signal acquisition module; 50. Spatial resource allocation acquisition module. Detailed Implementation

[0017] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0018] like Figure 1 As shown, the first embodiment of the present invention proposes a spatiotemporally complementary wind-solar grid-connected inverter control method, including the following steps: Step S100: Obtain real-time data from regional wind speed data and solar intensity data through a data acquisition device, and construct a spatiotemporal prediction model by combining it with a historical meteorological database. The spatiotemporal prediction model is based on time series analysis to determine the power output pattern in order to predict the fluctuation characteristics of renewable energy.

[0019] Using professional data acquisition devices (such as wind speed sensors, light meters, and meteorological monitoring terminals), dynamic wind speed data and real-time light intensity data of the target area are captured in real time to form complete real-time acquisition information. By integrating long-term meteorological data of the region from historical meteorological databases (such as diurnal variations, seasonal fluctuations, and characteristics of extreme weather changes in past wind speed / sunshine), a spatiotemporal prediction model covering both time and space dimensions is constructed. This model is based on time series analysis methods (such as ARIMA (Autoregressive Integrated Moving Average Model), LSTM (Long Short-Term Memory) neural network, Prophet model, etc.). By mining the correlation between real-time and historical data, the model analyzes the power output changes of wind and solar energy over time and spatial distribution, and determines the power output patterns of renewable energy. Based on these power output patterns, the model accurately predicts the power fluctuation characteristics of renewable energy (wind and solar energy) (such as fluctuation amplitude, fluctuation frequency, peak occurrence time, etc.), requiring a prediction accuracy of ≥88% and a short-term fluctuation prediction error of ≤10%, providing a scientific data basis for subsequent power optimization allocation and grid connection regulation.

[0020] Step S200: Based on the power output pattern determined by the spatiotemporal prediction model, complementary characteristic data is obtained from renewable energy resources. Power optimization algorithm is used to dynamically allocate power output weights to obtain short-term power buffer demand. The complementary characteristic data reflects the energy complementary power generation mode, and the short-term power buffer demand represents the buffer capacity required for power fluctuations.

[0021] Based on the wind and solar power output patterns (such as output periods, peak ranges, and fluctuation trends) output by the spatiotemporal prediction model in step S100, complementary characteristic data is extracted from renewable energy resources. This complementary characteristic data intuitively reflects the complementary power generation modes of wind and solar power (such as high solar power output during the day when there is sufficient sunlight, and wind power output being dominant at night / when there is abundant wind, or the power peaks of the two being staggered and their fluctuation trends being opposite, etc.). Power optimization algorithms adapted to wind-solar complementary scenarios (such as particle swarm optimization, genetic algorithms, and model predictive control algorithms) are adopted, combined with real-time power output status and grid acceptance requirements, to dynamically adjust the output weights of wind and solar power (i.e., the proportion of the two in the total power output). Through weight optimization, some natural power fluctuations are offset. Then, the buffer capacity required for the remaining power fluctuations is quantitatively calculated, i.e., the short-term power buffer demand. The weight allocation response delay is required to be ≤50ms and the buffer demand calculation error is required to be ≤8%, providing clear quantitative targets for subsequent energy storage call-up and power output regulation.

[0022] Step S300: Determine whether the short-term power buffer demand exceeds the threshold. If it does, obtain backup energy from the energy storage module, regulate the energy output through a decoupling control strategy, and obtain the regulated output data. The energy storage module includes a battery system.

[0023] Using the short-term power buffer demand obtained in step S200 as the judgment object, a clear power buffer threshold is first set based on the grid acceptance capacity, equipment operating limits, and grid connection quality standards. Through quantitative comparison, it is determined whether the short-term power buffer demand exceeds this threshold; if it does, the energy storage module (which may be paired with auxiliary units such as supercapacitors) containing the battery system is activated to retrieve its stored reserve energy to compensate for the output gap or absorb excess power. Simultaneously, a decoupling control strategy (such as PID (Proportional-Integral-Derivative) decoupling and model prediction decoupling) is adopted to separate the output control of wind power, solar power, and energy storage modules into independent channels, respectively regulating the output magnitude and response rhythm of each channel to achieve coordinated matching of multi-energy output. The final output data after regulation meets the grid connection requirements, requiring a regulation response delay ≤100ms and output fluctuation controlled within ±5%, providing stable energy input for subsequent inverter conversion.

[0024] Step S400: For the output data after regulation, a multi-level conversion structure is used to obtain the current signal from the inverter layer, and the harmonics are suppressed by the harmonic suppression device to determine that the output current is in phase and frequency with the power grid. The inverter layer is responsible for DC to AC conversion.

[0025] Using the stable output data (DC form) obtained in step S300 as input, a multi-level conversion structure (such as a three-level NPC (Neutral Point Clamped) topology or a modular multi-level topology) is adopted. Relying on the inverter layer responsible for DC-to-AC conversion, the stable DC power is converted into an AC current signal, improving the voltage level adaptability and waveform smoothness of the conversion process. By connecting harmonic suppression devices (such as passive filters or active power filters (APF)), harmonic components (such as higher harmonics and distorted waveforms) contained in the current signal are filtered out, reducing harmonic interference to the power grid. At the same time, phase-locked loop (PLL) technology is used to synchronize the grid frequency and phase in real time, accurately calibrating the frequency and phase parameters of the output AC current to ensure that the output current is completely in phase with the grid. The current waveform distortion rate is required to be ≤3% and the phase synchronization error is required to be ≤1°, providing compliant current guarantee for subsequent high-quality grid connection.

[0026] Step S500: Obtain total harmonic distortion (THD) data from the current signal after harmonic suppression. If the THD data is below the threshold, it is incorporated into cross-regional scheduling optimization to obtain spatial resource allocation to improve grid connection conversion efficiency. The THD data measures current quality, and cross-regional scheduling optimization involves resource allocation algorithms.

[0027] The AC current signal after harmonic suppression in step S400 is collected, and the Total Harmonic Distortion (THD) data is calculated. THD is a core indicator for measuring current quality, reflecting the degree of interference of harmonic components on the fundamental current. A preset THD threshold conforming to grid connection standards is used to determine if the calculated result is below this threshold. If it meets the standard, the qualified power output data for that region is integrated into the cross-regional dispatch optimization system. Through resource allocation algorithms (such as distributed supply and demand matching algorithms and global optimization dispatch models), combined with the distribution of wind and solar resources, load demand differences, and grid carrying capacity in different regions, the energy output allocation and transmission paths of each region are planned in a coordinated manner to form an optimal spatial resource allocation scheme. Ultimately, this improves the conversion efficiency and resource utilization of the entire wind and solar grid-connected system, requiring a THD compliance rate ≥99% and a cross-regional dispatch response delay ≤200ms.

[0028] Furthermore, in the spatiotemporally complementary wind-solar grid-connected inverter control method proposed in this embodiment, step S100 includes: Step S110: Obtain wind speed data and light intensity data from the target area through a data acquisition device, and use a pre-established sensor network to collect wind speed data and light intensity data at regular intervals to obtain a preliminary environmental monitoring dataset.

[0029] The preliminary environmental monitoring dataset was obtained through the following: (1) In formula (1), Indicates time The obtained environmental monitoring dataset, This indicates the total number of data collection time intervals. Indicates at time Collected wind speed data, Indicates at time The collected light intensity data, This represents the weighting coefficient of the wind speed data. The weighting coefficients represent the light intensity data.

[0030] Within the target area of ​​a wind farm, real-time data is acquired by deploying data acquisition devices such as anemometers and light sensors. These devices are typically installed on the top of wind turbine towers or near solar panels to capture changes in wind speed and fluctuations in light intensity. A pre-established sensor network can consist of multiple wireless sensor nodes interconnected via IoT protocols such as ZigBee or LoRa (Long Range Radio, a low-power long-range wireless communication technology) to ensure stable and low-power data transmission. This sensor network is configured to collect wind speed data, for example, in meters per second, and record light intensity in lux. After initial filtering to remove noise, the collected data forms a preliminary environmental monitoring dataset. This dataset includes timestamps, location coordinates, and corresponding wind speed and light intensity values, providing a foundation for subsequent analysis.

[0031] Step S120: Based on the preliminary environmental monitoring dataset, integrate the historical meteorological database to construct a spatiotemporal prediction model.

[0032] The formula for calculating the output of the spatiotemporal prediction model is defined as follows: (2) In formula (2), Indicates the future The prediction result vector at time step, and This represents the output gating parameter matrix. This represents element-wise multiplication. and The weight matrices represent the spatial and temporal features, respectively. Represents the spatial feature input vector. Represents the time feature input vector. This represents the bias vector. This represents the sigmoid activation function. This represents the hidden state weight matrix. This indicates the hidden state at the previous moment. Represents the context weight matrix. This represents the data fusion vector at the current moment. This represents the context bias vector.

[0033] When integrating historical meteorological databases, preliminary environmental monitoring datasets are combined with long-term data provided by the National Meteorological Administration. This historical data covers wind speed and sunshine records over the past 10 years, including seasonal variations and extreme weather events. The process of building a spatiotemporal prediction model involves integrating this historical data into a multidimensional dataset, where the spatiotemporal dimension considers geographical latitude and longitude and the continuity of the time series. The spatiotemporal prediction model employs time series analysis methods such as Long Short-Term Memory (LSTM) networks. First, the historical data is preprocessed, such as normalization and missing value imputation. Then, the model is trained to identify patterns, such as predicting the average and fluctuation range of wind speed over specific time periods, thereby predicting the fluctuating characteristics of renewable energy sources such as wind and solar power. This helps optimize power generation scheduling. For example, in practical applications, for a wind-solar hybrid power generation system located in a coastal area, wind speed data collected by the sensor network may show lower wind speeds and higher sunshine intensity in the morning. After integrating the historical database, the model can analyze the specific pattern that peak summer wind speeds occur in the afternoon.

[0034] Time series analysis methods here determine power output patterns by decomposing the series into trend, seasonal, and residual components. For example, an ARIMA (Autoregressive Integrated Moving Average) model is used to fit the data. The process includes stationary testing, parameter estimation, and model validation, ultimately predicting power output peaks caused by energy fluctuations such as sudden increases in wind speed. This can improve energy efficiency and prevent overload or underload in business operations. For instance, extending to urban rooftop solar projects, the initial dataset collection emphasizes the diurnal variation of solar irradiance, while the spatiotemporal model integrates satellite meteorological data to predict fluctuations caused by cloud movement. This prediction helps operators adjust energy storage systems in advance, enhancing system robustness.

[0035] Furthermore, in the spatiotemporally complementary wind-solar grid-connected inverter control method proposed in this embodiment, step S200 includes: Step S210: Based on the power output pattern determined by the spatiotemporal prediction model, obtain complementary characteristic data from the target renewable energy resources, classify them according to different power generation modes, and obtain the classified resource characteristic set.

[0036] The predicted output value in the output law determined by the spatiotemporal prediction model is obtained by the following formula: (3) In formula (3), Indicates time The predicted output value, This indicates the number of components in the spatiotemporal prediction model. Indicates the first The weight coefficients of each model component, Indicates the first A prediction function, Indicates the first The parameter set of a model Indicates a reference time point. This represents the time decay factor.

[0037] Complementary property data are obtained using the following formula. (4) In formula (4), Representing resources and resources The complementary characteristic coefficients between them and Representing resources and output sequence, Represents the covariance function. Represents the variance function. and Representing resources and The peak occurrence time, This indicates the maximum time difference.

[0038] The categorized set of resource characteristics is derived using the following formula: (5) In formula (5), Indicates the first The set of resource characteristics corresponding to each power generation mode Indicates the first A characteristic vector of a renewable energy resource. Indicates the first The central feature vector of the power generation mode Represents the distance metric function. Indicates the first The classification threshold for this pattern This indicates the classification threshold for other patterns.

[0039] Starting with the power output patterns determined by the spatiotemporal prediction model, the first step is to extract complementary characteristic data from target renewable energy resources such as wind and solar power. Here, power output patterns refer to the energy output patterns derived by the model through time series analysis, such as the alternation between peak wind speed periods and off-peak solar radiation periods. Complementary characteristic data includes the intermittency of wind power and the diurnal periodicity of solar power, acquired through a data acquisition system from historical and real-time data. For example, in a wind-solar hybrid power plant, wind power data shows stable output at night, while solar power data highlights daytime peaks. This extracted data reflects the complementarity of the two resources.

[0040] Different power generation modes are categorized into single-wind mode, single-solar mode, and hybrid complementary mode. The classification process involves grouping the data using clustering algorithms such as K-means. First, features are extracted from the complementary characteristic data, such as calculating the average output value and fluctuation coefficient. Then, the data is assigned to the corresponding categories based on thresholds, ultimately resulting in a categorized set of resource characteristics. This categorized resource characteristic set is a structured dataset containing resource attributes for each category, such as peak output and duration, thus providing a foundation for subsequent optimization. For example, in a wind and solar power project located in a desert region, the output pattern shows that wind power output increases in the evening while solar power output reaches its peak at noon. When acquiring complementary characteristic data, wind speed and solar radiation sequences from the past week are retrieved from the sensor network. Complementary characteristic data is recorded hourly; for example, an average wind speed of 5 m / s corresponds to a solar output of 300 watts per square meter. Thus, the complementary characteristic data is classified as a complementary set within the hybrid mode.

[0041] Step S220: Based on the resource characteristic set, use a power optimization tool to dynamically adjust the output weight. If the dynamically adjusted output weight exceeds the preset threshold, reduce the corresponding ratio to obtain the weight allocation scheme.

[0042] The dynamically adjusted output weight is obtained using the following formula: (6) In formula (6), Indicates the first The output weights of each resource are dynamically adjusted. Indicates the first The initial weights of each resource Indicates based on resource characteristics The power optimization value, Indicates the reference power value. Indicates the first The efficiency coefficient of a resource.

[0043] The final weight allocation value in the weight allocation scheme is obtained by the following formula: (7) In formula (7), Indicates the first The final weight allocation value for each resource. This represents the adjusted weight value. This indicates the preset upper limit of the weight threshold. Indicates the first The reduction ratio of each resource.

[0044] Based on the resource characteristic set, a power optimization tool dynamically adjusts the output weights. This power optimization tool is a software module based on linear programming. It calculates the optimal weight allocation by inputting resource set data. For example, the tool first loads the output data from the set, then uses an objective function to minimize the fluctuation variance. The dynamic adjustment process involves iterative calculations. For instance, the initial weights are 0.6 for wind power and 0.4 for solar power. If the simulated output exceeds the system capacity threshold, such as the total power limit of 1000 kilowatts, the corresponding proportions are reduced, for example, the wind power weight is reduced to 0.5. The final weight allocation scheme is a weight vector that ensures overall output balance. In coastal wind farm operations, the resource characteristic set shows large wind energy fluctuations. The power optimization tool monitors this in real time. If a wind power weight of 0.7 causes the total output to exceed the threshold by 1.2 times, it is automatically reduced to 0.55, forming a new allocation scheme.

[0045] Step S230: Based on the weight allocation scheme, use power balance calculation tools to monitor fluctuations in real time, obtain short-term power fluctuation data, and determine the initial buffer demand range.

[0046] Short-term power fluctuation data are obtained using the following formula: (8) In formula (8), The root mean square value represents the power fluctuation. Indicates the length of the monitoring time window. Indicates instantaneous power value. This represents the average power value within the time window.

[0047] The initial buffer demand range is determined using the following formula: (9) In formula (9), Indicates the required buffer capacity. Indicates the safety factor. The standard deviation of power fluctuations This represents the system response time constant.

[0048] According to the weighting scheme, fluctuations are monitored in real time using a power balance calculation tool. This tool is an embedded algorithm engine that calculates short-term power fluctuations by inputting the weighting scheme and real-time data. For example, the tool samples the output value every minute, calculates the fluctuation as the difference between the current value and the previous value, and then determines an initial buffer demand range. For instance, fluctuations within ±50 kW are considered low demand. This data helps identify potential instability. For example, in an urban solar roof system, after the scheme assigns a solar weight of 0.8, the tool monitors short-term fluctuations of 20 kW per 5 minutes. The initial buffer demand range is set at 50-100 kW to cover peak values.

[0049] Step S240: For short-term power fluctuation data within the initial buffer demand range, a data integration platform is used for secondary calibration to determine whether the balance requirements are met, and the short-term power buffer demand value is obtained.

[0050] The short-time power buffer requirement is derived using the following formula: (10) In formula (10), This indicates the short-term power buffer requirement value. Indicates the sampling time window. Indicates time Power fluctuation data, This represents the baseline power value. The formula calculates the buffer requirement by integrating the power fluctuation deviation over time.

[0051] The following formula is used to determine whether the balance requirement is met: (11) In formula (11), Indicators representing balance judgments This represents the total number of data points within the evaluation period. Indicates the first Energy deviation value at each moment This represents the mean of the energy deviation. This represents the threshold for balance requirements; the balance requirement is met when the balance judgment index is less than or equal to the threshold.

[0052] For short-term power fluctuation data within the initial buffering demand range, a secondary calibration is performed using a data integration platform. This platform is a cloud-based system that integrates multi-source data, such as weather forecasts and historical records, for calibration. For example, the platform first imports the fluctuation data, then applies filtering algorithms such as Kalman filtering to correct noise, and determines whether the balance requirement is met, i.e., whether the fluctuation is within the threshold after calibration. If so, it outputs the short-term power buffering demand, such as requiring an additional 50 kWh of energy storage. For instance, in a mountainous wind and solar power project, after the data integration platform performs secondary calibration on the fluctuation data within the range and confirms that the requirement is met, the buffering demand is determined to be 80 kWh to maintain stability.

[0053] Furthermore, in the spatiotemporally complementary wind-solar grid-connected inverter control method proposed in this embodiment, step S300 includes: Step S310: Determine whether the short-term power buffer demand exceeds the threshold. If the short-term power buffer demand exceeds the preset threshold range, trigger the call command and obtain the energy scheduling signal.

[0054] The following formula is used to define the triggering conditions for a call instruction: (12) In formula (12), This indicates that the command triggers a signal. This indicates the short-term power buffer requirement value. This indicates a preset threshold range. The call command is triggered when the absolute value of the short-term power buffer demand exceeds the threshold; otherwise, it is not triggered.

[0055] The energy dispatch signal of appropriate strength is generated based on the degree to which the threshold is exceeded using the following formula: (13) In formula (13), Indicates the strength of the energy dispatch signal. Represents the scheduling gain coefficient. This represents the power difference exceeding the threshold. Indicates the scheduling response index. Represents a symbolic function.

[0056] First, it's necessary to assess the short-term power buffer demand. This short-term power buffer demand refers to the estimated temporary energy storage capacity in a renewable energy system to cope with power fluctuations. For example, it can be calculated by real-time monitoring of wind and solar power output changes. If fluctuations cause the demand to reach 150 kWh, while the preset threshold is within 100 kWh, it's considered to exceed the threshold. This assessment involves using an embedded monitoring module integrated into the energy management system. This module collects power data every minute, first extracting a baseline value from historical buffer records, and then comparing the current demand with the threshold. For example, in a wind-solar hybrid power plant, if a sudden drop in wind speed leads to insufficient solar power output, the calculated buffer demand might be 180 kWh, exceeding the threshold by 80 kWh. In this case, the system automatically triggers a call command, a sequence of digital signals used to activate the downstream scheduling process, thereby generating an energy dispatch signal. This energy dispatch signal contains specific energy release instructions, such as specifying how much capacity to extract from the battery pack. For example, in a wind power project located in a plain area, the embedded monitoring module detects large fluctuations in wind power output at night. The buffer demand is calculated to be 220 kWh, which exceeds the threshold of 150 kWh. Therefore, an instruction is triggered, and the signal carries a priority code to ensure a fast response.

[0057] Once an energy dispatch signal is received, the system will draw backup energy from the energy storage module, typically a lithium-ion battery array or supercapacitor system, designed to store excess renewable energy. The acquisition process includes a signal parsing step. First, the signal is transmitted to the module controller. The controller extracts energy from the available capacity of the energy storage module based on parameters in the signal, such as demand and duration. For example, if the signal indicates a need for 100 kWh of energy, the controller will check the module's current storage status. If the total capacity is 500 kWh and the availability rate is 80%, it will prioritize extraction from high-efficiency cells, avoiding inefficient portions. For instance, in a coastal solar farm, if the dispatch signal specifies the acquisition of 150 kWh of backup energy, the energy storage module will use internal switching circuitry to evenly extract energy from multiple battery cells, ensuring that overall stability is not affected.

[0058] Step S320: Obtain backup energy from the energy storage module according to the energy dispatch signal, use the allocation tool to perform stratified processing of the backup energy, and determine the stratified allocation scheme.

[0059] The backup energy obtained from the energy storage module is calculated using the following formula: (14) In formula (14), This represents the total reserve energy obtained from the energy storage module. Indicates the number of energy storage modules. Indicates the first The strength of the energy scheduling signal received by each module Indicates the first The available capacity of each energy storage module, This represents the energy extraction efficiency coefficient.

[0060] The allocation scheme after stratification is determined using the following formula: (15) In formula (15), This represents the optimal allocation scheme. Indicates the total number of allocation units. Indicates the first The utility function value of each unit. Indicates the first The proportion of resources allocated to each unit Indicates the first Demand coefficient for each unit A set of possible allocation schemes.

[0061] After acquiring reserve energy, a distribution tool is used to stratify it. This tool is a software framework based on a stratification algorithm, dividing the energy into different tiers, such as a rapid response tier and a continuous supply tier, to optimize distribution efficiency. The stratification process first classifies the total energy amount. For example, 200 kWh of energy might be divided into a rapid response tier of 100 kWh for immediate fluctuation compensation and a buffer tier of 100 kWh for medium- to long-term balancing. Then, a scheme is determined based on the system load, outputting a stratified distribution scheme. This scheme is a vector list listing the energy proportions and release order of each tier. For instance, in a mountainous hybrid energy project, the tool stratifies the acquired 180 kWh of energy, with the rapid response tier accounting for 60% to cope with sudden wind changes. The final scheme is a three-tier structure to ensure coverage across different time scales.

[0062] Step S330: For the stratified allocation scheme, use control tools to dynamically adjust the energy output and obtain the output data after regulation.

[0063] The dynamic output adjustment mechanism based on control theory is described by the following formula: (16) In formula (16), This indicates the energy output after regulation. Indicates the basic allocation of output. This represents the proportional control coefficient. Represents the differential control coefficient. Indicates an energy deviation signal. This indicates the rate of change in energy deviation.

[0064] For the tiered energy allocation scheme, a control tool is used to dynamically adjust energy output. This control tool is a feedback control engine that adjusts output based on real-time input scheme data. The adjustment process involves iterative feedback. For example, the control tool monitors the current output; if the scheme indicates a rapid tier release of 50 kW, the control tool will adjust the speed of the wind turbines or the output of the solar inverters to obtain the adjusted output data, such as stabilizing it at a level of 800 kW. For instance, in a city rooftop energy system, after the scheme is tiered, the tool dynamically adjusts the solar output, reducing it from an initial fluctuating value to a stable curve. Data shows that the hourly output is balanced after adjustment.

[0065] Furthermore, in the spatiotemporally complementary wind-solar grid-connected inverter control method proposed in this embodiment, step S400 includes: Step S410: Based on the output data after regulation, a preliminary current signal is extracted from the inverter layer using a multi-level conversion structure. The preliminary current signal is then preliminarily processed to obtain the processed basic current data.

[0066] The initial current signal extracted from the inverter layer is obtained using the following formula: (17) In formula (17), This represents the initial current signal extracted from the inverter layer. Indicates the number of levels in a multi-level conversion structure. Indicates the first The voltage value of each level, Indicates the first Each level at time The on / off state, This represents the conversion gain coefficient of the inverter layer.

[0067] The adjusted baseline current data is obtained using the following formula: (18) In formula (18), This indicates the adjusted basic current data at the frequency. The amplitude at that point, This represents the weighting coefficient of the fundamental component. Represents the fundamental current component. The weighting coefficients representing harmonic components. Indicates the highest harmonic order considered. Indicates the first Second harmonic current components Represents the harmonic order.

[0068] Based on the regulated output data, a multilevel converter (MLC) structure is used to extract the initial current signal from the inverter layer. This MLC is an advanced power electronic topology, typically composed of multiple switching devices and capacitors, used to convert DC to AC while reducing harmonic distortion. When operating in the inverter layer, the MLC structure first receives the regulated output data. For example, in a wind-solar hybrid power generation system, the output data stabilizes at 750 kW. Then, it gradually constructs a voltage waveform through cascaded H-bridge units. Each unit is independently controlled to generate a multi-stage stepped output, thereby extracting the initial current signal. This process involves synchronous pulse width modulation (PWM) technology to ensure that the initial signal shape approximates a sine wave. In a solar power plant project located in a desert region, the regulated output data indicates a requirement for a 500 amp current output. When the MLC extracts the signal from the inverter layer, it processes the DC-side input in layers, first boosting the voltage to a medium level, and then gradually superimposing the signals to obtain an initial current signal of approximately 480 amps. This contributes to the stability of subsequent processing.

[0069] The initial current signal is processed to obtain the processed basic current data. This initial processing usually includes signal sampling and preliminary filtering steps, such as using a digital signal processor to discretize the signal and remove obvious noise peaks.

[0070] Step S420: If there are interference fluctuations in the basic current data, the basic current data is filtered by a harmonic suppression device to determine the filtered stable current data.

[0071] The filtered stable current data is obtained using the following formula: (19) In formula (19), This represents the stable current data after processing by the harmonic suppression device. This represents the original base current data. Indicates the first The amplitude of the second harmonic. Indicates the harmonic order. Indicates the fundamental angular frequency. Indicates the first Phase angle of the subharmonic, This indicates the highest harmonic order that needs to be suppressed.

[0072] The following formula is used to quantify the intensity of interference fluctuations to determine whether harmonic suppression devices need to be activated: (20) In formula (20), This represents the standard deviation of disturbance fluctuations in the base current data. Indicates the total number of sampling points. Indicates the first One original current sample value, Indicates the first The current value after smoothing.

[0073] If the base current data exhibits interference fluctuations, a harmonic suppression device filters the data to determine the stable current data. This device is a hardware component based on an active filter that dynamically injects reverse current to cancel harmonics. Specifically, the process first detects the spectrum of the base current data, analyzing the 3rd and 5th harmonic components using Fourier transform. If the fluctuation exceeds 10%, the device activates a control algorithm to calculate a compensation current and superimpose it onto the original signal, ultimately outputting stable current data. In a wind power system within an industrial park, the base current data showed a 15% interference fluctuation. The device, through real-time monitoring and sampling data 1000 times per second, identified the main harmonic frequency as 150 Hz, then generated a compensation signal to reduce the fluctuation to within 2%, resulting in a smooth waveform of approximately 450 amps for the stable current data. This processing ensures the reliability of grid connection, reduces equipment losses, and improves overall energy transmission efficiency.

[0074] Step S430: Use a frequency detection tool to compare and analyze the frequency characteristics of the stable current data to obtain adjusted current data that is consistent with the grid frequency.

[0075] The following formula is used to extract the frequency characteristics of steady-state current data for subsequent comparative analysis: (twenty one) In formula (21), This represents the frequency domain transformation result of the steady-state current data. This represents the steady-state current data in the time domain. Indicates the sampling time window. Represents frequency variables. Represents a complex exponential function; The following formula is used to determine whether the current frequency matches the power grid frequency: (twenty two) In formula (22), Indicates the standard frequency of the power grid. This indicates the frequency of the measured current. Indicates frequency deviation. This indicates the permissible frequency error threshold.

[0076] The following formula is used to generate adjusted current data consistent with the grid frequency: (twenty three) In formula (23), This indicates the adjusted current data. This represents the original steady-state current data. This represents the adjustment factor based on the power grid frequency. Indicates the power grid frequency. This indicates the phase correction angle.

[0077] A frequency detection tool is used to compare and analyze the frequency characteristics of stable current data to obtain adjusted current data that is consistent with the grid frequency. This frequency detection tool is a software algorithm framework integrated into a microcontroller, used to compare the deviation of the signal frequency from the standard grid frequency, such as 50 Hz.

[0078] Step S440: Use a phase calibration tool to perform phase comparison and correction on the adjusted current data, determine whether it has reached the state of being in phase with the power grid, and obtain the final output current signal.

[0079] The following formula is used to generate the final current signal that is in phase with the power grid: (twenty four) In formula (24), This indicates the corrected output current. Indicates the compensation current component. Indicates the angular frequency of the power grid. Indicates the reference phase of the power grid. Represents a time variable.

[0080] A phase calibration tool is used to compare and correct the phase of the adjusted current data to determine whether it has reached a state of phase synchronization with the power grid, thus obtaining the final output current signal. This phase calibration tool is a phase-locked loop (PLL) based system that can track and adjust the phase difference in real time. Specifically, the process involves extracting the phase information of the adjusted current data, such as calculating the phase angle using a zero-crossing detection method. If the phase deviation from the power grid exceeds 5 degrees, the tool applies proportional-integral control to gradually correct the deviation until it is less than 1 degree, ultimately confirming the synchronization state and outputting the signal. In a city-wide distributed energy network, the initial phase of the adjusted current data lagged by 8 degrees. The phase calibration tool, through iterative comparison and adjustment every cycle, quickly corrected to synchronization, resulting in a final output current signal stable at 400 amps. This achieves seamless grid connection, avoids power factor degradation, and thus improves the overall system performance and energy utilization.

[0081] Furthermore, in the spatiotemporally complementary wind-solar grid-connected inverter control method proposed in this embodiment, step S500 includes: Step S510: Extract total harmonic distortion rate data from the current signal after harmonic suppression using signal analysis tools to obtain a standardized distortion rate index.

[0082] The following formula is used to calculate the degree of distortion of all harmonic components in a current signal relative to the fundamental frequency: (25) In formula (25), The distortion rate metric represents the normalized distortion rate. Indicates the first Effective value of subharmonic current This represents the effective value of the fundamental current. This indicates the highest harmonic order considered.

[0083] Total Harmonic Distortion (THD) data is extracted from the harmonic-suppressed current signal using a signal analysis tool to obtain a standardized distortion rate index. This signal analysis tool is a software module based on digital signal processing (DSP) that performs spectral decomposition on the current waveform to calculate the ratio of harmonic components to the fundamental frequency. Specifically, the tool first acquires the harmonic-suppressed current signal. For example, in a grid-connected photovoltaic power station system, the current signal is a filtered AC current waveform. Then, a Fast Fourier Transform (FFT) method is applied to convert the time-domain signal to the frequency domain, identifying the amplitude of each harmonic. For example, if the fundamental frequency is 50 Hz, the content of the 3rd, 5th, and other harmonics is extracted. The THD is calculated as the square root of the sum of the squares of these harmonic amplitudes divided by the fundamental frequency amplitude, yielding a percentage value, such as 5.2%. This value is then normalized to a value between 0 and 1 for easier comparison later. In an application at a coastal wind farm, the current signal after harmonic suppression showed a fundamental amplitude of 600 amperes. The signal analysis tool extracted a total harmonic distortion rate of 4.8%, which, after normalization, was 0.048. This helps in assessing signal quality.

[0084] Step S520: If the standardized distortion rate index is lower than the preset threshold, then the data comparison tool is used to perform condition verification to determine the available distortion rate reference value.

[0085] The following formula is used to determine whether the verification passes by comparing data: (26) In formula (26), This indicates the conditional validation results of the data comparison tool. This indicates the total number of data points compared. Indicates the first One reference data value, Indicates the first One data value to be verified. This indicates the error tolerance that passed the verification.

[0086] The following formula is used to select the optimal distortion rate reference value from the candidate values ​​that meet the conditions: (27) In formula (27), This indicates the available distortion rate reference value. Indicates the first One candidate distortion rate value, Indicates the target distortion rate. Indicates the first Quality score of each candidate value This indicates the minimum quality requirement.

[0087] If the normalized distortion rate is lower than a preset threshold, a data comparison tool is used for condition verification to determine a usable distortion rate reference value. This data comparison tool is a comparison algorithm framework integrated into the control system to verify whether the indicator meets specific conditions. Specifically, the preset threshold might be set to 0.05. If the indicator is 0.03, which is lower than the threshold, the data comparison tool will match it with reference values ​​in a historical database, such as comparing the average distortion rate over the past week. The verification process includes calculating the deviation; if the deviation is less than 0.01, the indicator is confirmed as usable as a reference value. In the scenario of a mountainous hydropower project, the normalized indicator is 0.025, which is lower than the threshold of 0.05. After comparison, the data comparison tool determines the reference value to be 0.022, ensuring the reliability of the data.

[0088] Step S530: Use a data fusion tool to integrate the available distortion rate reference value with cross-regional scheduling information to obtain the fused scheduling input data.

[0089] The merged scheduling input data is obtained through the following formula: (28) In formula (28), This represents the merged scheduling input data. This indicates the number of available distortion rate reference values. Indicates the first The weighting coefficients of each distortion rate reference value Indicates the first A distortion rate reference value, This indicates the number of cross-regional scheduling messages. Indicates the first Weighting coefficients for cross-regional scheduling information Indicates the first Cross-regional scheduling information data.

[0090] A data fusion tool is used to integrate available distortion rate reference values ​​with cross-regional scheduling information to obtain fused scheduling input data. This data fusion tool is a multi-source data integration algorithm that can merge data from different dimensions into a unified format. Specifically, it first performs weighted fusion of the distortion rate reference value (e.g., 0.022) with cross-regional scheduling information, such as load forecast data and real-time power demand from neighboring provincial power grids. A comprehensive input is then generated using averaging or Kalman filtering methods. For example, the fused data includes a vector of distortion rate and scheduling demand, facilitating subsequent optimization.

[0091] Step S540: The fused scheduling input data is processed by a resource allocation algorithm tool to perform spatial resource allocation processing, resulting in an optimized regional resource distribution scheme.

[0092] The optimized regional resource distribution scheme is derived using the following formula: (29) In formula (29), Representing region coordinates The optimal resource allocation scheme at the location. Indicates the total number of resource types. Indicates the first Weight coefficients of resource classes Indicates the region For the The demand for such resources Indicates the area The Class resource capacity limit, Indicates the first The efficiency factor for the utilization of various resources.

[0093] The resource allocation algorithm tool processes the fused scheduling input data to perform spatial resource allocation, resulting in an optimized regional resource distribution scheme. This algorithm is a linear programming-based optimization engine that considers spatial factors such as geographical location and resource availability. Specifically, after inputting the fused data, the resource allocation algorithm defines the objective function as minimizing resource waste, with constraints including a distortion rate not exceeding a threshold. The algorithm then iteratively solves the problem. For example, in a multi-regional power grid, power can be allocated to different substations, resulting in an output scheme such as allocating 40% of resources to the eastern region and 60% to the western region, thus optimizing the distribution.

[0094] Step S550: For the optimized regional resource distribution scheme, continuously monitor the changes in the distortion rate index to ensure the matching degree with cross-regional scheduling information.

[0095] The following formula is used to evaluate the consistency of cross-regional scheduling information: (30) In formula (30), Indicates the first The degree of cross-regional scheduling information matching in each region Indicates the total number of dimensions of scheduling information. Indicates the first Weight coefficients for each dimension Indicates an indicator function, Indicates the area In the The actual scheduling status in each dimension Indicates the area In the The expected scheduling status in each dimension.

[0096] The following formula is used to continuously monitor the dynamic changes in the distortion rate metric: (31) In formula (31), Indicates the first Distortion rate change indicators for each monitoring period This represents the static deviation weighting coefficient. Indicates dynamically changing weighting coefficients. This represents the distortion rate value for the current period. This represents the baseline distortion rate value. This indicates the rate of change of the distortion rate.

[0097] For the optimized regional resource distribution scheme, changes in the distortion rate index are continuously monitored to ensure its consistency with cross-regional dispatch information. This monitoring process uses a real-time tracking system, sampling the distortion rate every minute. For example, if the index rises from 0.022 to 0.035 after the scheme is implemented, the system checks its consistency with the dispatch information. If the deviation exceeds 5%, an alarm is triggered to adjust the scheme. In a case study of a city smart grid, monitoring showed that the index remained stable at 0.028, achieving a 98% consistency with the dispatch information and maintaining system balance.

[0098] Please see Figure 2This invention provides a spatiotemporally complementary wind-solar grid-connected inverter control system for executing the aforementioned spatiotemporally complementary wind-solar grid-connected inverter control method. It includes a spatiotemporal prediction model construction module 10, a short-term power buffer demand acquisition module 20, an output data acquisition module 30, a current signal acquisition module 40, and a spatial resource allocation acquisition module 50. The spatiotemporal prediction model construction module 10 acquires real-time information from regional wind speed and solar irradiance data using a data acquisition device, and constructs a spatiotemporal prediction model by combining it with a historical meteorological database. This model determines the output pattern based on time series analysis to predict the fluctuation characteristics of renewable energy. The short-term power buffer demand acquisition module 20, based on the output pattern determined by the spatiotemporal prediction model, acquires complementary characteristic data from renewable energy resources, dynamically allocates output weights using a power optimization algorithm, and obtains the short-term power buffer demand. The complementary characteristic data reflects the energy complementary power generation model. The formula is as follows: Short-term power buffer demand represents the buffer capacity required for power fluctuations; Output data acquisition module 30 is used to determine whether the short-term power buffer demand exceeds the threshold. If it does, it obtains backup energy from the energy storage module and regulates the energy output through a decoupling control strategy to obtain the regulated output data. The energy storage module includes a battery system; Current signal acquisition module 40 is used to obtain the current signal from the inverter layer based on the regulated output data using a multi-level conversion structure. Harmonic suppression devices are used to suppress harmonics and determine that the output current is in phase and frequency with the grid. The inverter layer is responsible for DC-AC conversion; Spatial resource allocation acquisition module 50 is used to obtain total harmonic distortion (THD) data from the harmonic-suppressed current signal. If the THD data is lower than the threshold, it is incorporated into cross-regional scheduling optimization to obtain spatial resource allocation to improve grid connection efficiency. The THD data measures current quality, and cross-regional scheduling optimization involves resource allocation algorithms.

[0099] This embodiment discloses a spatiotemporally complementary wind-solar grid-connected inverter control method and system. Compared with existing technologies, it predicts power output patterns by constructing a spatiotemporal prediction model, dynamically allocates power output weights by combining complementary characteristic data and power optimization algorithms, accurately calculates short-term power buffer demand, and regulates power output through energy storage modules and decoupling control strategies when demand exceeds a threshold, ensuring power stability. Simultaneously, it optimizes the current signal using a multi-level conversion structure and harmonic suppression device to ensure that the output current is in phase and frequency with the grid. Furthermore, it improves resource allocation efficiency through cross-regional scheduling optimization, ultimately achieving a significant improvement in grid-connected conversion efficiency and stable current quality. This embodiment effectively solves the volatility and stability problems in the renewable energy grid connection process, providing reliable technical support for large-scale renewable energy grid integration.

[0100] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A spatiotemporal complementary wind-solar grid-connected inverter control method, characterized in that, The method comprises the following steps: S100, acquiring real-time collection information from regional wind speed data and light intensity data through a data acquisition device, and combining a historical meteorological database to construct a space-time prediction model, wherein the space-time prediction model determines an output law based on a time series analysis method to predict renewable energy fluctuation characteristics; S200, acquiring complementary characteristic data from renewable energy resources according to the output law determined by the space-time prediction model, dynamically allocating an output weight using a power optimization algorithm, and obtaining short-time power buffer demand, wherein the complementary characteristic data reflects an energy complementary power generation mode, and the short-time power buffer demand represents a buffer capacity required for power fluctuation; S300, determining whether the short-time power buffer demand exceeds a threshold value, and if so, acquiring standby energy from an energy storage module, regulating energy output through a decoupling control strategy, and acquiring regulated output data, wherein the energy storage module comprises a battery system; S400, for the regulated output data, acquiring a current signal from an inverter layer using a multi-level conversion structure, suppressing harmonics through a harmonic suppression device, and determining an output current that is the same frequency and in phase with a power grid, wherein the inverter layer is responsible for direct current to alternating current conversion; S500, acquiring total harmonic distortion rate data from the current signal after harmonic suppression, and if the total harmonic distortion rate data is below a threshold value, then integrating cross-regional dispatch optimization to obtain spatial resource allocation to improve grid conversion efficiency, wherein the total harmonic distortion rate data measures current quality, and the cross-regional dispatch optimization involves a resource allocation algorithm.

2. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 1, characterized in that, Step S100 comprises: S110, acquiring wind speed data and light intensity data from a target area using a pre-established sensor network to collect the wind speed data and light intensity data at regular intervals, and obtaining a preliminary environmental monitoring data set; S120, constructing a space-time prediction model by fusing a historical meteorological database according to the preliminary environmental monitoring data set.

3. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 1, characterized in that, Step S200 comprises: S210, acquiring complementary characteristic data from a target renewable energy resource according to the output law determined by the space-time prediction model, classifying different power generation modes, and obtaining a classified resource characteristic set; S220, dynamically adjusting an output weight using a power optimization tool according to the resource characteristic set, and if the dynamically adjusted output weight exceeds a preset threshold value, reducing the corresponding proportion to obtain a weight allocation scheme; S230, using a power balance calculation tool to monitor fluctuations in real time according to the weight allocation scheme, acquiring short-time power fluctuation data, and determining a preliminary buffer demand range; S240, using a data integration platform to perform secondary calibration on the short-time power fluctuation data within the preliminary buffer demand range, determining whether the balance requirement is met, and obtaining a short-time power buffer demand value.

4. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 1, characterized in that, Step S300 comprises: S310, determining whether the short-time power buffer demand exceeds a threshold value, and if the short-time power buffer demand exceeds a preset threshold value range, triggering a call instruction to obtain an energy dispatch signal; S320, obtaining standby energy from the energy storage module according to the energy scheduling signal, and performing hierarchical processing on the standby energy by using a distribution tool to determine a hierarchical distribution scheme; S330, dynamically adjusting the energy output by using a control tool for the hierarchical distribution scheme to obtain regulated output data.

5. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 1, characterized in that, Step S400 includes: S410, extracting a preliminary current signal from the inverter layer by using a multi-level conversion structure according to the regulated output data, and performing preliminary arrangement on the preliminary current signal to obtain arranged basic current data; S420, if the basic current data has interference fluctuations, performing interference filtering processing on the basic current data by using a harmonic suppression device to determine filtered stable current data; S430, comparing and analyzing the frequency characteristics of the stable current data by using a frequency detection tool to obtain adjusted current data consistent with the grid frequency; S440, performing phase comparison and correction processing on the adjusted current data by using a phase calibration tool to determine whether it reaches a state consistent with the grid phase to obtain the final output current signal.

6. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 1, characterized in that, Step S500 includes: S510, extracting total harmonic distortion rate data from the harmonic-suppressed current signal by using a signal analysis tool to obtain a normalized distortion rate index; The following formula is used to calculate the distortion degree of all harmonic components in the current signal relative to the fundamental wave: ； wherein denotes the normalized distortion rate indicator, denotes the effective value of the second harmonic current, denotes the effective value of the fundamental current, denotes the highest harmonic order considered; S520, if the normalized distortion rate index is lower than the preset threshold, performing conditional verification by using a data comparison tool to determine a usable distortion rate reference value; S530, integrating the usable distortion rate reference value with the cross-regional scheduling information by using a data fusion tool to obtain fused scheduling input data; S540, performing spatial resource configuration processing on the fused scheduling input data by using a resource allocation algorithm tool to obtain an optimized regional resource distribution scheme; S550, continuously monitoring the change of the distortion rate index for the optimized regional resource distribution scheme to ensure the matching degree with the cross-regional scheduling information.

7. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 6, characterized in that, In step S520, the following formula is used to determine whether the verification passes by data comparison: ； wherein, represents a condition check result of a data comparison tool, represents a total number of comparison data points, represents a first reference data value, represents a first to-be-checked data value, represents an error tolerance for passing the check; The following formula is used to select the optimal distortion rate reference value from the candidate values that meet the conditions: ； in, This indicates the available distortion rate reference value. Indicates the first One candidate distortion rate value, Indicates the target distortion rate. Indicates the first Quality score of each candidate value This indicates the minimum quality requirement.

8. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 7, characterized in that, In step S530, the fused scheduling input data is obtained by the following formula: ； wherein, represents the fused scheduling input data, represents the number of available distortion rate reference values, represents a weight coefficient of the th distortion rate reference value, represents the th distortion rate reference value, represents the number of cross-region scheduling information, represents a weight coefficient of the th cross-region scheduling information, represents the th cross-region scheduling information data; In step S540, the optimized regional resource distribution scheme is obtained by the following formula: ； wherein, represents the optimal resource configuration scheme at the region coordinate , represents the total number of resource types, represents the weight coefficient of the th resource type, represents the demand amount of the th resource type in the region , represents the capacity limit of the th resource type in the region , represents the utilization efficiency factor of the th resource type.

9. The spatiotemporal complementary wind-solar grid-connected inverter control method according to claim 8, characterized in that, In step S550, the following formula is used to evaluate the consistency degree of the cross-regional scheduling information: ； in, Indicates the first The degree of cross-regional scheduling information matching in each region Indicates the total number of dimensions of scheduling information. Indicates the first Weight coefficients for each dimension Indicates an indicator function, Indicates the area In the The actual scheduling status in each dimension Indicates the area In the The expected scheduling status in each dimension; The following formula is used to continuously monitor the dynamic change characteristics of the distortion rate index: ； wherein, denotes a distortion rate change indicator of the monitoring period, denotes a static deviation weight coefficient, denotes a dynamic change weight coefficient, denotes a distortion rate value of the current period, denotes a reference distortion rate value, denotes a change rate of the distortion rate.

10. A time-space complementary wind-solar grid-connected inverter control system for executing the time-space complementary wind-solar grid-connected inverter control method according to any one of claims 1 to 9, comprising: a time-space prediction model construction module (10) for obtaining real-time collection information from regional wind speed data and illumination intensity data by using a data collection device, and constructing a time-space prediction model in combination with a historical weather database, wherein the time-space prediction model determines the output law based on a time series analysis method to predict the fluctuation characteristics of renewable energy; The short-time power buffer demand acquisition module (20) is configured to acquire complementary characteristic data from the renewable energy resource according to the output law determined by the space-time prediction model, dynamically allocate output weight by using a power optimization algorithm, and obtain short-time power buffer demand, wherein the complementary characteristic data reflects an energy complementary power generation mode, and the short-time power buffer demand represents a buffer capacity required by power fluctuation. The output data acquisition module (30) is configured to determine whether the short-time power buffer demand exceeds a threshold value, acquire standby energy from an energy storage module if the short-time power buffer demand exceeds the threshold value, control energy output by using a decoupling control strategy, and acquire regulated output data, wherein the energy storage module includes a battery system. The current signal acquisition module (40) is configured to acquire a current signal from an inverter layer by using a multi-level conversion structure for the regulated output data, suppress harmonics by using a harmonic suppression device, and determine an output current and a power grid to be the same frequency and in phase, wherein the inverter layer is responsible for direct current to alternating current conversion. The spatial resource configuration acquisition module (50) is configured to acquire total harmonic distortion rate data from the current signal after the harmonics are suppressed, integrate cross-regional scheduling optimization if the total harmonic distortion rate data is lower than a threshold value, and obtain spatial resource configuration to improve grid conversion efficiency, wherein the total harmonic distortion rate data measures current quality, and the cross-regional scheduling optimization involves a resource allocation algorithm.

Images