A method, device and equipment for integrated collaborative regulation of optical storage and charging based on load characteristic response, and a readable storage medium

Abstract

The present application relates to the field of electric power technology, in particular to a kind of light storage fills integrated collaborative regulation method, device and equipment based on net load characteristic response and readable storage medium, the method is to the time sequence modeling and data preprocessing of distribution area net load in day, extract multi-dimensional characteristic index;Based on the quantitative mapping relationship of net load and substation point voltage of net load multi-dimensional characteristic index, the safe operation constraint of gateway point voltage is converted into the adjustable boundary of net load in each period by inverse mapping;With net load adjustable boundary as core constraint, construct multi-objective day-ahead scheduling model, obtain optimal day-ahead scheduling strategy;The remaining net load obtained is substituted into the quantitative mapping relationship of net load and substation point voltage established, complete gateway point voltage calculation, constraint check and out-of-limit risk assessment, the scheduling result that does not satisfy voltage safety requirement is closed-loop feedback and re-optimized.

Description

Technical Field

[0001] This invention relates to a method, apparatus, equipment, and readable storage medium for integrated photovoltaic-storage-charging control based on net load characteristic response, belonging to the field of power technology. Background Technology

[0002] In distribution substations with a high proportion of photovoltaic (PV) power, the temporal and spatial misalignment between PV power generation and substation load easily leads to reverse power flow. The substation's net load exhibits a significant duck-shaped curve characteristic, with the net load continuously decreasing or even turning negative during the midday peak PV generation period, and then experiencing a steep climb in a short period during the evening peak electricity consumption period. This results in frequent voltage exceedances and excessive voltage fluctuations at the substation's critical points. The voltage exceedance problem is particularly prominent during weather conditions with sudden changes in irradiance and during the evening peak hours when electric vehicles are charging intensively. This seriously affects the lifespan of power distribution equipment and the power quality on the user side, and has become a core bottleneck restricting the absorption of high-proportion renewable energy in distribution substations.

[0003] To address the aforementioned issues, extensive research has been conducted both domestically and internationally on voltage regulation and multi-source collaborative optimization in distribution transformer areas. This research primarily includes: First, power smoothing control based on energy storage systems, utilizing the charging and discharging capabilities of energy storage to mitigate fluctuations in photovoltaic output and load, thus alleviating voltage deviation problems; Second, demand response management based on flexible and adjustable loads, achieving peak shaving and valley filling by guiding orderly charging of electric vehicles and adjusting interruptible loads, thereby reducing the risk of peak-valley differences and voltage fluctuations in the distribution transformer area; Third, day-ahead scheduling models based on intelligent optimization algorithms, achieving coordinated scheduling of photovoltaic, energy storage, and adjustable loads through the construction of optimization models, improving the economic efficiency and stability of distribution transformer area operation. These existing technologies have, to a certain extent, improved the voltage control capabilities of distribution transformer areas, alleviated some operational problems caused by the integration of new energy sources, and provided a theoretical and technical foundation for multi-source collaborative regulation of distribution transformer areas.

[0004] In practical engineering applications of integrated photovoltaic-storage-charging distribution substations, the quantitative characterization system for net load characteristics is incomplete, failing to accurately map the impact mechanism of load fluctuations on voltage. The characterization of net load in substations is often limited to single indicators such as peak-valley difference, failing to fully explore multi-dimensional characteristics such as the coefficient of variation, maximum ramp rate, and continuous power changes in adjacent time periods. This fails to comprehensively depict the amplitude fluctuations, relative dispersion, instantaneous impact intensity, and continuous stability of net load, making it difficult to accurately reflect the impact of dynamic net load behavior on the voltage at the substation's cutoff point. Consequently, control strategies lack precise characteristic basis. Furthermore, the deep coupling between voltage safety constraints and the net load control process is lacking, resulting in a severe disconnect between dispatch optimization and voltage control. Existing control strategies often treat voltage constraints as ex-post verification conditions, failing to establish a quantitative mapping relationship between net load and cutoff point voltage. This prevents the direct conversion of voltage safety operation constraints stipulated in power grid regulations into executable and quantifiable net load control boundaries during dispatch. Consequently, voltage control is disconnected from day-ahead dispatch optimization, easily leading to situations where the theoretically optimal dispatch result results in voltage exceeding limits during actual operation, thus compromising the engineering feasibility of control strategies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method, apparatus, device, and readable storage medium for integrated photovoltaic-storage-charging control based on net load characteristic response, aiming to solve the above problems.

[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for integrated photovoltaic-storage-charging control based on net load characteristic response, comprising the following steps: Time-series modeling and data preprocessing are performed on the day-ahead net load of the distribution substation to extract multi-dimensional characteristic indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. Based on the multidimensional characteristic indicators of net load, a data-driven method is used to establish a quantitative mapping relationship between net load and the voltage at the cut-off point of the transformer area. Through inverse mapping, the safety operation constraint of the cut-off point voltage is transformed into the adjustable boundary of net load for each time period. Using the adjustable net load boundary as the core constraint, a multi-objective day-ahead scheduling model integrating multiple adjustable resources such as photovoltaic, energy storage, and adjustable load is constructed to obtain the optimal day-ahead scheduling strategy; The remaining net load is substituted into the established quantitative mapping relationship between the net load and the voltage at the control point of the transformer area to complete the voltage calculation, constraint verification and over-limit risk assessment at the control point, and the closed-loop feedback of the dispatch results that do not meet the voltage safety requirements is re-optimized.

[0007] In a preferred embodiment of this application, establishing a quantitative mapping relationship between net load and transformer substation voltage using a data-driven method includes: employing a Gaussian process regression method to convert historical net load sequences... Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

[0008] In a preferred embodiment of this application, the step of converting the threshold voltage safety operation constraint into a net load adjustable boundary for each time period through inverse mapping includes: Under voltage constraint conditions Using the inverse mapping capability of Gaussian process regression, the boundary values ​​of net load for each time period are calculated: In the formula, and They represent in The net load limit is ensured at all times, ensuring that the voltage at the cut-off point is not lower than the minimum value and does not exceed the maximum value, thus forming an adjustable boundary for the net load under the voltage constraint of the transformer area.

[0009] In a preferred embodiment of this application, the cutoff voltage calculation step includes: The net load sequence corresponding to the optimized scheduling result is input into the established net load-gate voltage mapping model. The gate voltage for each time period is calculated. Based on the Gaussian process regression model, the predicted distribution of the gate voltage is obtained. The following formula is used for calculation: In the formula, for Predicted average voltage at time-bound thresholds To predict variance.

[0010] In a preferred embodiment of this application, the voltage constraint verification includes the following steps: The voltage prediction values ​​at each time point are constrained and checked to determine whether they meet the voltage operating range. The calculation formula is as follows: ; If the predicted average voltage at any time point exceeds the allowable range, the scheduling scheme is deemed not to meet the voltage constraint requirements.

[0011] In a preferred embodiment of this application, the steps of assessing the risk of voltage exceeding limits include: Based on the prediction variance output by the Gaussian process regression model, a probabilistic assessment of the risk of voltage exceedance at the threshold point is performed. The calculation formula is as follows: In the formula, This is the cumulative distribution function of the standard normal distribution.

[0012] A photovoltaic-storage-charging integrated coordinated control device based on net load characteristic response, comprising: The feature extraction module is used to perform time-series modeling and data preprocessing of the day-ahead net load of the distribution area, and extract multi-dimensional feature indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. The mapping boundary module is used to establish a quantitative mapping relationship between net load and transformer area threshold voltage based on multi-dimensional net load characteristic indicators and using a data-driven method. Through inverse mapping, the safety operation constraints of threshold voltage are transformed into adjustable net load boundaries for each time period. The day-ahead scheduling module is used to construct a multi-objective day-ahead scheduling model that integrates multiple adjustable resources such as photovoltaics, energy storage, and adjustable load, with the net load adjustable boundary as the core constraint, and obtain the optimal day-ahead scheduling strategy. The verification module is used to substitute the obtained remaining net load into the established quantitative mapping relationship between net load and transformer area threshold voltage, complete the threshold voltage calculation, constraint verification and over-limit risk assessment, and re-optimize the scheduling results that do not meet voltage safety requirements through closed-loop feedback.

[0013] In a preferred embodiment of this application, the mapping boundary module includes: Quantitative units are used to perform Gaussian process regression on historical net load sequences. Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

[0014] Furthermore, to achieve the above objectives, the present invention also provides a photoelectric storage and charging integrated coordinated control device based on net load characteristic response. The photoelectric storage and charging integrated coordinated control device based on net load characteristic response includes a processor, a memory, and a photoelectric storage and charging integrated coordinated control program based on net load characteristic response stored in the memory and executable by the processor. When the photoelectric storage and charging integrated coordinated control program based on net load characteristic response is executed by the processor, it implements the steps of the photoelectric storage and charging integrated coordinated control method based on net load characteristic response as described above.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a photoelectric storage-charging integrated coordinated control program based on net load characteristic response, wherein when the photoelectric storage-charging integrated coordinated control program based on net load characteristic response is executed by a processor, the steps of the photoelectric storage-charging integrated coordinated control method based on net load characteristic response as described above are implemented.

[0016] The beneficial effects of this invention are as follows: By performing time-series modeling and data preprocessing on the day-ahead net load of distribution substations, multi-dimensional characteristic indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change, and continuous stability of change are extracted. Based on the multi-dimensional characteristic indicators of net load, a data-driven method is used to establish a quantitative mapping relationship between net load and the voltage at the substation threshold. Through inverse mapping, the safety operation constraint of the threshold voltage is transformed into the adjustable boundary of net load for each time period. With the adjustable boundary of net load as the core constraint, a multi-objective day-ahead scheduling model integrating multiple adjustable resources such as photovoltaic, energy storage, and adjustable load is constructed to obtain the optimal day-ahead scheduling strategy. The obtained remaining net load is substituted into the established quantitative mapping relationship between net load and the voltage at the substation threshold to complete the calculation of the threshold voltage, constraint verification, and risk assessment of exceeding limits. The scheduling results that do not meet the voltage safety requirements are re-optimized through closed-loop feedback. A full-dimensional quantitative characterization system for net load characteristics is constructed, breaking through the characterization limitations of existing technologies. By extracting multi-dimensional characteristic indicators covering net load amplitude fluctuation, relative dispersion, instantaneous rate of change, and continuous stability, a complete characterization of the dynamic behavior of the net load is achieved. This accurately reflects the impact mechanism of net load fluctuations on the threshold voltage, significantly improving the accuracy of the net load characteristics in representing voltage changes and providing reliable characteristic support for coordinated regulation. Deep coupling between voltage safety constraints and net load regulation is realized, solving the industry pain point of disconnect between existing scheduling and voltage control technologies. By establishing a quantitative mapping between net load and threshold voltage, voltage safety constraints are directly transformed into executable adjustable boundaries of net load during scheduling, changing voltage control from traditional post-event verification to pre-event constraint. This fundamentally avoids the problem of "theoretically optimal scheduling results but actual operating voltage exceeding limits," significantly improving the engineering feasibility of the regulation strategy. Coordinated optimization of multi-source resources (photovoltaics, energy storage, and charging) is achieved, balancing voltage safety and operational economy. Using the net load boundary corresponding to voltage constraints as the core constraint, a multi-objective scheduling model is constructed by integrating multiple adjustable resources such as photovoltaics, energy storage, and adjustable loads. Under the premise of ensuring voltage safety, it simultaneously optimizes net load fluctuation characteristics and system operating costs, improving the distribution area's ability to absorb high-proportion distributed photovoltaics and reducing the system's full-cycle operating costs. A complete closed-loop control mechanism is formed, significantly improving the safety and stability of the distribution area's operation. Through a closed-loop process of "optimized scheduling - voltage verification - feedback correction," the voltage response effect of the scheduling results is quantitatively verified and iteratively optimized, ensuring that the scheduling results meet voltage safety requirements throughout the entire time period. This effectively solves the core problems of voltage exceeding limits and power quality fluctuations in distribution areas under high-proportion photovoltaic access, improving the reliability and anti-disturbance capability of the integrated photovoltaic-storage-charging system. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the hardware structure of the integrated photovoltaic-storage-charging coordinated control device based on net load characteristic response involved in the present invention. Figure 2 This is a schematic diagram of the integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response involved in this invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response involved in the embodiments of the present invention is mainly applied to integrated photovoltaic-storage-charging coordinated control equipment based on net load characteristic response. This integrated photovoltaic-storage-charging coordinated control equipment based on net load characteristic response can be a PC, a portable computer, a mobile terminal, or other devices with display and processing functions.

[0021] Reference Figure 1 , Figure 1 This is a schematic diagram of the hardware structure of the integrated photovoltaic-storage-charging coordinated control device based on net load characteristic response involved in the embodiments of the present invention. In this embodiment, the integrated photovoltaic-storage-charging coordinated control device based on net load characteristic response may include a processor 1001 (e.g., CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize communication between these components; the user interface 1003 may include a display screen or an input unit such as a keyboard; the network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface); the memory 1005 may be a high-speed RAM memory or a stable non-volatile memory, such as a disk storage device, and the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0022] Those skilled in the art will understand that Figure 1 The hardware structure shown does not constitute a limitation on the integrated photovoltaic-storage-charging coordinated control device based on net load characteristic response. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0023] Continue to refer to Figure 1 , Figure 1 The memory 1005, which is a computer-readable storage medium, may include an operating system, a network communication module, and an integrated optical storage and charging control program based on net load characteristic response.

[0024] exist Figure 1In this embodiment, the network communication module is mainly used to connect to the server and communicate with the server for data; while the processor 1001 can call the integrated optical storage and charging control program based on net load characteristic response stored in the memory 1005 and execute the integrated optical storage and charging control method based on net load characteristic response provided in this embodiment of the invention.

[0025] This invention provides a method for integrated photovoltaic-storage-charging control based on net load characteristic response, referring to... Figure 2 The steps include: S10 performs time-series modeling and data preprocessing on the day-ahead net load of the distribution area, and extracts multi-dimensional characteristic indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. S20, based on the multi-dimensional characteristic index of net load, adopts a data-driven method to establish a quantitative mapping relationship between net load and the voltage at the cut-off point of the transformer area, and transforms the safety operation constraint of the cut-off point voltage into the adjustable boundary of net load for each time period through inverse mapping. S30, with the net load adjustable boundary as the core constraint, constructs a multi-objective day-ahead scheduling model that integrates multiple adjustable resources such as photovoltaic, energy storage and adjustable load, and obtains the optimal day-ahead scheduling strategy; S40 substitutes the obtained remaining net load into the established quantitative mapping relationship between net load and transformer area threshold voltage to complete the threshold voltage calculation, constraint verification and over-limit risk assessment, and re-optimizes the closed-loop feedback of dispatch results that do not meet voltage safety requirements.

[0026] In this embodiment, by using net load time-series modeling and multi-dimensional characteristic index extraction, the limitations of the single peak-valley difference index in existing technologies are overcome. The dynamic fluctuation characteristics of net load are comprehensively quantified from four core dimensions: amplitude fluctuation, relative dispersion, instantaneous rate of change, and continuous stability. This accurately captures the impact of net load dynamic behavior on the voltage at critical points, providing precise characteristic evidence and data foundation for subsequent full-process control. Based on the extracted multi-dimensional characteristics of net load, a data-driven method is used to establish a quantitative mapping relationship between net load and the voltage at critical points in the distribution area. Then, through inverse mapping, the hard constraints for voltage safety operation stipulated in the power grid regulations are directly transformed into quantifiable and executable adjustable boundaries for net load in each time period. This solves the core problem in existing technologies where voltage constraints cannot be implemented as controllable variables for scheduling, realizing the pre-emptive and operational nature of voltage safety constraints. Using the adjustable boundary of net load transformed from voltage constraints as the core constraint, a multi-objective day-ahead scheduling model integrating photovoltaic, energy storage, and adjustable load resources is constructed. Voltage safety requirements are integrated throughout the entire scheduling optimization process. Under the premise of ensuring voltage safety, the net load fluctuation characteristics and system operation economy are synergistically optimized to obtain the optimal day-ahead scheduling strategy that balances safety and efficiency. The remaining net load after scheduling optimization is substituted back into the established net load-cut-off voltage mapping model to complete the quantitative calculation of cut-off voltage, constraint compliance verification, and limit exceedance risk assessment. For scheduling results that do not meet voltage safety requirements, closed-loop feedback is fed back to the scheduling optimization stage for iterative solution, forming a complete control closed loop. This completely solves the disconnect problem of existing technologies that "emphasize scheduling optimization, neglect voltage verification, and lack feedback correction," ultimately achieving the safe, economical, and stable operation of the photovoltaic-storage-charging integrated system.

[0027] In some embodiments, establishing a quantitative mapping relationship between net load and transformer substation voltage using a data-driven method includes: employing a Gaussian process regression method to convert historical net load sequences... Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

[0028] The nonlinear mapping model is constructed using the Gaussian process regression method. It does not rely on the precise physical topology, line parameters, equipment impedance and other prior information of the distribution substation. The model can be trained using only the historical operation data of the substation. It is adaptable to distribution substations with different grid structures, different photovoltaic installation scales and different load characteristics, and has strong generalization ability and engineering applicability.

[0029] Gaussian process regression has excellent nonlinear fitting capabilities and can accurately capture the complex nonlinear coupling relationship between net load and threshold voltage. Compared with linear regression, traditional curve fitting and other methods, it significantly improves the prediction accuracy of net load-voltage mapping relationship and provides an accurate model basis for the subsequent quantitative transformation of voltage constraints.

[0030] The Gaussian process regression model can simultaneously output the mean of voltage prediction and the Gaussian noise term. It can not only achieve a quantitative mapping of net load to voltage, but also quantify the uncertainty of prediction results. This provides data support for the probability assessment of voltage over-limit risk in subsequent steps, realizing a technical extension from deterministic mapping to probabilistic safety assessment.

[0031] The step of transforming the safety operation constraint of the threshold voltage into an adjustable net load boundary for each time period through inverse mapping includes: Under voltage constraint conditions Using the inverse mapping capability of Gaussian process regression, the boundary values ​​of net load for each time period are calculated: In the formula, and They represent in The net load limit is ensured at all times, ensuring that the voltage at the cut-off point is not lower than the minimum value and does not exceed the maximum value, thus forming an adjustable boundary for the net load under the voltage constraint of the transformer area.

[0032] The steps for calculating the threshold voltage include: The net load sequence corresponding to the optimized scheduling result is input into the established net load-gate voltage mapping model. The gate voltage for each time period is calculated. Based on the Gaussian process regression model, the predicted distribution of the gate voltage is obtained. The following formula is used for calculation: In the formula, for Predicted average voltage at time-bound thresholds To predict variance.

[0033] The steps for constraining and verifying the voltage include: The voltage prediction values ​​at each time point are constrained and checked to determine whether they meet the voltage operating range. The calculation formula is as follows: ; If the predicted average voltage at any time point exceeds the allowable range, the scheduling scheme is deemed not to meet the voltage constraint requirements.

[0034] The steps for conducting a voltage over-limit risk assessment include: Based on the prediction variance output by the Gaussian process regression model, a probabilistic assessment of the risk of voltage exceedance at the threshold point is performed. The calculation formula is as follows: In the formula, This is the cumulative distribution function of the standard normal distribution.

[0035] The present invention provides a specific embodiment as follows: This invention relates to a photovoltaic-storage-charging integrated coordinated control method based on net load characteristic response, which specifically includes the following steps: Step 1: Process the day-ahead forecast data of the load and photovoltaic output of the 0.4kV distribution substation to construct the net load time series. Extract four indicators from the series: peak-to-valley difference, coefficient of variation, maximum ramp rate, and average power change in adjacent time periods to form a complete net load characteristic vector.

[0036] Specifically, step 1 includes the following steps: S101: Collect historical operating data of the 0.4kV distribution transformer substation, including: load power data of the distribution transformer substation. Photovoltaic power output data Electric vehicle charging and discharging data, and power data of adjustable load equipment. All data are sampled at 15-minute intervals, forming a time-series data sequence of 96 time points over 24 hours.

[0037] S102: Process the collected raw data to ensure data integrity and continuity. Specific methods are as follows: Outlier removal: Remove outlier points from the load or photovoltaic power data. To perform identification, it must satisfy the following equation (1): (1); treat it as an outlier and replace it with the mean of the nearest time period, where Adjacent The average of each time period, Standard deviation, To set a threshold; missing value imputation: for missing data points Linear interpolation is used for completion, as shown in equation (2): (2); Duplicate value merging: If multiple sampled values ​​exist at the same time point... Then, the mean value is taken as the unique value, as shown in equation (3): (3) Time alignment processing: All data are aligned to a 15-minute time scale to form a continuous net load time series.

[0038] S103: Net load calculation. Based on the processed load data and photovoltaic output data, the net load sequence of the distribution area is calculated as shown in equation (4): (4); where, express Net load at any given moment.

[0039] S104: Net Load Feature Extraction. Four characteristic indicators are extracted from the net load sequence, including peak-to-valley difference. As shown in equation (5): (5); coefficient of variation As shown in equation (6): (6); In the formula, For the standard deviation operator, This is the mean operator. Maximum gradeability. As shown in equation (7): (7); Average power change over adjacent time periods As shown in equation (8): (8).

[0040] Step 2, Net Load Voltage Mapping Construction Method: Based on the transformer substation topology and historical operating data, a data-driven approach is used to establish the mapping relationship between net load and threshold voltage. The mapping model converts the upper and lower voltage limits into corresponding net load boundaries, enabling voltage constraints to be incorporated into the scheduling optimization model in the form of net load constraints, providing operable quantitative constraints for scheduling decisions. Specifically, Step 2 includes the following steps: S201, Dataset partitioning: Collect the net load sequence and transformer area threshold voltage sequence obtained in step 1 to form a training set. All data are aligned according to a 15-minute sampling interval to ensure input-output consistency, as shown in equation (9): (9).

[0041] S202, The mapping model is established. The radial basis function kernel function is selected, and the kernel matrix is ​​defined as shown in equation (10): In the formula, Let the kernel width be the parameter; establish a Gaussian process regression model, as shown in equation (10): (10); Vt represents the voltage at the junction of the transformer area at time t; Wt represents the net load at time t; f(·) is a nonlinear mapping function obtained by Gaussian process regression fitting; ε is a Gaussian noise term, following a normal distribution with mean 0 and variance σ², i.e., ε∼N(0,σ²). The kernel parameters are determined by maximizing the log-likelihood function. and noise variance As shown in equation (11): (11); where, , , It is the identity matrix. The log-likelihood function can be maximized to obtain the result. and .

[0042] S203: Voltage constraint boundary calculation, under voltage constraint conditions, as shown in equation (12): (12); Calculate the upper and lower limits of net load, as shown in equation (13): (13) By solving the mapping function in reverse, the voltage constraint is mapped to the net load range to obtain the adjustable net load upper and lower limit sequence.

[0043] S204: Boundary sequence generation, generating each time point and Arranged in 15-minute intervals, a complete net load adjustable boundary sequence is formed, providing constraints for the multi-objective day-ahead scheduling model in step 3.

[0044] Step 3 involves multi-objective day-ahead scheduling modeling based on net load response, constructing a multi-objective day-ahead scheduling model for photovoltaic, energy storage, and adjustable load devices. The scheduling objectives include minimizing the net load peak-to-valley difference, minimizing the net load standard deviation, minimizing power variation between adjacent time periods, and minimizing system operating costs. Model constraints include energy storage state of charge and power limitations, electric vehicle charging demand and power constraints, photovoltaic output upper limit constraints, and net load upper and lower limits and peak-to-valley difference constraints. By solving this model, the optimized net load time series that satisfies the constraints is obtained, i.e., the remaining net load sequence. Specifically, Step 3 includes the following steps: S301, Scheduling Time Scale and Variable Definition: Constructing the day-ahead scheduling time domain, discretizing the 24 hours into 96 time periods in 15-minute intervals, and defining the time period set. Powered by photovoltaics Energy storage charging and discharging power and adjustable load regulation power The decision variable is used to form the net load after scheduling, i.e., the remaining net load, as shown in equation (14): (14) At the same time, the interaction power with the upper-level power grid is defined as... This satisfies the power balance relationship.

[0045] S302, Objective function construction: Constructing a multi-objective optimization model containing five sub-objectives: The peak-valley difference target is shown in equation (15): (15); The number of variants is shown in equation (16): (16); where, For the standard deviation operator, The mean operator is used; the target for the maximum gradeability is shown in equation (17): (17); The target for the average power change over adjacent time periods is shown in equation (18): (18); The operating cost target is shown in equation (19): (19); where, This is the unit operating cost coefficient for energy storage. This is the unit adjustment cost coefficient for adjustable load. for Electricity purchase price coefficient for different time periods for The power purchased by the distribution area from the superior power grid during the time period.

[0046] S303: Constraint modeling. In multi-objective day-ahead scheduling, constraints are established, including: power balance constraint of distribution area: after scheduling, the net load is balanced with photovoltaic, energy storage, adjustable load and purchased power, as shown in equation (20): (20); Voltage constraint boundary: The net load must satisfy the voltage constraint mapping established in step 2, as shown in equation (21): (21); Energy storage operation constraints: the charging and discharging power is shown in equation (22) and the SOC constraint is shown in equation (23): (twenty two); (23); Adjustable load constraints: Adjustable power is subject to upper and lower limits, as shown in equation (24): (24); Photovoltaic output constraints are shown in equation (25): (25); where, for The predicted maximum power generation of photovoltaic power at any given time; Flexibility management constraints: In order to achieve flexible management of the distribution area, it is necessary to ensure that the charging energy of the energy storage in the area is equal to the discharging energy within a day, and to ensure that the state of charge of the energy storage remains unchanged, as shown in equation (26): (26); where, The energy used for energy storage and charging. The energy stored during discharge.

[0047] S304, Solution Method: The NSGA-III algorithm is used for multi-objective optimization, suitable for handling problems with multiple conflicting objectives. The algorithm steps include: Initialize the population and reference direction; calculate the fitness of each individual in the target space and perform non-dominated sorting; update the population by selecting the crowding distance with the reference point assistance; iterate until the convergence condition is met to obtain the Pareto front solution set.

[0048] S305: Equilibrium solution selection in the Pareto solution set. For the Pareto solution set obtained by NSGA-III, the fuzzy decision method based on normalized membership degree is used to select the equilibrium solution. For each objective function Each solution below Calculate the normalized membership values, as shown in equation (27): (27); The fuzzy decision value for each solution is calculated as shown in equation (28): (28); Choice The largest solution is taken as the balanced solution, which is the final scheduling scheme.

[0049] Step 4, Calculation and Verification of Remaining Net Load Voltage Response: Substitute the optimized remaining net load into the net load-voltage mapping model to calculate the voltage response at each time point and evaluate the upper and lower voltage limits and fluctuations. By comparing the net load characteristics and voltage indicators before and after optimization, the effectiveness of the scheduling strategy is determined. If voltage exceedance occurs, the scheduling model parameters can be adjusted and the solution recalculated until all threshold voltages meet the constraints, achieving closed-loop control and operational verification. Specifically, Step 4 includes the following steps: S401: Net load input and voltage calculation. Input the net load sequence corresponding to the optimized scheduling result obtained in step 3 into the net load-gate voltage mapping model established in step 2, and calculate the gate voltage for each time period. Based on the Gaussian process regression model, the predicted distribution of the gate voltage is obtained, as shown in equation (29): (29); where, for Predicted average voltage at time-bound thresholds To predict variance.

[0050] S402: Voltage constraint check, the voltage prediction value at the threshold point in each time period is constrained and checked to determine whether it meets the voltage operating range, as shown in equation (30): (30); If the predicted average voltage at any time point exceeds the allowable range, the scheduling scheme is deemed not to meet the voltage constraint requirements.

[0051] S403: Voltage over-limit risk assessment. Based on the prediction variance output by the Gaussian process regression model, the probability assessment of the voltage over-limit risk at the threshold point is performed, as shown in Equation (31): (31); where, The cumulative distribution function is a standard normal distribution; the operational risk level of the scheduling scheme is quantified by calculating the voltage over-limit probability in each time period.

[0052] S404: Control effect judgment and feedback mechanism. When the voltage prediction at all time points meets the constraint range and the probability of exceeding the limit is lower than the set threshold, the scheduling result is judged to meet the voltage safety operation requirements. When there are time periods that do not meet the conditions, the corresponding net load scheduling result is fed back to step 3 as input to re-optimize the scheduling variables until the voltage constraint conditions are met.

[0053] This invention constructs a net load characteristic system that includes peak-to-valley difference, coefficient of variation, maximum ramp rate, and average power change over adjacent time periods. This system enables a unified quantitative characterization of the fluctuations, relative dispersion, instantaneous rate of change, and continuous stability of the net load in a distribution substation. The constructed multidimensional characteristic index comprehensively depicts the dynamic behavior of the net load, effectively reflects the impact mechanism of load fluctuations on the voltage at the cut-off point, and improves the accuracy of the net load characteristics in representing voltage changes.

[0054] This invention establishes a quantitative mapping relationship between net load and threshold voltage based on Gaussian process regression. Furthermore, it transforms voltage operation constraints into upper and lower limits of net load for each time period, achieving a direct mapping of voltage constraints to scheduling variables. This method embeds voltage safety constraints into the net load regulation process, transforming voltage control from ex-post verification to ex-ante constraint, thus improving the feasibility and consistency of scheduling results.

[0055] This invention constructs a multi-objective day-ahead scheduling model that incorporates peak-to-valley difference, coefficient of variation, maximum ramp rate, average power variation in adjacent time periods, and operating costs to achieve coordinated and optimized scheduling of photovoltaic, energy storage, and adjustable loads. By introducing the NSGA-III algorithm to solve the multi-objective optimization problem and combining it with the normalized membership method to select the optimal scheduling scheme, coordinated optimization among multiple objectives is achieved, improving the scheduling model's comprehensive control capability over net load fluctuation characteristics and economic efficiency.

[0056] After scheduling optimization, this invention substitutes the optimized net load result into the net load-voltage mapping model for threshold voltage calculation and over-limit risk assessment, and uses a feedback mechanism to achieve closed-loop verification and correction of the scheduling results. This method realizes consistency verification between the net load regulation process and voltage response, forming a closed-loop control system of "characteristic extraction-mapping modeling-optimized scheduling-voltage verification," which improves the voltage safety operation level and control reliability of distribution transformer areas under conditions of high proportion of distributed power source access.

[0057] This invention provides a photovoltaic-storage-charging integrated coordinated control device based on net load characteristic response, comprising: The feature extraction module is used to perform time-series modeling and data preprocessing of the day-ahead net load of the distribution area, and extract multi-dimensional feature indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. The mapping boundary module is used to establish a quantitative mapping relationship between net load and transformer area threshold voltage based on multi-dimensional net load characteristic indicators and using a data-driven method. Through inverse mapping, the safety operation constraints of threshold voltage are transformed into adjustable net load boundaries for each time period. The day-ahead scheduling module is used to construct a multi-objective day-ahead scheduling model that integrates multiple adjustable resources such as photovoltaics, energy storage, and adjustable load, with the net load adjustable boundary as the core constraint, and obtain the optimal day-ahead scheduling strategy. The verification module is used to substitute the obtained remaining net load into the established quantitative mapping relationship between net load and transformer area threshold voltage, complete the threshold voltage calculation, constraint verification and over-limit risk assessment, and re-optimize the scheduling results that do not meet voltage safety requirements through closed-loop feedback.

[0058] The mapping boundary module includes: Quantitative units are used to perform Gaussian process regression on historical net load sequences. Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

[0059] In addition, embodiments of the present invention also provide a computer-readable storage medium.

[0060] The present invention stores a photoelectric storage and charging integrated coordinated control program based on net load characteristic response on a computer-readable storage medium, wherein when the photoelectric storage and charging integrated coordinated control program based on net load characteristic response is executed by a processor, the steps of the photoelectric storage and charging integrated coordinated control method based on net load characteristic response described above are implemented.

[0061] The method implemented when the integrated photovoltaic-storage-charging coordinated control program based on net load characteristic response is executed can be referred to in various embodiments of the integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response of the present invention, and will not be repeated here.

[0062] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0063] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0064] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0065] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0066] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are exhaustively listed. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0067] For those skilled in the art, various modifications and improvements can be made without departing from the concept of the present invention, and these modifications and improvements are all within the scope of protection of the present invention. The scope of protection of the present invention is defined by the appended claims.

Claims

1. A method for integrated photovoltaic-storage-charging control based on net load characteristic response, comprising the following steps: Time-series modeling and data preprocessing are performed on the day-ahead net load of the distribution substation to extract multi-dimensional characteristic indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. Based on the multidimensional characteristic indicators of net load, a data-driven method is used to establish a quantitative mapping relationship between net load and the voltage at the cut-off point of the transformer area. Through inverse mapping, the safety operation constraint of the cut-off point voltage is transformed into the adjustable boundary of net load for each time period. Using the adjustable net load boundary as the core constraint, a multi-objective day-ahead scheduling model integrating multiple adjustable resources such as photovoltaic, energy storage, and adjustable load is constructed to obtain the optimal day-ahead scheduling strategy; The remaining net load is substituted into the established quantitative mapping relationship between the net load and the voltage at the control point of the transformer area to complete the voltage calculation, constraint verification and over-limit risk assessment at the control point, and the closed-loop feedback of the dispatch results that do not meet the voltage safety requirements is re-optimized.

2. The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response according to claim 1, characterized in that, The method of establishing a quantitative mapping relationship between net load and transformer substation voltage using a data-driven approach includes: using a Gaussian process regression method to analyze the historical net load sequence. Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

3. The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response according to claim 2, characterized in that, The step of transforming the safety operation constraint of the threshold voltage into an adjustable net load boundary for each time period through inverse mapping includes: Under voltage constraint conditions Using the inverse mapping capability of Gaussian process regression, the boundary values ​​of net load for each time period are calculated: In the formula, and They represent in The net load limit is ensured at all times, ensuring that the voltage at the cut-off point is not lower than the minimum value and does not exceed the maximum value, thus forming an adjustable boundary for the net load under the voltage constraint of the transformer area.

4. The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response according to claim 1, characterized in that, The steps for calculating the threshold voltage include: The net load sequence corresponding to the optimized scheduling result is input into the established net load-gate voltage mapping model. The gate voltage for each time period is calculated. Based on the Gaussian process regression model, the predicted distribution of the gate voltage is obtained. The following formula is used for calculation: In the formula, for Predicted average voltage at time-bound thresholds To predict variance.

5. The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response according to claim 4, characterized in that, The steps for constraining and verifying the voltage include: The voltage prediction values ​​at each time point are constrained and checked to determine whether they meet the voltage operating range. The calculation formula is as follows: ; If the predicted average voltage at any time point exceeds the allowable range, the scheduling scheme is deemed not to meet the voltage constraint requirements.

6. The integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response according to claim 5, characterized in that, The steps for conducting a voltage over-limit risk assessment include: Based on the prediction variance output by the Gaussian process regression model, a probabilistic assessment of the risk of voltage exceedance at the threshold point is performed. The calculation formula is as follows: In the formula, This is the cumulative distribution function of the standard normal distribution.

7. A photovoltaic-storage-charging integrated coordinated control device based on net load characteristic response, characterized in that, include: The feature extraction module is used to perform time-series modeling and data preprocessing of the day-ahead net load of the distribution area, and extract multi-dimensional feature indicators that can comprehensively characterize the fluctuation of net load amplitude, relative dispersion, instantaneous rate of change and continuous stability of change. The mapping boundary module is used to establish a quantitative mapping relationship between net load and transformer area threshold voltage based on multi-dimensional net load characteristic indicators and using a data-driven method. Through inverse mapping, the safety operation constraints of threshold voltage are transformed into adjustable net load boundaries for each time period. The day-ahead scheduling module is used to construct a multi-objective day-ahead scheduling model that integrates multiple adjustable resources such as photovoltaics, energy storage, and adjustable loads, with the net load adjustable boundary as the core constraint, and obtain the optimal day-ahead scheduling strategy. The verification module is used to substitute the obtained remaining net load into the established quantitative mapping relationship between net load and transformer area threshold voltage, complete the threshold voltage calculation, constraint verification and over-limit risk assessment, and re-optimize the scheduling results that do not meet voltage safety requirements through closed-loop feedback.

8. The integrated photovoltaic-storage-charging coordinated control device based on net load characteristic response according to claim 7, characterized in that, The mapping boundary module includes: Quantitative units are used to perform Gaussian process regression on historical net load sequences. Voltage at the corresponding time point Using the data as the training set, a nonlinear regression model is constructed through a kernel function to obtain the continuous mapping relationship between net load and threshold voltage: In the formula, express Voltage at the switch point in the time zone Net load, For Gaussian noise, the function It was determined by fitting historical data using Gaussian process regression.

9. A photovoltaic-storage-charging integrated coordinated control device based on net load characteristic response, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the integrated photovoltaic-storage-charging control method based on net load characteristic response as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the integrated photovoltaic-storage-charging coordinated control method based on net load characteristic response as described in any one of claims 1 to 6.

Images