Multi-source data fusion photovoltaic power scheduling control system

Abstract

The invention relates to the technical field of power management, in particular to a multi-source data fusion photovoltaic power scheduling control system, which comprises a sensing extraction module, a disturbance identification module, an offset trend analysis module, a path switching module and a frequency band suppression module. According to the method, a disturbance pre-judgment interval is identified by constructing a time derivative sequence and combining illumination and wind speed trend change in a continuous period, a disturbance time period is accurately discriminated, the disturbance interval is confirmed and disturbance characteristics are extracted through conjoint analysis of photovoltaic and energy storage current change trend and grid-connected point fluctuation amplitude, and the disturbance accuracy is improved. An offset trend curve is formed by calculating photovoltaic and energy storage current change differences, the energy supply offset degree is judged in real time to trigger scheduling conditions, dynamic suppression adjustment is performed on regulation and control signal parameters according to spectrum energy distribution, the accuracy and timeliness of disturbance recognition, energy supply offset judgment and frequency band regulation and control are improved from multiple levels, and the energy storage and energy supply efficiency is improved. And more stable and efficient optical storage combined scheduling control is realized.

Description

Technical Field

[0001] This invention relates to the field of power management technology, and in particular to a photovoltaic power dispatch and control system that integrates multi-source data. Background Technology

[0002] Power management technology involves the efficient scheduling, allocation, and conversion of energy, aiming to improve the overall operational efficiency and stability of energy systems. This technology primarily includes the intelligent coordination and optimized management of energy generation, storage, transmission, and utilization. Its core aspects include grid power flow control, renewable energy access management, power quality regulation, load forecasting and scheduling, battery management, and optimized configuration of multi-energy complementary systems. Especially in scenarios such as smart grids, distributed energy systems, and new energy grid integration, power management technology ensures stable system operation and optimal energy efficiency through the comprehensive scheduling of various power sources, loads, and energy storage devices. Traditional multi-source data fusion photovoltaic power dispatch control systems refer to systems that achieve dynamic scheduling control of photovoltaic power generation system power output by comprehensively analyzing and processing data from multiple sources, including photovoltaic power plants, environmental monitoring equipment, grid operating status, and load data. Traditional systems allocate power by setting fixed scheduling strategies or rule-based algorithms. The inputs include collecting environmental parameters such as light intensity, temperature, and wind speed, combining this with current power generation conditions to predict power output, and determining photovoltaic power output based on grid load demand. Its control methods mainly rely on time series analysis, physical model calculations, or expert experience rules to formulate strategies, aiming to coordinate the dynamic balance between photovoltaic power output and grid supply and demand.

[0003] Traditional photovoltaic power dispatch control systems based on multi-source data fusion mainly rely on preset rules, expert experience, and time series analysis for strategy formulation during operation. They lack the ability to identify and predict sudden disturbances in real time. Under complex weather conditions, they are slow to respond to nonlinear changes in light intensity and wind speed, resulting in delayed dispatch response and low control accuracy. When judging the source of photovoltaic output fluctuations, they fail to fully distinguish between short-term disturbances caused by environmental changes and power shifts caused by load fluctuations. This can easily lead to frequent switching of energy storage dispatch, increasing the system's operating burden and reducing energy storage efficiency. Especially when the grid connection point current fluctuates frequently, the system is unable to make timely and effective adjustments, affecting the stable supply and demand coordination between the photovoltaic power generation system and the grid. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a photovoltaic power dispatch and control system based on multi-source data fusion.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a photovoltaic power dispatch control system with multi-source data fusion includes:

[0006] The perception and extraction module obtains the irradiance value sequence and wind speed value sequence, constructs the time derivative sequence based on the difference in irradiance values ​​at adjacent time points, calls the deviation between the time derivative sequence and the continuous period average derivative, and judges whether the direction of wind speed continuous period trend change is consistent, and generates the fusion disturbance prediction interval.

[0007] The disturbance identification module calls the fusion disturbance prediction interval, obtains the photovoltaic current output value change sequence and the energy storage current change sequence, performs the consistency judgment of the current increase and decrease direction in the same period, analyzes the total fluctuation amplitude of the grid connection point current within the time period, and generates a disturbance feature confirmation window.

[0008] The offset trend analysis module calls the disturbance feature confirmation window to calculate the difference between the change in photovoltaic current and the change in energy storage current at the same time point, and determines whether the real-time power supply offset meets the scheduling start conditions to obtain the photovoltaic-storage offset scheduling trigger point.

[0009] Based on the photovoltaic-storage offset scheduling trigger point, the path switching module obtains the power converter channel status and the remaining available energy storage time, and combines the real-time photovoltaic output trend to determine whether the power supply path switching meets the main channel switching conditions, and generates a path master-slave switching control command.

[0010] As a further aspect of the present invention, the fusion disturbance prediction interval includes disturbance start time, disturbance end time, disturbance amplitude determination basis, and disturbance trend consistency label; the disturbance feature confirmation window includes current change consistency mark, total current fluctuation amplitude threshold type, and disturbance cycle identification label; the photovoltaic-storage offset scheduling trigger point includes offset degree value, offset trend direction, and scheduling trigger threshold status; and the path master-slave switching control command includes energy storage channel identifier, photovoltaic channel auxiliary status, and control signal command type.

[0011] As a further aspect of the present invention, the perception extraction module includes:

[0012] The irradiance calculation submodule obtains the irradiance value sequence and wind speed value sequence, performs item-by-item difference calculation on the irradiance values ​​at adjacent time points, establishes the irradiance change derivative sequence under equal time intervals, divides the derivative sequence into multiple period segments according to the set time window, solves the mean value of the derivative value in each period segment, calls the deviation of the period derivative mean value and the time point derivative to calculate the difference, and obtains the irradiance derivative offset.

[0013] The wind speed trend analysis submodule calls the illumination derivative offset to obtain the wind speed value sequence for the time period corresponding to the illumination derivative offset. It divides the wind speed value sequence into continuous period intervals, calculates the difference between the average wind speed values ​​within the period intervals, determines the positive or negative sign of the change in the average wind speed value during the period, identifies the direction of the continuous change trend, and obtains the wind speed trend consistency indicator sequence.

[0014] The disturbance interval marking submodule calculates the disturbance enhancement index based on the wind speed trend consistency marker sequence, in order to meet the need for adjusting and judging the disturbance signal in the wind-solar joint disturbance identification scenario. It then compares the disturbance enhancement index with the set disturbance marking threshold point by point, identifies the time points that are greater than the disturbance marking threshold as disturbance points, and aggregates the disturbance points according to the time continuity condition to establish a fusion disturbance prediction interval.

[0015] As a further aspect of the present invention, the disturbance identification module includes:

[0016] The sequence change extraction submodule obtains the photovoltaic current output value change sequence and the energy storage current change sequence based on the fusion disturbance prediction interval. It divides the photovoltaic current output value change sequence and the energy storage current change sequence into a periodic dimension, and extracts the current increase and decrease direction of the photovoltaic current output value change sequence and the energy storage current change sequence at adjacent time points to obtain the periodic current change direction sequence.

[0017] The direction consistency judgment submodule calls the periodic current change direction sequence, and performs synchronization judgment on the photovoltaic and energy storage current change directions according to the period. Periods with consistent directions within the same period are marked as same direction, and the remaining periods are marked as opposite direction. Combined with the timing information, the continuous same direction period segments are integrated to obtain the continuous same direction period interval set.

[0018] The fluctuation amplitude filtering submodule extracts the numerical change sequence of the grid-connected point current in the corresponding time period based on the set of continuous unidirectional periodic intervals, calculates the fluctuation offset amplitude value, and performs a judgment operation on the continuous unidirectional periodic intervals based on the difference between the fluctuation offset amplitude value and the set disturbance identification benchmark fluctuation amplitude to obtain the disturbance feature confirmation window.

[0019] As a further aspect of the present invention, the fluctuation offset amplitude value is expressed by the formula:

[0020] ;

[0021] in, Representing the The amplitude of fluctuation offset in each unidirectional periodic segment. Indicates the number of periodic segments in the same direction. Indicates the first Changes in photovoltaic current value within a period Indicates the first Changes in energy storage current value within each cycle Indicates the first The disturbance identification benchmark fluctuation amplitude corresponding to each period segment.

[0022] As a further aspect of the present invention, the offset trend analysis module includes:

[0023] The data stream receiving submodule calls the disturbance feature confirmation window to obtain the photovoltaic output current time series and the energy storage output current time series. According to the internally set time step parameters, the photovoltaic output current time series and the energy storage output current time series are aligned to a unified time axis to generate a photovoltaic-storage synchronous current time segment.

[0024] The difference extraction submodule calculates the difference sequence between the photovoltaic current value and the energy storage current value at time points based on the photovoltaic current value and the energy storage current value in the photovoltaic-storage synchronous current time segment, constructs a curve-shaped data structure with time as the horizontal axis and difference as the vertical axis, and generates a photovoltaic-storage current difference trend vector.

[0025] The trend determination submodule calls the photovoltaic-storage current difference trend vector, combines the sign and magnitude of the slope of the difference change within a continuous time period, extracts the change direction of the real-time difference trend, and matches the change direction value with the set power supply offset start judgment benchmark value for difference direction judgment. If the trend direction of multiple consecutive points is consistent and the magnitude exceeds the judgment benchmark value, it is determined that the power supply offset start is established, and the photovoltaic-storage offset scheduling trigger point is obtained.

[0026] As a further aspect of the present invention, the path switching module includes:

[0027] Based on the photovoltaic-storage offset scheduling trigger point, the channel information extraction submodule obtains the real-time channel working status identifier of the power converter and the remaining energy storage power monitoring value, calculates the remaining available time of energy storage under real-time discharge power conditions, and simultaneously obtains the photovoltaic module output power time series within the time period. Based on the time series change trend, it extracts the photovoltaic power change direction and generates a channel operation status information set.

[0028] The path switching condition judgment submodule calls the channel operation status information set, extracts the remaining available time value of energy storage and the photovoltaic power change direction value, compares the available time with the path switching continuous benchmark value, and determines whether it has the ability to maintain the main channel power supply. At the same time, it detects whether the photovoltaic power trend is in a continuous downward state. If both conditions are met, it is determined that the path switching requirements are met in real time, and the path switching status judgment result is obtained.

[0029] Based on the path switching status determination result, if the status meets the conditions, the command control submodule constructs a control signal structure and sets the energy storage path as the main channel identifier and the photovoltaic path as the auxiliary channel identifier. The control signal is then uploaded to the edge controller control terminal through the data interface to generate the path master-slave switching control command.

[0030] As a further aspect of the present invention, the system also includes a frequency band suppression module:

[0031] The frequency band suppression module calls the path master-slave switching control command, obtains real-time data of grid connection point voltage and current, constructs complex impedance response spectrum within the cycle, extracts the main frequency position and identifies the frequency response drop bandwidth, obtains real-time periodic energy storage output power data and performs Fourier spectrum decomposition. If the main energy component of the spectrum is within the frequency drop bandwidth, the modulation time parameter in the control signal is adjusted to suppress the frequency band output energy density and generate the frequency band power injection compression result.

[0032] The frequency band power injection compression result includes the frequency band of the main spectral components, the energy density compression amplitude, and the modulation time adjustment parameters.

[0033] As a further aspect of the present invention, the frequency band suppression module includes:

[0034] The complex impedance spectrum construction submodule calls the path master-slave switching control command to obtain real-time three-phase voltage and current sampling data of the grid connection point, performs complex domain amplitude and phase angle calculation according to the set period, constructs a complex impedance response value sequence of time segment, and forms a response spectrum curve according to the frequency domain distribution. It extracts the position of the main frequency of the complex impedance modulus in the response spectrum curve, uses the main frequency as the center to detect the width of the frequency domain energy drop interval, and generates a frequency drop bandwidth parameter group.

[0035] The spectrum principal component identification submodule obtains the time-series data of active and reactive power output of energy storage within the real-time period through the frequency drop bandwidth parameter group, performs Fourier transform operation to obtain the spectrum energy distribution map, identifies the main energy concentration frequency band in the spectrum energy distribution map, and performs frequency interval matching with the frequency drop bandwidth parameter group to determine whether the main energy component is within the drop bandwidth and generates the main frequency energy matching status value.

[0036] The frequency energy suppression submodule, based on the main frequency energy matching status value, if the status determines that the main frequency energy is within the falling bandwidth, calls the original control signal field, performs compression adjustment processing on the modulation time parameter item, adjusts the modulation window length and amplitude factor according to the bandwidth range, re-encapsulates the control field and completes the energy density control configuration, and generates the frequency band power injection compression result.

[0037] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0038] In this invention, a disturbance prediction interval is identified by constructing a time derivative sequence and combining it with the trend changes of illumination and wind speed within a continuous period. The disturbance period is accurately identified. By jointly analyzing the trend of photovoltaic and energy storage current changes and the fluctuation amplitude of grid connection point, the disturbance interval is confirmed and the disturbance characteristics are extracted. By calculating the difference in changes of photovoltaic and energy storage currents to form an offset trend curve, the degree of energy supply offset is judged in real time to trigger scheduling conditions. The path switching is evaluated in combination with the power channel status and energy storage duration. The control signal parameters are dynamically suppressed and adjusted according to the spectrum energy distribution. The accuracy and timeliness of disturbance identification, energy supply offset judgment and frequency band control are improved from multiple levels, so as to achieve more stable and efficient photovoltaic and energy storage joint scheduling control. Attached Figure Description

[0039] Figure 1 This is a system flowchart of the present invention;

[0040] Figure 2 This is a flowchart of the perception extraction module in this invention;

[0041] Figure 3 This is a flowchart of the disturbance identification module in this invention;

[0042] Figure 4 This is a flowchart of the offset trend analysis module in this invention;

[0043] Figure 5 This is a flowchart of the path switching module in this invention;

[0044] Figure 6 This is a flowchart of the mid-frequency band suppression module of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0046] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0047] Please see Figure 1A photovoltaic power dispatch control system with multi-source data fusion includes:

[0048] The perception and extraction module obtains the irradiance value sequence and wind speed value sequence, constructs the time derivative sequence based on the difference in irradiance values ​​at adjacent time points, calls the deviation between the time derivative sequence and the continuous period average derivative, and judges whether the direction of wind speed continuous period trend change is consistent. The time period that meets the threshold judgment condition is marked as the disturbance time period, and a fusion disturbance prediction interval is generated.

[0049] The disturbance identification module calls the fusion disturbance prediction interval, obtains the photovoltaic current output value change sequence and the energy storage current change sequence, performs the consistency judgment of the current increase and decrease direction in the same period, analyzes the total fluctuation amplitude of the grid-connected point current in the time period, and if the fluctuation amplitude is higher than the disturbance identification benchmark fluctuation amplitude and the current change direction is consistent, the time period is confirmed as a disturbance interval and a disturbance feature confirmation window is generated.

[0050] The offset trend analysis module calls the disturbance feature confirmation window to obtain the time series of photovoltaic output current and energy storage output current, calculates the difference between the change in photovoltaic current and the change in energy storage current at the same time point, constructs the photovoltaic-energy storage offset trend curve, and judges whether the real-time power supply offset degree meets the scheduling start condition based on the change direction of the photovoltaic-energy storage offset trend curve, and obtains the photovoltaic-energy storage offset scheduling trigger point.

[0051] The path switching module obtains the power converter channel status and the remaining available time of energy storage based on the photovoltaic-storage offset scheduling trigger point. It combines the real-time photovoltaic output trend to determine whether the power supply path switching meets the main channel switching conditions. If the switching conditions are met, it issues a regulation control signal, designates the energy storage path as the main channel, marks the photovoltaic channel as the auxiliary path, and uploads the regulation signal instruction to the edge controller control terminal to generate the path master-slave switching control instruction.

[0052] The frequency band suppression module calls the path master-slave switching control command, obtains real-time data of grid connection point voltage and current, constructs the complex impedance response spectrum within the cycle, extracts the main frequency position and identifies the frequency response drop bandwidth, obtains real-time periodic energy storage output power data and performs Fourier spectrum decomposition. If the main energy component of the spectrum is within the frequency drop bandwidth, the modulation time parameter in the control signal is adjusted to suppress the frequency band output energy density and generate the frequency band power injection compression result.

[0053] The integrated disturbance prediction range includes disturbance start time, disturbance end time, disturbance amplitude determination criteria, and disturbance trend consistency label. The disturbance feature confirmation window includes current change consistency mark, total current fluctuation amplitude threshold type, and disturbance cycle identification label. The photovoltaic-storage offset scheduling trigger point includes offset degree value, offset trend direction, and scheduling trigger threshold status. The path master-slave switching control command includes energy storage channel identifier, photovoltaic channel auxiliary status, and control signal command type. The frequency band power injection compression result includes spectrum principal component frequency band, energy density compression amplitude, and modulation time adjustment parameters.

[0054] Please see Figure 2 The perception extraction module includes:

[0055] The irradiance calculation submodule obtains the irradiance value sequence and wind speed value sequence, performs item-by-item difference calculation on the irradiance values ​​at adjacent time points, establishes the irradiance change derivative sequence under equal time intervals, divides the derivative sequence into multiple period segments according to the set time window, solves the mean value of the derivative value in each period segment, calls the deviation of the period derivative mean value and the time point derivative to calculate the difference, and obtains the irradiance derivative offset.

[0056] A representative monitoring point is selected from the area to be analyzed. Assuming data is collected at each monitoring point every minute, a time-series data set is formed. The difference between adjacent time points at the monitoring point is calculated, with a difference window length of 5 minutes (i.e., a period of 5 minutes). This yields a derivative sequence. During the difference calculation, each time point... Illumination value Compared to the previous time point The difference is calculated, and the derivative is... To reflect the variation in light intensity per unit time, the total sequence is divided into multiple equal-length periodic segments based on the user-defined time window length, such as a 60-minute analysis window. The light derivative sequence within each periodic segment is processed to obtain the mean, standard deviation, and extreme value range of each segment's derivative sequence, which are used to reflect the trend of light change within the time period. After obtaining the mean of the derivative for each periodic segment, they are arranged in chronological order to form a periodic derivative value sequence. The deviation between the periodic derivative value and the derivative at a given time point is used to calculate the difference. For example, if the derivative at the current time point is 6.5 and the mean of the derivative in its periodic segment is 3.2, then the deviation value is 3.3. This deviation value is used as the light derivative offset at the current time point for wind speed trend comparison processing to obtain the light derivative offset.

[0057] The wind speed trend analysis submodule calls the illumination derivative offset to obtain the wind speed value sequence for the time period corresponding to the illumination derivative offset. It divides the wind speed value sequence into continuous period intervals, calculates the difference between the mean wind speed values ​​within the period intervals, determines the positive or negative sign of the change in the mean wind speed during the period, identifies the direction of the continuous change trend, and obtains the wind speed trend consistency indicator sequence.

[0058] Obtain the wind speed value sequence corresponding to the time period of the light derivative numerical offset. Assuming the acquisition frequency is consistent with the light intensity, the wind speed value is recorded once per minute. Divide the wind speed value sequence into multiple sub-segments according to the light intensity cycle. Each segment is consistent with the light derivative cycle. Calculate the difference between the wind speed value in each segment and the average wind speed in the cycle to determine whether the wind speed shows a continuous upward or downward trend. Obtain the starting and ending wind speed values ​​in each segment. If the ending value is greater than the starting value, it is recorded as a positive trend; otherwise, it is a negative trend. For example, if the starting wind speed of the first segment is 2.3 m / s and the ending wind speed is 3.1 m / s, it is recorded as an upward trend. Record the trend direction of the cycle segment in the form of "+1" or "1". That is, if the trend direction of three consecutive cycles is +1, the trend consistency is "upward". If it is "+1, 1, +1", it is considered as an inconsistent trend. Generate a wind speed trend consistency flag sequence.

[0059] The disturbance interval marking submodule, based on the wind speed trend consistency marker sequence, adopts the following formula to address the need for adjusting and judging disturbance signals in wind-solar joint disturbance identification scenarios:

[0060] ,

[0061] Calculate the disturbance enhancement index, compare the disturbance enhancement index with the set disturbance marker threshold point by point, identify the time points that are greater than the disturbance marker threshold as disturbance points, aggregate the disturbance points according to the time continuity condition, and establish a fusion disturbance prediction interval.

[0062] in, Indicates the first The perturbation enhancement index corresponding to time . Indicates the first The luminous derivative value at any given time. Indicates the first Mean of illumination derivative over the period, Indicates the first The standard deviation of the illumination derivative within the corresponding period at any given time. Represents a constant. Indicates the first Corresponding wind speed value at any time Display window Wind speed value inside, Number of windows;

[0063] The formula's calculation logic consists of two parts. First, the difference between the illumination derivative value and the mean illumination derivative within the corresponding period is divided by the sum of the standard deviation and a constant term within the period to form a standardized offset, expressing the relative intensity of the current illumination change relative to the regular changes within the period. Second, the ratio of the current wind speed value to the sum of wind speed values ​​within the analysis window is incremented by 1 to form a wind speed enhancement coefficient, used to enhance and modulate the illumination disturbance amplitude according to wind speed weights, reflecting the moderating effect of wind speed on the disturbance's cause. Multiplying these two parts yields the disturbance enhancement index. This value can be amplified by combining sudden changes in light intensity with sudden increases in wind speed, and can quantify the significance of disturbances on a unified scale.

[0064] The disturbance amplification index is a comprehensive indicator that combines changes in light intensity and wind speed to numerically characterize disturbances. It is calculated by the deviation of the light derivative and the wind speed weighting term to reflect whether there is a light disturbance at the current moment that is amplified by wind speed. The larger the value, the more significant the disturbance. Based on this, it can be determined whether the moment should be marked as a critical disturbance point.

[0065] Meaning of parameters and calculation process:

[0066] Indicates the first The illumination derivative value at any given time;

[0067] : No. The mean value of the illumination derivative for the current time period is 3.3. Assuming the current period is the 3rd period and the derivative values ​​within the period are {3.2, 3.5, 3.1, 3.4, 3.3}.

[0068] Indicates the first The standard deviation of the derivative within the period of a given time is expressed by the formula: calculate;

[0069] To represent a small constant and prevent the denominator from being zero, it is set to 0.01;

[0070] Indicates the first Wind speed value at any time;

[0071] Indicates the first in the analysis window The wind speed values ​​of each segment are summed and normalized.

[0072] Calculations are performed using the example parameters;

[0073] Table 1: Example Parameter Table

[0074]

[0075] Substituting into the formula, we get:

[0076] Light intensity derivative term:

[0077] ;

[0078] Wind speed is classified as one item:

[0079] ;

[0080] Substitute into the formula to calculate:

[0081] ;

[0082] The results show that the disturbance intensity index at time 5 is 3.199. If the disturbance intensity threshold is set to 2.5, the current time is marked as a significant disturbance point for use in region aggregation and prediction.

[0083] Table 2: Calculation Parameters for Disturbance Intensity Index

[0084]

[0085] As shown in Table 2, by quantifying multiple parameters at time 5 and substituting them into the calculation formula, the disturbance index value was successfully obtained, and the disturbance judgment at that time was completed.

[0086] The advantage of the formula is that by introducing two channels, illumination and wind speed, into the calculation, and comprehensively considering the disturbance judgment factor composed of illumination shift and wind speed increase at a given time point, the accuracy of disturbance identification is significantly improved. In addition to the original illumination derivative amplitude, a wind speed disturbance modulation term is added to enhance the sensitivity to the moment of rapid illumination change at high wind speed.

[0087] Please see Figure 3 The disturbance identification module includes:

[0088] The sequence change extraction submodule obtains the photovoltaic current output value change sequence and the energy storage current change sequence based on the fusion disturbance prediction interval. It divides the photovoltaic current output value change sequence and the energy storage current change sequence into a periodic dimension, and extracts the current increase and decrease direction of the photovoltaic current output value change sequence and the energy storage current change sequence at adjacent time points to obtain the periodic current change direction sequence.

[0089] The photovoltaic current output value variation sequence is represented as a time-sampled current value sequence. For example, the current is sampled from the photovoltaic inverter output at 10ms intervals, resulting in the sequence {4.2A, 4.4A, 4.7A, 4.3A, 4.1A}. Simultaneously, the energy storage current variation sequence can be obtained by sampling the charging and discharging current of the energy storage device at the same time intervals, for example, the sequence {−1.8A, −2.0A, −2.4A, −1.9A, −1.7A}. Both sets of current sequences are divided into periodic dimensions. Based on the load-side power consumption mode or grid-connected control strategy, the period duration is set to 50ms, and the sequence is divided into five periods, resulting in multiple current variations. Based on the periodic segments, the rate of change of the increasing and decreasing trends of photovoltaic current and energy storage current in each periodic segment is calculated. This is done by performing a difference operation on two sets of current data at adjacent time points. For example, after performing adjacent difference operations on the photovoltaic sequence, the sequence {+0.2, +0.3, −0.4, −0.2} is obtained, while the difference in energy storage current is {−0.2, −0.4, +0.5, +0.2}. The difference in photovoltaic current and the difference in energy storage current at the same time are compared directionally. If the signs are the same, they are marked as having the same direction; if the signs are opposite, they are marked as having opposite directions. In this process, a mapping sequence needs to be established for the directional comparison results to obtain the periodic current change direction sequence.

[0090] The direction consistency judgment submodule calls the periodic current change direction sequence, and judges the synchronization of the photovoltaic and energy storage current change directions according to the period. Periods with consistent directions within the same period are marked as same direction, and the remaining periods are marked as opposite direction. Combined with the timing information, the continuous same direction period segments are integrated to obtain the continuous same direction period interval set.

[0091] The photovoltaic current and energy storage current direction sequences are traversed according to the time segment order. For each period segment, the current period direction is set as the reference direction. For example, the direction of the first period is "consistent direction". In the process of judging the period, if the photovoltaic direction and the energy storage direction are both "consistent direction" in a certain period, the period direction is judged to be the same. If they are "inconsistent direction", it is recorded as a direction deviation. The comparison period continues to slide and the number of consecutive consistent direction periods is counted. The threshold for consecutive consistent period is set to 3, that is, the direction must be consistent for 3 consecutive periods to be recorded as a continuous period interval. In a specific example, if the direction sequence is {consistent, consistent, consistent, inconsistent, consistent, consistent}, the system identifies the first to third periods as a continuous consistent direction segment. The fourth period is skipped because it is inconsistent. The fifth to sixth periods form another continuous period interval. To ensure the rigor of the judgment process, a direction judgment threshold needs to be set. The direction judgment threshold is set to 3 periods according to experimental data, indicating that the system is considered valid only when the directionality is consistent within 3 periods. See Table 3 for the recognition of consistent direction periods under different threshold conditions.

[0092] Table 3: Comparison Table for Consistency Judgment of Continuous Directions

[0093]

[0094] As shown in Table 3, when the threshold period is set to 3, only segments with consistent periods of 3 consecutive periods can be identified. The threshold selection has a significant impact, and this period segment is used for fluctuation amplitude screening in subsequent processes.

[0095] The fluctuation amplitude screening submodule extracts the numerical change sequence of the grid-connected point current within the corresponding time period based on a set of continuous unidirectional periodic intervals, using the following formula:

[0096] ;

[0097] Calculate the fluctuation offset amplitude value, and based on the difference between the fluctuation offset amplitude value and the set disturbance identification benchmark fluctuation amplitude, perform a judgment operation on the continuous same-direction period interval to obtain the disturbance feature confirmation window;

[0098] in, Representing the The amplitude of fluctuation offset in each unidirectional periodic segment. Indicates the number of periodic segments in the same direction. Indicates the first Changes in photovoltaic current value within a period Indicates the first Changes in energy storage current value within each cycle Indicates the first The disturbance identification benchmark fluctuation amplitude corresponding to each period segment;

[0099] Formula calculation logic: For periodic segments Each sampling point The photovoltaic current change value at the sampling point With the change in energy storage current The summation is then squared, and the square root is taken. This is essentially equivalent to the absolute value of the summation of the photo-storage current at the sampling point. Subtract the benchmark fluctuation range of the current period from this value. This represents the distance of the sampling point from the standard disturbance reference. The absolute value of the above difference is taken to prevent positive and negative offsets from canceling each other out. Sum the absolute values ​​of the differences between the sampling points and divide by the total number of points. Calculate the average fluctuation deviation within the period and output the result. That is, the first The average perturbation offset intensity of the period;

[0100] The fluctuation deviation amplitude value is the average value of the difference between the actual current fluctuation and the reference disturbance amplitude after the photovoltaic current and the energy storage current are superimposed within a cycle. It reflects the degree of deviation between the cycle and the preset stable state. The larger the deviation, the more abnormal the current disturbance in the cycle. This value is obtained by averaging the fluctuation deviation values ​​of multiple sampling points within the cycle. The larger the value, the stronger the fluctuation, which is used for disturbance identification and judgment.

[0101] Meaning of parameters and calculation process:

[0102] This represents the total number of sampling points within the current period. For example, if sampling is performed at 10ms intervals within a 50ms period, then... ;

[0103] Indicates the first The photovoltaic current change value at each sampling point;

[0104] Indicates the first The change in energy storage current at each sampling point;

[0105] Indicates the first The disturbance identification benchmark fluctuation amplitude corresponding to the period segment, benchmark fluctuation amplitude The data was pre-sampled under stable operating conditions and set to 3.0A.

[0106] The following current sequence is selected as an example: , The formula is as follows: Sum and square each point individually, then take the square root and subtract it from the benchmark value.

[0107] ;

[0108] This result indicates that the current number The actual fluctuation offset amplitude of each period segment is 0.62A. The value is compared with the set disturbance identification tolerance. If this value exceeds the set disturbance judgment threshold... If the value is positive, it is identified as a fluctuation disturbance; otherwise, it is determined to be a normal periodic segment. In this example, since 0.62A > 0.5A, the current period is determined to be a disturbance period and participates in the disturbance identification window selection process.

[0109] The advantage of the formula is that by summing and squaring the photovoltaic current and the energy storage current point by point, and then comparing it with the difference of the reference offset, it can comprehensively reflect the changing trend of the actual disturbance intensity within the cycle, thereby enhancing the accuracy of the cycle disturbance judgment, especially in the case of small disturbances but with cumulative effects.

[0110] Please see Figure 4 The offset trend analysis module includes:

[0111] The data stream receiving submodule calls the disturbance feature confirmation window to obtain the photovoltaic output current time series and the energy storage output current time series. Based on the internally set time step parameters, it performs unified time axis alignment processing on the photovoltaic output current time series and the energy storage output current time series to generate a photovoltaic-storage synchronization current time segment.

[0112] To ensure consistent comparison and processing of the two sets of data, the time series of output current generated by the photovoltaic module and the energy storage device are read. To achieve this, a unified time axis alignment operation is required according to the system's internal time step. The step size can be 1 second, 5 seconds, or 10 seconds, depending on the system's real-time requirements. Assuming the photovoltaic current sampling frequency is 1 second, and the data is sequentially collected from 00:00:00 at 5.1A, 5.0A, 5.2A, 5.4A, etc., while the energy storage current data is collected at different time intervals, there will be a misalignment between the two sets of data at different time points. The system uses linear interpolation or nearest-value imputation to process the data. For example, if the energy storage current data is missing at 00:00:01, the estimated value at that time is calculated by using data from the preceding and following time points. A set of photovoltaic-storage current correspondences is formed under a unified time series. In actual scenarios, data from different devices can be collected uniformly by PLC acquisition devices or data concentrators. The system filters valid data based on timestamps and combines them to construct a time segment data set. Each time segment includes two values: photovoltaic side current and energy storage side current. At the same time, the original time tag is retained to generate a photovoltaic-storage synchronous current time segment.

[0113] The difference extraction submodule calculates the difference sequence between the photovoltaic current value and the energy storage current value at time points based on the photovoltaic current value and the energy storage current value in the photovoltaic-storage synchronous current time segment, constructs a curve-shaped data structure with time as the horizontal axis and difference as the vertical axis, and generates a photovoltaic-storage current difference trend vector.

[0114] The numerical difference between two current values ​​needs to be compared point by point. For each time segment, the photovoltaic output current value and the energy storage output current value are extracted and directly subtracted to obtain the difference value between them. The difference value represents the relative energy supply difference of the photovoltaic and energy storage system at a given time point. The difference values ​​at each time point are combined sequentially to form a new difference sequence. For example, if the photovoltaic current is 5.0A, 5.1A, and 5.3A and the energy storage current is 3.0A, 3.2A, and 3.4A in a certain time period, the three sets of differences are 2.0A, 1.9A, and 1.9A respectively. This sequence is organized into a time-difference pair structure for processing. In actual deployment, the difference data structure can be stored in a cache queue or memory in the form of key-value pairs or list arrays, or it can be directly written to the database for trend analysis. During the data structure construction process, the direction and magnitude of the difference change can also be recorded in parallel to preliminarily detect whether the equipment is operating in a balanced state and generate a photovoltaic-energy storage current difference trend vector.

[0115] The trend determination submodule calls the photovoltaic-storage current difference trend vector, combines the sign and magnitude of the slope of the difference change within a continuous time period, extracts the change direction of the real-time difference trend, and matches the change direction value with the set power supply offset start judgment benchmark value for difference direction judgment. If the trend direction of multiple consecutive points is consistent and the magnitude exceeds the judgment benchmark value, it is determined that the power supply offset start is established, and the photovoltaic-storage offset scheduling trigger point is obtained.

[0116] The direction and degree of change of the difference over a continuous time period need to be analyzed. During the analysis, the system compares the difference between adjacent time points to see if it shows a continuous increase, decrease, or remains stable, and determines whether the direction of change is consistent. For example, if the difference sequence is 2.1A, 1.9A, 1.7A, and 1.5A, it can be initially judged as a continuous downward trend. The system then determines whether the magnitude of each change exceeds the preset change threshold. The change threshold can be set based on the operating data, such as by extracting the average difference change over a certain period and then setting the threshold by a multiple. If the direction of change is the same for multiple consecutive points and the magnitude of change is greater than the threshold, it can be considered that the current system operation has a relatively obvious power supply offset characteristic, and it is determined to be an offset start state. In practical applications, it is deployed inside the scheduling control and the trend vector is called for real-time monitoring. Once the judgment condition is met, the system outputs an offset start flag signal, records the current time point, and uses it as a trigger signal for adjusting the scheduling control strategy. Based on this, the energy storage discharge strategy can be readjusted to adapt to the operating requirements of the offset state, ensuring that the scheduling can react quickly according to the current state and obtain the photovoltaic-storage offset scheduling trigger point.

[0117] Please see Figure 5 The path switching module includes:

[0118] The channel information extraction submodule obtains the real-time channel working status identifier of the power converter and the remaining energy storage power monitoring value based on the photovoltaic-storage offset scheduling trigger point, calculates the remaining available time of energy storage under real-time discharge power conditions, and simultaneously obtains the time series of photovoltaic module output power within the time period. Based on the time series change trend, it extracts the direction of photovoltaic power change and generates a set of channel operation status information.

[0119] The system reads the current channel operating status identifier information of the power converter. The channel status includes the working channel number and whether the current channel is in main power supply, standby, or standby state. Status bit data is obtained through the communication protocol of the internal device, such as Modbus or CAN bus interface. At the same time, the real-time remaining power value of the energy storage device is collected. The real-time remaining power value can be comprehensively calculated from the voltage, current, and SOC (state of charge percentage) data provided by the energy storage management device BMS. Based on the known current discharge power level of the energy storage and the remaining power information, the system calculates the duration for which the energy storage can continue to discharge and maintain power supply while keeping the current discharge rate unchanged. This duration is of reference value for judging the stability of power supply. After completing the calculation of the available energy storage time, the system synchronously reads the output power time series of the photovoltaic module. For example, by continuously sampling the photovoltaic inverter data, the photovoltaic power is organized into a sequence data according to the time point. According to the continuous change trend of the power value in the sequence, the current photovoltaic power is identified as rising, falling, or oscillating. The photovoltaic trend direction is combined and encapsulated with the available energy storage time, channel status identifier, and other information to generate a set of channel operation status information.

[0120] The path switching condition judgment submodule calls the channel operation status information set, extracts the remaining available time of energy storage and the direction of photovoltaic power change, compares the available time with the path switching continuous benchmark value, and determines whether it has the ability to maintain the main channel power supply. At the same time, it detects whether the photovoltaic power trend is in a continuous downward state. If both conditions are met, it is determined that the path switching requirements are met in real time, and the path switching status judgment result is obtained.

[0121] The system extracts the remaining available time of energy storage and the trend direction of photovoltaic power change. The remaining available time is expressed in minutes or hours, representing how long the energy storage device can maintain power supply under the current discharge state. The value is compared with the system's preset path switching duration benchmark value. The benchmark value can be determined based on operating parameters such as load level, response time, and backup power access duration. If the available time is higher than the benchmark value, the system considers that the energy storage has sufficient energy to support subsequent power supply. At the same time, the trend direction of photovoltaic power is also judged. If the photovoltaic power shows a downward trend in multiple consecutive sampling periods, the system determines that the photovoltaic no longer meets the conditions to continuously serve as the main power supply path. At this time, the two core conditions in the judgment condition, namely sufficient available time of energy storage and continuous decline of photovoltaic power, are both met, and the real-time conditions for path switching are considered to be met. The judgment result is output to the module as a path switching status identifier. In specific applications, this judgment process is generally deployed in energy management equipment or scheduling platform, and works with edge controllers to realize the judgment logic of fast response channel switching and obtain the path switching status judgment result.

[0122] The instruction control submodule determines the path switching status based on the results. If the status meets the conditions, it constructs a control signal structure and sets the energy storage path as the main channel identifier and the photovoltaic path as the auxiliary channel identifier. It then uploads the control signal to the edge controller control terminal through the data interface and generates a path master-slave switching control instruction.

[0123] The system determines whether the switching conditions are met. If the status confirms that the channel switching conditions are met, the system immediately constructs a set of control signal structures. The structure contains multiple fields, including channel role identifier, control type, priority, execution timestamp, etc. The energy storage path is marked as the main channel and the photovoltaic path as the auxiliary channel in the structure. The settings reflect the current system's strategy of prioritizing energy storage power supply. After the signal structure is constructed, it is uploaded through the data interface connected to the edge controller. The interface protocol is Modbus TCP / IP or IEC standard communication. The interface service module is called to transmit the control signals to the execution port of the edge controller in real time. After receiving the control command, the edge controller will execute the path role switching operation according to the predetermined logic. For example, it will disconnect the channel connected to the photovoltaic as the main power supply path and connect the channel connected to the energy storage to the main power supply line. At the same time, it will notify the monitoring to update the channel status. During the execution process, the control behavior is recorded in the log and a path master-slave switching control command is generated.

[0124] Please see Figure 6 The frequency band suppression module includes:

[0125] The complex impedance spectrum construction submodule calls the path master-slave switching control command to obtain real-time three-phase voltage and current sampling data of the grid connection point. It performs complex domain amplitude and phase angle calculations according to the set period, constructs a complex impedance response value sequence for the time segment, and forms a response spectrum curve according to the frequency domain distribution. It extracts the position of the main frequency of the complex impedance modulus in the response spectrum curve, uses the main frequency as the center to detect the width of the frequency domain energy drop interval, and generates a frequency drop bandwidth parameter set.

[0126] The system retrieves real-time sampling data of three-phase voltage and current at the grid connection point. This data is collected by high-frequency sampling equipment, such as a three-phase power meter or a multi-channel sampling module. The voltage and current waveforms are recorded periodically within each set period. After preprocessing, this sampling data needs to undergo complex domain calculations. Specifically, the amplitude and phase angle information of the voltage and current signals in each time segment are extracted to represent the complex form of the voltage and current. Through time-segment analysis, a sequence of complex impedance response values ​​is formed. Each complex impedance value is calculated from the complex values ​​of the voltage and current at each time point and combined in chronological order. The data of each time segment forms a complex impedance response spectrum. The complex impedance response shows the change of complex impedance at each time point. The system identifies the frequency position where the complex impedance magnitude peaks from the response spectrum curve, sets the main frequency as the core center frequency point of the frequency domain detection, and then extends to both sides to find the boundary position of the interval where the frequency domain energy continuously decreases, forming a set of bandwidth parameters to represent the frequency domain width covered by the energy decrease, generating a frequency decrease bandwidth parameter set.

[0127] The spectrum principal component identification submodule obtains the time-series data of active and reactive power output of energy storage within the real-time period through the frequency drop bandwidth parameter group, performs Fourier transform operation to obtain the spectrum energy distribution map, identifies the main energy concentration frequency band in the spectrum energy distribution map, and performs frequency interval matching with the frequency drop bandwidth parameter group to determine whether the main energy component is within the drop bandwidth and generates the main frequency energy matching status value.

[0128] The system reads the output power data of the energy storage device within a set real-time period, including time-series data of active and reactive power. The collected data is recorded with high time resolution, and the energy management platform can track the energy storage output at the second level. The system inputs the time-series data into the spectrum analysis process, performs discrete Fourier transform processing, and maps the power data in the time domain to the frequency domain to identify the distribution of power energy at different frequency points, obtaining a power spectrum energy distribution map. After the power spectrum energy distribution map is formed, it will show the degree of energy concentration at each frequency point. Based on this, the system locates the core frequency region of the main energy distribution, identifies the main energy frequency band by searching for the most significant concentrated frequency band in the frequency domain, and then compares and matches the frequency range covered by the frequency band with the aforementioned frequency drop bandwidth parameter group item by item to determine whether the main energy in the current power output is completely within the energy drop range. If the match is successful, the main frequency energy matching status is recorded as a valid status, which will be used to determine whether to enter the next stage of modulation control processing and generate the main frequency energy matching status value.

[0129] The frequency energy suppression submodule matches the main frequency energy status value. If the status determines that the main frequency energy is within the falling bandwidth, it calls the original control signal field, performs compression adjustment on the modulation time parameter, adjusts the modulation window length and amplitude factor according to the bandwidth range, repackages the control field and completes the energy density control configuration, and generates the frequency band power injection compression result.

[0130] If the current identification results indicate that the main energy frequency band is indeed within the previously identified frequency drop bandwidth range, the system initiates the control signal field invocation process to adjust the modulation parameters contained in the original control signal. Specifically, this involves reading the modulation time parameter, which represents the time window range used by the current system for signal conditioning and control response; performing compression processing based on the specific frequency range of the frequency band drop range, i.e., shortening the duration of the original modulation time window; and simultaneously adjusting the modulation amplitude factor based on the bandwidth position to reduce or reset the influence range of amplitude adjustment, making the modulation behavior more concentrated in the area involved in the main frequency change. After completing the parameter compression and reset, the system re-encapsulates the control field, generating a set of updated energy conditioning instruction structures, representing the output form of the new control strategy. This can be transmitted to the execution layer via an interface, where the injection conditioning task is actually executed on the controller or power conversion device, generating the frequency band power injection compression result.

[0131] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A multi-source data fusion photovoltaic power scheduling control system, characterized in that, The system comprises: The perception extraction module obtains the light value sequence and the wind speed value sequence, constructs a time derivative sequence according to the adjacent time point light value difference, calls the deviation degree between the time derivative sequence and the continuous period average derivative, judges whether the wind speed continuous period trend change direction is consistent, generates a fusion disturbance pre-judgment interval; The disturbance identification module calls the fusion disturbance pre-judgment interval, obtains the photovoltaic current output value change sequence and the energy storage current change sequence, performs the same period current increase and decrease direction consistency judgment, analyzes the total current fluctuation amplitude in the time period, and generates a disturbance feature confirmation window; The offset trend analysis module calls the disturbance feature confirmation window, calculates the difference between the photovoltaic current change amount and the energy storage current change amount at the same time point, judges whether the real-time energy supply offset degree meets the scheduling start condition, and obtains a photovoltaic and energy storage offset scheduling trigger point; The path switching module obtains the power converter channel state and the energy storage remaining available time length based on the photovoltaic and energy storage offset scheduling trigger point, judges whether the power supply path switching meets the master channel switching condition in combination with the real-time photovoltaic output trend, and generates a path master-slave switching control instruction.

2. The multi-source data fused photovoltaic power dispatching control system according to claim 1, characterized in that, The fusion disturbance pre-judgment interval includes a disturbance starting time, a disturbance ending time, a disturbance amplitude judgment basis, and a disturbance trend consistency label, the disturbance feature confirmation window includes a current change consistency label, a total current fluctuation amplitude threshold type, and a disturbance period identification label, the photovoltaic and energy storage offset scheduling trigger point includes an offset degree value, an offset trend direction, and a scheduling trigger threshold state, and the path master-slave switching control instruction includes an energy storage channel identifier, a photovoltaic channel auxiliary state, and a control signal instruction type.

3. The multi-source data fused photovoltaic power dispatch control system of claim 1, wherein, The perception extraction module comprises: The light value calculation submodule obtains the light value sequence and the wind speed value sequence, calculates the difference between the light values of adjacent time points item by item, establishes a light change derivative sequence under equal time intervals, divides the derivative sequence into multiple period segments according to a set time window, respectively solves the derivative value mean in the period segment, calls the deviation amplitude of the period derivative mean and the time point derivative for difference calculation, and obtains the light derivative offset amount; The wind speed trend analysis submodule calls the light derivative offset amount, obtains the wind speed value sequence of the time period corresponding to the light derivative offset amount, divides the wind speed value sequence into continuous period intervals, calculates the difference between the wind speed value means in the period interval, judges the positive and negative signs of the wind speed mean change between the periods, identifies the continuous change trend direction, and obtains a wind speed trend consistency label sequence; The disturbance interval marking submodule calculates a disturbance enhancement index according to the wind speed trend consistency label sequence, compares the disturbance enhancement index with a set disturbance label threshold point by point in sequence, identifies the time point greater than the disturbance label threshold as a disturbance point, aggregates the disturbance points according to the time continuity condition, and establishes a fusion disturbance pre-judgment interval.

4. The multi-source data fused photovoltaic power dispatching control system according to claim 3, characterized in that, The disturbance identification module comprises: The sequence change extraction submodule obtains a photovoltaic current output value change sequence and an energy storage current change sequence based on the fusion disturbance pre-judgment interval, divides the photovoltaic current output value change sequence and the energy storage current change sequence in a cycle dimension, and extracts current increase and decrease change directions of the photovoltaic current output value change sequence and the energy storage current change sequence at adjacent time points to obtain a cycle current change direction sequence; The direction consistency judgment submodule calls the cycle current change direction sequence, judges the synchronization of the photovoltaic and energy storage current change directions according to a cycle, marks the cycles with the same direction in the same cycle as homodirectional, marks the remaining cycles as heterodirectional, integrates continuous homodirectional cycle segments in combination with time sequence information to obtain a continuous homodirectional cycle interval set, and extracts a numerical value change sequence of the grid-connected point current in the corresponding time period according to the continuous homodirectional cycle interval set, calculates a fluctuation offset amplitude value, performs a determination operation on the continuous homodirectional cycle interval according to the difference between the fluctuation offset amplitude value and a set disturbance identification reference fluctuation amplitude, and obtains a disturbance feature confirmation window. The fluctuation offset amplitude value is calculated by the following formula:

5. The multi-source data fused photovoltaic power dispatching control system according to claim 4, characterized in that, The offset trend analysis module includes: ； wherein, represents the wave offset amplitude value of the first periodic section, represents the number of periodic sections, represents the change of the photovoltaic current value in the first period, represents the change of the energy storage current value in the first period, represents the disturbance identification reference wave amplitude corresponding to the first periodic section.

6. The multi-source data fused photovoltaic power dispatch control system of claim 4, wherein, The data stream receiving submodule calls the disturbance feature confirmation window, obtains a photovoltaic output current time sequence and an energy storage output current time sequence, performs unified time axis alignment processing on the photovoltaic output current time sequence and the energy storage output current time sequence according to an internally set time step parameter, and generates a photovoltaic and energy storage synchronous current time segment; The difference amount extraction submodule calculates a difference sequence of photovoltaic current values and energy storage current values at time points according to the photovoltaic current values and the energy storage current values in the photovoltaic and energy storage synchronous current time segment, constructs a curve form data structure with time as the horizontal axis and the difference amount as the vertical axis, and generates a photovoltaic and energy storage current difference trend vector; The trend judgment submodule calls the photovoltaic and energy storage current difference trend vector, extracts the change direction of the real-time difference trend in combination with the sign and amplitude of the difference amount change slope in a continuous time period, and performs difference amount direction matching judgment on the change direction value and a set energy supply offset start judgment reference value. If the trend directions of continuous multiple points are consistent and the amplitudes exceed the judgment reference value, it is determined that the energy supply offset start state is established, and a photovoltaic and energy storage offset scheduling trigger point is obtained. The path switching module includes:

7. The multi-source data fused photovoltaic power dispatch control system of claim 6, wherein, The channel information extraction submodule obtains real-time channel working state identifiers of a power converter and energy storage residual capacity monitoring values based on the photovoltaic and energy storage offset scheduling trigger point, calculates a residual available duration of the energy storage under the condition of real-time discharge power, synchronously obtains a photovoltaic module output power time sequence in a time period, extracts a photovoltaic power change direction according to the time sequence change trend, and generates a channel operation situation information set. ​ The path switching condition judgment submodule calls the channel running situation information set, respectively extracts the remaining available duration value and the photovoltaic power change direction value, compares the available duration with the path switching duration reference value, judges whether the main channel power supply capability is maintained, and simultaneously detects whether the photovoltaic power trend is in a continuous decline state, and if both conditions are met, it is determined that the real-time path switching requirement is met, and a path switching state judgment result is obtained; The instruction control submodule constructs a control signal structure according to the path switching state judgment result, sets the energy storage path as the main channel identifier and the photovoltaic path as the auxiliary channel identifier, uploads the control signal to the edge controller control end through the data interface, and generates a path master-slave switching control instruction.

8. The multi-source data fused photovoltaic power dispatch control system of claim 1, wherein, The system further comprises a frequency band suppression module: The frequency band suppression module calls the path master-slave switching control instruction, obtains real-time grid point voltage and current data, constructs a complex impedance response spectrum in a period, extracts the main frequency position and identifies the frequency response bandwidth, obtains real-time period energy storage output power data and performs Fourier spectrum decomposition, if the spectrum main energy component is within the frequency drop bandwidth, adjusts the modulation time parameter in the control signal, suppresses the frequency band output energy density, and generates a frequency band power injection compression result; The frequency band power injection compression result includes the spectrum main component frequency band, the energy density compression amplitude, and the modulation time adjustment parameter.

9. The multi-source data fused photovoltaic power dispatch control system of claim 8, wherein, The frequency band suppression module comprises: The complex impedance spectrum construction submodule calls the path master-slave switching control instruction, obtains real-time three-phase voltage and current sampling data of the grid point, calculates the complex domain amplitude and phase angle according to the set period, constructs a complex impedance response value sequence of the time segment, and forms a response spectrum curve according to the frequency domain distribution, extracts the main frequency corresponding position of the complex impedance modulus in the response spectrum curve, detects the frequency energy drop interval width using the main frequency as the center, and generates a frequency drop bandwidth parameter group; The spectrum main component identification submodule obtains active power and reactive power time series data of energy storage output in a real-time period through the frequency drop bandwidth parameter group, performs Fourier transform operation to obtain a frequency spectrum energy distribution map, identifies the main energy concentrated frequency band in the frequency spectrum energy distribution map, and matches the frequency interval with the frequency drop bandwidth parameter group, judges whether the main energy component is within the drop bandwidth, and generates a main frequency energy matching state value; The frequency energy suppression submodule adjusts the modulation time parameter according to the main frequency energy matching state value, if the state is determined to be within the drop bandwidth, performs compression adjustment processing on the original control signal field, adjusts the modulation window length and amplitude factor according to the bandwidth interval, re-encapsulates the control field and completes the energy density control configuration, and generates a frequency band power injection compression result.

Images