A source-grid-load-storage distributed collaborative scheduling system considering wind energy and solar energy uncertainty

Abstract

The present application relates to power supply and distribution circuit device technical field, disclose a kind of source network load storage distribution collaborative scheduling system considering wind energy solar uncertainty, including power regulation subsystem and ripple feature detection, impedance distribution correction, dynamic response filter and collaborative scheduling control module, ripple feature detection module extracts high-frequency current ripple and determines branch mutual conductance parameter, impedance distribution correction module updates sensitivity matrix impedance weight coefficient accordingly, dynamic response filter module executes delay filter according to response time constant to adjustment instruction, collaborative scheduling control module is based on the adjustment physical phase angle control parameter offset of matrix after correction and filter result, induces node power flow automatic change direction, the present application realizes scheduling model and topology self-calibration by ripple inversion, using phase angle offset mechanism makes each node present collaborative relay characteristics at capacity boundary, effectively suppresses adjustment speed mismatch problem, improves operating steady state.

Description

Technical Field

[0001] This invention relates to a source-grid-load-storage distributed collaborative scheduling system that takes into account the uncertainties of wind and solar energy, belonging to the technical field of power supply and distribution circuit devices. Background Technology

[0002] Currently, the power balance regulation of various components on the source side, grid side, load side, and energy storage side in the distribution network typically uses a central station's predictive algorithm to generate dispatch commands. This method assumes that power fluctuations are statistically stationary within a preset period and maintains the stability of bus voltage and system frequency by calculating the dispatch target values ​​for active and reactive power. As the proportion of distributed energy access increases, power fluctuations exhibit wide-frequency step characteristics. In low-inertia networks lacking synchronous machine support, the timescale of voltage reduction caused by power drops is at the cycle level. The central dispatch mode is limited by physical channel bandwidth and computing power convergence speed, resulting in regulation delay. By the time the regulation command reaches the execution mechanism, the system state has already shifted. To cope with such state shifts, the system usually needs to be configured with redundant energy storage capacity.

[0003] While increasing the complexity of prediction algorithms or building high-bandwidth communication networks can shorten the response cycle, prediction accuracy is limited by meteorological characteristics, and communication links pose a risk of physical delay. Simply increasing hardware investment cannot solve the structural contradiction of distributed nodes lacking local physical self-healing capabilities. At the hardware level, physical self-healing capabilities are insufficient, and existing control and monitoring algorithms struggle to balance real-time performance with topology adaptability. For example, Chinese invention patent CN117791853A discloses a fault monitoring method for the end coefficient matrix of a distribution network based on high-frequency power flow. It uses high-frequency power flow information to construct a Laplace matrix to improve the sensitivity of end fault monitoring, focusing on post-event diagnosis rather than pre-event scheduling. It cannot perform active admittance compensation for fluctuations in wind and solar power output. The construction of the coefficient matrix relies on preset model parameters of cable characteristic impedance and does not consider the dynamic drift of line impedance caused by environmental thermal effects and contact resistance aging. This leads to a mismatch between the scheduling model and the actual physical topology. Existing technologies lack consideration for the differences in response time constants of heterogeneous energy nodes, and when dealing with extreme disturbances, power oscillations are easily induced by speed mismatch.

[0004] Therefore, how to construct a mechanism that enables each node of the source, grid, load, and storage to generate a power balance closed loop based on the deviation of the local bus physical state, so as to eliminate the risk of system instability caused by scheduling decisions lagging behind physical phenomena, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy, the system comprising a power regulation subsystem, a ripple characteristic detection module, an impedance distribution correction module, a dynamic response filtering module, and a collaborative scheduling control module: The power conditioning subsystem is deployed at the bus nodes of the power distribution network. Each power conditioning subsystem includes a bidirectional power conversion module and a local control module. The ripple feature detection module is configured to extract the high-frequency current ripple generated by the bidirectional power conversion module during the switching operation, and determine the branch mutual conduction parameters between the bus nodes based on the attenuation of the transmission amplitude of the high-frequency current ripple between two adjacent bus nodes. The impedance distribution correction module is configured to acquire branch mutual conductance parameters and update the impedance weight coefficients in the preset sensitivity mapping matrix accordingly to generate a corrected sensitivity matrix that reflects the actual physical impedance distribution of the distribution network. The dynamic response filtering module is configured to acquire the response time constants of the energy storage converter and the thermally inert load, and perform hysteresis filtering on the power regulation command of the local control module according to the response time constant, so as to transfer the transient power regulation from the fast response node to the slow response node in the time domain. The collaborative scheduling control module is configured to, based on the correction sensitivity matrix and the filtering results of the dynamic response filtering module, change the physical phase angle control parameters of the bidirectional power conversion module through the local control module to generate an offset. It utilizes the automatic reversal of power flow direction at the voltage boundary of the bus node to form a closed-loop voltage regulation between the power regulation subsystems. When changing the physical phase angle control parameters, the collaborative scheduling control module uses the deviation between the measured voltage of the bus node and the preset voltage threshold as the feedback control gradient to dynamically adjust the output equivalent admittance of the bidirectional power conversion module, so as to maintain voltage steady state of the distribution network under the condition of distributed power output fluctuation.

[0006] Preferably, the impedance distribution correction module follows the following calculation rules when updating the impedance weighting coefficients: ,in, For the corrected bus node With bus node Sensitivity weighting coefficients between them The preset initial weighting coefficients, bus node The initial amplitude of the high-frequency current ripple emitted at the location. bus node The remaining amplitude of the high-frequency current ripple received at the point.

[0007] Preferably, the system also includes an operation status monitoring module, which is configured to extract the broadband ripple residual component from the output current of the bidirectional power conversion module, determine the physical fault when the broadband ripple residual component exhibits asymmetric harmonic characteristics, and output a blocking command to the collaborative scheduling control module to shut down the dynamic response filtering module.

[0008] Preferably, after determining that a physical fault has occurred, the operation status monitoring module performs physical isolation of the fault point by adjusting the reactive power output current of the bidirectional power conversion module and sends a lockout command to the collaborative scheduling control module.

[0009] Preferably, the sampling frequency of the ripple feature detection module is synchronized with the switching frequency of the bidirectional power conversion module, and the ripple feature detection module filters out the 50Hz fundamental component in the bus node through a filter circuit.

[0010] Preferably, the dynamic response filtering module configures the cutoff frequency of the controlled low-pass filter for the bus node where the energy storage converter is located based on the response time of the energy storage converter and the response time of the thermally inert load, so as to ensure that each power regulation subsystem exhibits a uniform power regulation synchronization characteristic.

[0011] Preferably, the system further includes a disturbance injection module, which is configured to inject physical disturbance signals into the bidirectional power conversion module and verify the calculation accuracy of the sensitivity mapping matrix by detecting the voltage response gradient of the bus node to the physical disturbance signals.

[0012] Preferably, the collaborative scheduling control module enables the power regulation subsystem to update model parameters without introducing external probe links by offsetting the physical phase angle control parameters, and allocates reactive power compensation between each bus node according to the updated impedance weight coefficient.

[0013] Preferably, each power regulation subsystem also includes an energy storage status acquisition submodule. When the collaborative scheduling control module calculates the offset of the physical phase angle control parameters, it uses the state of charge value output by the energy storage status acquisition submodule as the capacity constraint condition for the power regulation subsystem to participate in voltage regulation.

[0014] Preferably, the collaborative scheduling control module locates the voltage over-limit risk node based on the modified sensitivity matrix, and determines the execution order of each power regulation subsystem participating in the collaborative regulation according to the impedance weighting coefficient, and issues the offset command of the physical phase angle control parameter in sequence.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the distributed coordinated scheduling of power generation, grid, load and storage, by establishing a sensitivity mapping matrix that reflects the correlation between node power disturbance and voltage vector, and transforming it into the physical adjustment weight of distributed execution units, the direct mapping of power uncertainty to node equivalent admittance compensation is realized. Under this mechanism, the access impedance of the energy storage side and the load side is adjusted in milliseconds with the micro changes of bus voltage and frequency. The external scheduling logic that originally relied on wide-area communication and prediction algorithms is internalized into the endogenous response of the controlled admittance unit. This instantaneous compensation mechanism at the physical level not only eliminates the power gap caused by prediction error, but also achieves spontaneous power redistribution among nodes by utilizing the admittance distribution characteristics of the circuit system itself, thereby improving the anti-interference recovery capability of the distribution network against broadband disturbances.

[0016] 2. By utilizing the impedance vector redirection mechanism of controlled nodes when the regulation capacity approaches the physical threshold, the common capacity saturation impact problem in distributed regulation is solved. When the regulation space of energy storage devices or flexible loads is limited, the redirection module adjusts the impedance phase angle of the local interface and uses the guiding effect of impedance on current in Kirchhoff's current law to actively transfer the excess power flow to adjacent nodes with regulation margin. This method does not rely on central computing power to redistribute tasks, but achieves automatic reversal of power flow through the micro-offset of physical phase angle, so that each node in the system exhibits flexible relay characteristics at the capacity boundary, effectively avoiding the risk of voltage over-limit caused by local regulation failure.

[0017] 3. By extracting and utilizing the high-frequency switching ripple signal associated with the inverter, a self-calibrating closed loop that aligns the algorithm model with the physical topology in real time is constructed. By utilizing the attenuation characteristics of the ripple signal during branch transmission, the system can invert the actual coupling strength between each physical node in real time and correct the weight coefficients in the sensitivity mapping matrix accordingly. This enables the perception of changes in the physical environment such as line impedance drift and contact resistance aging, ensuring that the scheduling logic is always anchored to the real power grid topology. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the closed-loop regulation principle and signal interaction of the present invention. Figure 2 This is a schematic diagram showing the overall functional architecture decomposition and module composition of the present invention. Detailed Implementation

[0019] The following disclosure is for illustrative purposes only and is not intended to limit the scope of protection of the present invention.

[0020] This invention provides a distributed grid-load-storage collaborative dispatch system considering the uncertainties of wind and solar energy. It includes a power regulation subsystem, a ripple characteristic detection module, an impedance distribution correction module, a dynamic response filtering module, and a collaborative dispatch control module. The power regulation subsystem is deployed at the bus nodes of the distribution network, and each subsystem includes a bidirectional power conversion module and a local control module. To address the issue of distribution network physical topology parameters drifting with the operating environment due to distributed power source integration, the ripple characteristic detection module extracts the current ripple generated by the bidirectional power conversion module during switching operations. The sampling frequency of the ripple characteristic detection module is synchronized with the switching frequency of the bidirectional power conversion module, and the ripple in the bus nodes is filtered out by a filtering circuit. Hz fundamental component, ripple characteristic detection module measures bus nodes Initial amplitude of the emitted current ripple and bus nodes The remaining amplitude received at the location The impedance distribution correction module obtains the branch mutual conductance parameters and updates the impedance weighting coefficients in the preset sensitivity mapping matrix to generate a corrected sensitivity matrix. The impedance distribution correction module follows the following calculation rules when updating the weighting coefficients: ;in, For the corrected bus node With bus node Sensitivity weighting coefficients between components, in dimensionless units. These are the preset initial weighting coefficients, in dimensionless units. bus node The initial amplitude of the high-frequency current ripple emitted at the location, in amperes (A). bus node The received high-frequency current ripple residual amplitude, in amperes (A), is obtained by acquiring the bus node voltage amplitude. With phase angle Establish the nodal power flow balance equation in polar coordinate form and construct the sensitivity mapping matrix. Jacobian matrix The initial values ​​of each element are set based on the reference power flow parameters of the IEEE 33-bus test system, and the impedance distribution correction module obtains the high-frequency ripple amplitude ratio. Then, the impedance weighting calibration procedure is executed to measure the changes in the mutual conductance of the branches. Mapping to sensitivity weighting coefficients If the calculated voltage sensitivity gradient exceeds the preset threshold pu / MW triggers dynamic recombination of non-zero elements in the Jacobian matrix, resulting in the corrected sensitivity matrix. Guide the collaborative scheduling control module to generate physical phase angle offset instructions This ensures that the execution gradient of dispatch instructions is anchored to the real-time physical impedance distribution of the distribution network. Through this correction procedure, the system maps the amplitude attenuation during the transmission process of physical branches to the dispatch model in real time, so that the model parameters are consistent with the actual physical impedance distribution of the distribution network.

[0021] To address the computational bias in the sensitivity mapping matrix caused by uncertainties in physical parameters, the perturbation injection module executes an online calibration procedure based on impedance spectrum sweeping, controlling the bidirectional power conversion module to inject an amplitude of the rated voltage at the bus node. % and frequency is The Hz sinusoidal voltage perturbation signal and the ripple characteristic detection module synchronously acquire the current response vector of each bus node. The collaborative scheduling control module determines the measured increment of the transmission impedance based on the voltage response gradient between each node. And the measured increment is obtained by using the least squares fitting algorithm. Feedback is sent to the impedance distribution correction module to adjust the impedance weight coefficients in the sensitivity mapping matrix, ensuring that the execution gradient of the dispatch command is always anchored to the real-time physical impedance distribution of the distribution network. The measured increment of transmission impedance is expressed in units of 1 / 2 Ω. To address the power oscillation problem caused by the speed mismatch between the energy storage converter and the thermally inertial load, the dynamic response filtering module obtains the first response time constant of the energy storage converter. The second response time constant of thermal inert load The dynamic response filtering module determines the cutoff frequency of the controlled low-pass filter based on the response time constant. and utilize this cutoff frequency The slope of the power command adjustment from the local control module is constrained. When the system senses power output fluctuations, the energy storage nodes with faster response times prioritize transient regulation. The dynamic response filtering module uses the cutoff frequency... By performing hysteresis filtering, the regulation task is transferred to thermally inert load nodes with slower response speeds in the time domain. This synchronization procedure enables heterogeneous energy nodes to exhibit consistent synchronization characteristics during dynamic regulation, thereby improving the steady-state level of the bus voltage.

[0022] To address the risk of voltage exceedance caused by capacity saturation of local regulation units, the collaborative scheduling control module, based on the correction sensitivity matrix and the filtering results of the dynamic response filtering module, adjusts the physical phase angle control parameters of the bidirectional power conversion module through the local control module. The collaborative scheduling control module also acquires the state of charge (SOC) values ​​output by the energy storage SOC acquisition submodule. This is used as a capacity constraint condition for voltage regulation, when the state of charge value of the bus node is... Reaching the preset percentage threshold, either upper or lower, i.e. or At that time, the collaborative scheduling control module locates the voltage over-limit risk node based on the modified sensitivity matrix, determines the execution order according to the impedance weighting coefficient, and issues a physical phase angle offset command, thereby changing the phase angle of the output voltage vector of the bidirectional power conversion module. The power flow direction of the inducing bus node to automatically switch at the voltage boundary is induced. The collaborative scheduling control module uses the deviation between the measured voltage of the bus node and the preset voltage threshold as the feedback control gradient to dynamically adjust the output equivalent admittance of the bidirectional power conversion module.

[0023] The system also includes an operation status monitoring module, which extracts the broadband ripple residual component from the output current of the bidirectional power conversion module. When the broadband ripple residual component It exhibits asymmetric harmonic characteristics and its amplitude exceeds the preset fault current threshold. When a physical fault is detected, the operation status monitoring module sends a lockout command to the coordinated scheduling control module to shut down the dynamic response filter module. Simultaneously, it adjusts the reactive power output current of the bidirectional power conversion module to physically isolate the fault point. Then, it performs current spectrum analysis based on sliding window Fourier transform (SDFT) to calculate... Sum of the effective values ​​of odd harmonics within a ms period Sum of the effective values ​​of even harmonics Define the asymmetry index By comparing the current ripple residual vector with the pre-stored baseline fingerprint vector Euclidean distance Determine the fault attribute, if and The system identifies a branch metallic grounding fault and outputs a high-level blocking command. This drives the logic gate circuit to forcibly shut down the integrator output of the dynamic response filter module and synchronously adjusts the reactive excitation current of the bidirectional power conversion module, thereby achieving physical isolation of the fault point and providing transient energy support.

[0024] Example 1: In specific application scenarios, the power distribution network faces the challenge of distributed photovoltaic power stations' output being reduced due to large-area cloud cover. sudden drop within ms And the measured voltage at the bus node drops to the preset operating threshold. At this point, the system triggers dynamic sensing of the physical topology of the power distribution network, and the ripple feature detection module utilizes a sampling frequency of... kHz data acquisition unit, synchronously acquiring bus nodes The current ripple generated by the switching action of the bidirectional power conversion module is detected by the ripple characteristic detection module through a cutoff frequency of [missing information]. A high-pass filter at Hz filters out the fundamental component and extracts the initial amplitude. The high-frequency ripple signal was detected, and its transmission to adjacent bus nodes was determined. The remaining amplitude at that point is Due to the drastic change in light intensity causing a shift in the equivalent impedance of the line, the impedance distribution correction module corrects the impedance distribution according to the rules. The procedure calculates and updates the impedance weighting coefficients in the sensitivity mapping matrix. This process maps the amplitude attenuation during the transmission process of physical branches to the scheduling model in real time, so that the modified sensitivity matrix reflects the physical state of the branches affected by the current heating effect. For the power gap caused by the sudden drop in photovoltaic output, the dynamic response filtering module obtains the first response time constant of the local energy storage converter. and the second response time constant of building thermal inertia load The dynamic response filtering module utilizes the cutoff frequency of the controlled low-pass filter. The time domain allocation for the adjustment command is implemented, with energy storage nodes with response speeds in the millisecond range taking the lead in compensating for transient active power deficits.

[0025] As the hysteresis filtering process progresses, the dynamic response filtering module smoothly transitions the adjustment weights to thermally inert load nodes with high-capacity adjustment potential. When the energy storage state acquisition submodule detects the state of charge value... When the preset percentage threshold upper or lower limit is reached, the collaborative scheduling control module determines the execution order based on the modified sensitivity matrix, issues a physical phase angle offset command, and changes the phase angle of the output voltage vector of the bidirectional power conversion module. This causes a shift in the equivalent admittance of the controlled node, utilizing the physical property of impedance-guided current, phase angle The offset induces the automatic reversal of power flow direction at the voltage boundary of the bus node, and the adjacent power regulation subsystem with regulation margin takes over to maintain voltage steady state; relying on the coordinated operation of ripple feature extraction and physical phase angle offset, the distribution network realizes the redistribution of power flow in the transient process triggered by distributed power source fluctuations. Each bus node generates a closed-loop response based on the voltage deviation and local physical constraints, transforming the dynamic regulation differences of heterogeneous energy nodes into a coordinated relay feature. In an environment where the central control system does not need to distribute regulation commands over a wide area, the system maintains the voltage steady state of the bus node by utilizing the physical correlation characteristics between the bus nodes.

[0026] Example 2: In the verification experiment on the real-time identification accuracy of branch interconnection parameters and the closed-loop stability of coordinated dispatch control in the distribution network, a real-time digital simulation environment based on the solution of nodal analysis equations was constructed to simulate a network containing ten bus nodes. The system integrates a kV AC power distribution system, a distributed photovoltaic power generation unit, a lithium iron phosphate energy storage system, and a thermally inertial air conditioning load model. To simulate the measurement challenges under real industrial electromagnetic environments, the experiment actively injects a signal-to-noise ratio of [value missing] into the sampled signal. dB of Gaussian white noise, with a total harmonic distortion rate of 100 dB. The low-order harmonic interference was reduced by % to verify the extraction performance of the ripple feature detection module under non-ideal signal input; in the core parameter setting procedure, the sampling frequency was... The determination requires balancing the integrity of the ripple waveform reconstruction with the computational load of the local control module, based on the Nyquist sampling theorem and the bidirectional power conversion module. A switching frequency of kHz was used to experimentally set the sampling frequency. for kHz, while the cutoff frequency of the controlled low-pass filter The setting is based on a trade-off between suppressing high-frequency noise interference and maintaining the dynamic response speed of the power adjustment command. When the ambient noise power density increases, the cutoff frequency... The cutoff frequency was adjusted accordingly to enhance the stability of the control closed loop. Under the operating conditions of this experiment, the cutoff frequency was... Set as Hz; During the dynamic response test of distributed power generation output step fluctuation, the voltage change trend of the bus node after the power disturbance is monitored to obtain the table. The performance comparison data of the present invention sample group and the comparative sample group shown are as follows: the comparative sample group adopts a droop control strategy based on a preset static impedance parameter, while the present invention sample group activates the ripple feature detection module and the impedance distribution correction module. The output of distributed power sources dropped sharply. The data compares the operating indicators of each group at a certain percentage, and records the maximum voltage drop of the bus. The time required for the voltage to recover to the steady-state range and the relative error in the identification of branch mutual conductance parameters .

[0027] Table 1: Comparison of Branch Interconductance Parameter Identification Accuracy and Voltage Regulation Performance Analysis table It can be seen that the sample group of the present invention still maintains the branch mutual conduction parameters below a certain level under noise interference environment. The high identification accuracy provides a physical basis for correcting the sensitivity matrix, reducing the bus voltage drop relative to the control group and improving the voltage recovery time. The shortening of the impedance distribution correction module demonstrates its effectiveness in combating topology parameter drift at the physical level; this also verifies the cutoff frequency in the dynamic response filtering module. The reasonableness of the set range was ensured, and the experimental design included a gradient control group that crossed the performance inflection point. The test data are shown in the table. As shown, the overshoot includes the power adjustment command. Phase margin of the control loop .

[0028] Table 2: Verification Table of System Stability and Regulation Performance at Different Cutoff Frequencies Observation table The recorded data shows that when for When the frequency is Hz and below the preset lower limit, the system response lags, resulting in a large active power deficit during the dynamic adjustment process. for When the frequency exceeds a preset upper limit, the control closed loop becomes sensitive to high-frequency ripple, and the phase margin... Sharply down to The assembly and integration procedures of this dispatching system follow the standardized operating procedures below: Each power regulation subsystem is deployed at a designated bus node in the distribution network; the input of the ripple characteristic detection module is connected in parallel to the AC output of the bidirectional power conversion module via a current transformer; each local control module is connected to the collaborative dispatching control module via a communication link; and the data exchange cycle is set to [missing information]. ms, inject the preset initial sensitivity mapping matrix into the coordinated scheduling control module, and record the first response time constant of the energy storage converter. The second response time constant with thermal inert load Finally, the disturbance injection module is activated to perform a system self-test to identify the relative error. Stable for three consecutive sampling periods When the voltage is within % of the specified value, the system enters the distributed collaborative scheduling working state. The above test process and integration procedure jointly confirm that the physical model provided by the ripple feature detection module and the time-domain adjustment weight provided by the dynamic response filtering module produce synergistic effect under the scheduling of the physical phase angle offset command, transforming the passive response of isolated nodes into closed-loop regulation based on network interconnection, and finally achieving steady-state voltage maintenance of the distribution network under uncertain output fluctuations.

[0029] Example 3: This example combines Figures 1 to 2 This document describes a source-grid-load-storage distributed collaborative scheduling system that considers the uncertainties of wind and solar energy distribution. Figure 1As shown, the system includes a power regulation subsystem deployed at the bus nodes of the distribution network. This subsystem contains a bidirectional power conversion module and a local control module. The switching ripple generated during its operation is transmitted to the ripple feature detection module, which extracts the high-frequency current ripple and determines the branch conduction parameters. The branch conduction parameters are then sent to the impedance distribution correction module to obtain the branch conduction parameters and update the impedance weight coefficients of the sensitivity matrix. The generated corrected sensitivity matrix is ​​then sent to the collaborative dispatch control module. Simultaneously, the dynamic response filtering module receives the response time constant of the energy storage / thermal inert load, performs hysteresis filtering based on the response time constant, and adjusts the power time domain transfer. The resulting filtering result is sent to the collaborative dispatch control module. Based on the corrected sensitivity matrix and the filtering result, the collaborative dispatch control module adjusts the physical phase angle control parameter offset and induces automatic commutation of the node power flow. Finally, it feeds back the physical phase angle control parameter offset command to the power regulation subsystem.

[0030] like Figure 2 As shown, the source-grid-load-storage distributed collaborative scheduling system is divided into five main branches in its architecture. The power regulation subsystem includes a local control module and a bidirectional power conversion module. The ripple feature detection module is responsible for extracting high-frequency switching ripple and determining branch mutual conduction parameters. The impedance distribution correction module is responsible for updating the sensitivity matrix and correcting the weight coefficients. The dynamic response filtering module performs the operations of obtaining the response time constant and performing hysteresis filtering. The collaborative scheduling control module is responsible for the tasks of physical phase angle control parameter offset and power flow automatic commutation. These functional branches together constitute the overall logical architecture of the system.

[0031] Example 4: In a factory microgrid containing distributed wind turbines and high-power inductive loads, the system faces voltage drop interference caused by instantaneous short circuits in external power distribution feeders. To avoid misjudging physical faults as power uncertainties caused by wind fluctuations, the operation status monitoring module executes a collaborative arbitration procedure based on harmonic fingerprint characteristics. It collects the output current signal of the bidirectional power conversion module in real time and performs a Fast Fourier Transform, extracting the sum of the RMS values ​​of the third, fifth, and seventh odd harmonics as the first characteristic quantity, and simultaneously extracting the sum of the RMS values ​​of the second, fourth, and sixth even harmonics as the second characteristic quantity. The ratio of the first characteristic quantity to the second characteristic quantity is calculated, and this ratio is defined as the asymmetry index. When the asymmetry index In continuous The fault identification ratio continuously exceeds the preset threshold within milliseconds. And broadband ripple residual components The effective value exceeds the system rated current. When the power output drops to a certain percentage, the system determines it to be a physical fault. At this point, the collaborative scheduling control module issues a blocking command, switches to fault current limiting mode, and limits the short-circuit current by adjusting the local interface impedance of the bidirectional power conversion module. For power output fluctuations, the collaborative scheduling control module adjusts the power output based on the measured voltage at the bus node. Compared with the preset voltage reference value The difference Determine the offset increment of the physical phase angle control parameters Its calculation process is achieved through a discrete proportional-integral control loop, and its mathematical control law is as follows: ;in, This is the phase angle offset, in rad. This represents the measured voltage at the busbar node, in units of pu. This is a preset voltage reference value, in pu. This is the proportional adjustment coefficient. This is the integral adjustment coefficient.

[0032] To determine the proportional adjustment coefficient The system executes the following calibration procedure to determine the value of the branch mutual conductance parameter: In the simulation environment, the branch mutual conductance parameter is fixed, and... Increase step size The values ​​are recorded and the voltage response waveform is monitored. The recovery attenuation ratio of the bus node voltage after the disturbance is monitored. The recovery attenuation ratio is expected to be within a certain range. to Within the interval and phase margin Greater than At that time, the corresponding values ​​are determined as operating parameters; during the system deployment phase, the initial weight coefficients of the sensitivity mapping matrix are... The disturbance injection module sequentially injects voltage disturbance signals of specific frequencies into each bus node through an offline frequency sweep test program. The voltage response vectors of the remaining nodes are measured, and an initial sensitivity model is constructed by calculating the measured values ​​of the transmission mutual impedance between each node. Online impedance correction is performed through a ripple characteristic detection module, utilizing the amplitude attenuation of current ripple propagating on the physical branches. For the initial weight coefficients The update is performed in this way, and the physical phase angle offset command issued by the coordinated scheduling control module is applied to the actual physical topology, so as to realize the physical guidance of power flow under the condition of uncertain output of distributed power source.

[0033] Example 5: In the field deployment and commissioning of a newly built microgrid system, the power regulation subsystem records the reference feature vector before initiating coordinated scheduling, and the operation status monitoring module controls the bidirectional power conversion module to operate at rated power. %to Step switching is performed within the % range, and broadband ripple residual components at each power level are simultaneously collected and stored as a reference fingerprint vector. When the system monitors the broadband ripple residual components and the reference fingerprint vector in real time during operation... Euclidean distance between Exceeding the preset allowable deviation threshold When the PU (Publication Processing) occurs, the operation status monitoring module determines that a physical fault has occurred at the bus node, and the allowable deviation threshold is reached. The PU is used to shield the sensitivity mapping matrix trigger signal induced by voltage sag; when the system is connected to a load node with unknown dynamic characteristics, the disturbance injection module applies a voltage of rated amplitude to the bus node. % and pulse width is The system uses a perturbation signal of milliseconds to acquire active power response data between the energy storage converter and the thermally inert load. The dynamic response filtering module extracts the first-order inertial element characteristics from the active power response data through a numerical fitting algorithm to determine the first response time constant of the energy storage converter. The second response time constant with thermal inert load The cutoff frequency of the controlled low-pass filter is calculated based on the determined values. This ensures that the power allocation slope of the power adjustment command matches the physical response rate of the local execution unit.

[0034] With a total length of In a campus deployment scenario with km of low-voltage distribution feeders and heterogeneous converter networks, the system faces the problem of nonlinear deviation between the high-frequency ripple amplitude attenuation characteristics and geographical distance caused by uneven distribution of feeder impedance parameters. To determine the mapping relationship between branch mutual conduction parameters and physical attenuation, the ripple characteristic detection module initiates the attenuation coefficient calibration program during the system commissioning phase and controls the bus nodes. The frequency generated at this location is A kHz calibration ripple current signal at adjacent bus nodes The system synchronously acquires and receives signal amplitude, and calculates the attenuation factor per unit length by combining the geographical coordinate data between bus nodes. and the initial weight coefficients The initial calibration values ​​are corrected to match the branch conduction parameters to eliminate the influence of cable topology on topology sensing accuracy. When the system is connected to electric vehicle charging loads and inductive power loads with random fluctuations, the dynamic response filtering module performs discretization processing on the collected power step response curve data and determines the first response time constant of the energy storage converter by matching the maximum negative gradient point of the power change rate. and the second response time constant of thermal inert load. The coordinated scheduling control module obtains the short-circuit ratio parameters of the bus nodes. And based on the short-circuit ratio parameter The reciprocal adjustment of the offset step size of the physical phase angle control parameter Control the bidirectional power conversion module in to The phase angle of the output voltage vector is changed within the range. To guide power flow, the correlation between impedance distribution and phase angle shift is utilized to enable energy nodes to exhibit automatic compensation characteristics based on physical admittance during power fluctuations, resulting in an amplitude of [missing value] on the load side. During a pulsed disturbance of kW, the system voltage recovery time remains at Within ms.

[0035] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A source-grid-load-storage distributed collaborative scheduling system considering the uncertainties of wind and solar energy, characterized in that, The system includes a power regulation subsystem, a ripple characteristic detection module, an impedance distribution correction module, a dynamic response filtering module, and a collaborative scheduling control module. The power conditioning subsystem is deployed at the bus nodes of the power distribution network. Each power conditioning subsystem includes a bidirectional power conversion module and a local control module. The ripple feature detection module is configured to extract the high-frequency current ripple generated by the bidirectional power conversion module during the switching operation, and determine the branch mutual conduction parameters between the bus nodes based on the attenuation of the transmission amplitude of the high-frequency current ripple between two adjacent bus nodes. The impedance distribution correction module is configured to acquire branch mutual conductance parameters and update the impedance weight coefficients in the preset sensitivity mapping matrix accordingly to generate a corrected sensitivity matrix that reflects the actual physical impedance distribution of the distribution network. The dynamic response filtering module is configured to acquire the response time constants of the energy storage converter and the thermally inert load, and perform hysteresis filtering on the power regulation command of the local control module according to the response time constant, so as to transfer the transient power regulation from the fast response node to the slow response node in the time domain. The collaborative scheduling control module is configured to, based on the correction sensitivity matrix and the filtering results of the dynamic response filtering module, change the physical phase angle control parameters of the bidirectional power conversion module through the local control module to generate an offset. It utilizes the automatic reversal of power flow direction at the voltage boundary of the bus node to form a closed-loop voltage regulation between the power regulation subsystems. When changing the physical phase angle control parameters, the collaborative scheduling control module uses the deviation between the measured voltage of the bus node and the preset voltage threshold as the feedback control gradient to dynamically adjust the output equivalent admittance of the bidirectional power conversion module, so as to maintain voltage steady state of the distribution network under the condition of distributed power output fluctuation.

2. The source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The impedance distribution correction module follows these calculation rules when updating impedance weighting coefficients: ,in, For the corrected bus node With bus node Sensitivity weighting coefficients between them The preset initial weighting coefficients, bus node The initial amplitude of the high-frequency current ripple emitted at the location. bus node The remaining amplitude of the high-frequency current ripple received at the point.

3. The source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The system also includes an operation status monitoring module, which is configured to extract the broadband ripple residual component from the output current of the bidirectional power conversion module. When the broadband ripple residual component exhibits asymmetric harmonic characteristics, it is determined to be a physical fault, and a blocking command is output to the collaborative scheduling control module to shut down the dynamic response filtering module.

4. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 3, characterized in that, After determining that a physical fault has occurred, the operation status monitoring module performs physical isolation of the fault point by adjusting the reactive power output current of the bidirectional power conversion module and sends a lockout command to the collaborative scheduling control module.

5. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The sampling frequency of the ripple feature detection module is synchronized with the switching frequency of the bidirectional power conversion module. The ripple feature detection module filters out the 50Hz fundamental component in the bus node through a filter circuit.

6. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The dynamic response filtering module configures the cutoff frequency of the controlled low-pass filter for the bus node where the energy storage converter is located based on the response time of the energy storage converter and the response time of the thermally inert load, so as to ensure that each power regulation subsystem exhibits a unified power regulation synchronization characteristic.

7. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The system also includes a disturbance injection module, which is configured to inject physical disturbance signals into the bidirectional power conversion module. The accuracy of the sensitivity mapping matrix is ​​verified by detecting the voltage response gradient of the bus node to the physical disturbance signals.

8. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The collaborative scheduling control module enables the power regulation subsystem to update model parameters without introducing external probe links by offsetting the physical phase angle control parameters, and allocates reactive power compensation between each bus node according to the updated impedance weighting coefficient.

9. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, Each power regulation subsystem also includes an energy storage status acquisition submodule. When the collaborative scheduling control module calculates the offset of the physical phase angle control parameters, it uses the state of charge value output by the energy storage status acquisition submodule as the capacity constraint condition for the power regulation subsystem to participate in voltage regulation.

10. A source-grid-load-storage distribution collaborative scheduling system considering the uncertainties of wind and solar energy as described in claim 1, characterized in that, The collaborative scheduling control module locates the voltage over-limit risk node based on the modified sensitivity matrix, and determines the execution order of each power regulation subsystem participating in the collaborative regulation according to the impedance weighting coefficient, and issues offset commands for the physical phase angle control parameters in sequence.

Images