A micro-inverter cooperative power grid dispatching system with adaptive regulation capability

Abstract

The present application relates to the field of control adjustment technology, and discloses a micro-assembly coordinated power grid dispatching system with self-adaptive regulation and control capability, comprising a state monitoring module, an asynchronous triggering module and a self-adaptive regulation module, the state monitoring module acquires operation data of energy coupling nodes between a main controlled system and a sub controlled system and extracts impedance characteristic parameters, the asynchronous triggering module captures instantaneous jump of state trajectory slope to generate an asynchronous trigger pulse, the self-adaptive regulation module compares time intervals of adjacent pulses, establishes a physical buffer zone through reverse expansion dead zone when the interval reaches the lower limit of the essential physical response cycle of an actuator, and maps an error signal to a power regulation instruction by using time integral transformation, the present application establishes a dynamic association mechanism of instruction issuing frequency and physical execution boundary, alleviates regulation saturation phenomenon caused by instruction overlap, suppresses fluctuation of a controlled variable, and improves self-adaptive regulation and control performance of the system under complex topology.

Description

Technical Field

[0001] This invention relates to the field of control and regulation technology, and specifically to a distribution micro-coordinated power grid dispatching system with adaptive regulation capabilities. Background Technology

[0002] Currently, within the scope of various complex closed-loop control systems and associated discrete-time regulation logic, the industry typically employs a closed-loop control architecture based on deviation feedback to mitigate state fluctuations of the target variable in response to external random disturbances to the controlled object. This architecture relies on real-time monitoring of state variables and dynamically adjusting the error tolerance accordingly. When a transient offset is detected to be aggravated, the control unit tends to shrink the trigger dead zone to increase the frequency of control command issuance, thereby achieving close tracking of the target state using a dense sequence of control pulses. Existing research focuses on macro-level scheduling coordination and information-level monitoring. For example, Chinese invention patent application CN121395698A discloses a method and system for handling main-distribution-microgrid collaborative faults. By establishing a power grid mesh model, performing power outage monitoring, and generating collaborative handling schemes, it improves the collaborative efficiency of main-distribution-microgrid fault scenarios. Such technologies focus on information layer topology identification and task dispatch. When returning to the control layer logic, they are constrained by the inherent physical limitations of the underlying actuators.

[0003] As high-frequency, non-periodic disturbances increase within the controlled node, the conventional logic of relying on unidirectional sensitivity enhancement to improve tracking accuracy exhibits inherent limitations of the underlying physical mechanism. Because the computation cycle of the discrete control algorithm is much shorter than the maximum ramp-up time of the downstream controlled actuator, the continuously shrinking trigger envelope causes the time interval between instruction issuance by the scheduling node to exceed the lower throughput limit of the physical execution channel, resulting in new and old instructions overlapping and stacking at the actuator's entry point. Conventional engineering approaches to address this instruction stacking phenomenon typically involve increasing the precision of the sampling hardware or widening the communication network bandwidth, but these methods only reduce… The hysteresis of the low information transmission layer cannot eliminate the inherent electrical inertia of the physical execution side, and may even further aggravate the integral divergence of the control loop due to the lossless transmission of high-frequency noise. Analysis reveals the following inherent defects at the general control algorithm level: 1. There is a time-domain mismatch between the ultra-high-speed triggering requirements of the discrete control algorithm and the rigid action delay of the execution hardware; 2. The dead-zone logic of unidirectional contraction under extreme disturbance conditions lacks a perception defense against the actual stress state of the controlled physical node; 3. High-frequency triggering not only increases the load on communication resources but also easily induces low-frequency oscillations in the control loop.

[0004] Therefore, the technical problem to be solved by this invention is how to construct a general adaptive closed-loop regulation and control architecture based on the collaboration of general logic control components, which can identify the essential physical execution boundary of the underlying actuator in real time through the algorithm layer, and thereby dynamically intercept and perform nonlinear time integral equivalent transformation on high-frequency asynchronous triggering instructions. Summary of the Invention

[0005] This invention proposes a distribution-micro-grid coordinated dispatching system with adaptive control capabilities, comprising:

[0006] The status monitoring module is used to acquire real-time data streams characterizing the operating status of energy coupling nodes between the main controlled system and the sub-controlled system, and to extract impedance characteristic parameters characterizing the strength properties of the energy coupling nodes.

[0007] The asynchronous triggering module is used to construct a state trajectory based on real-time data stream, and generate an asynchronous triggering pulse to activate the adaptive adjustment module when the absolute value of the slope of the state trajectory exceeds the preset slope triggering threshold.

[0008] The adaptive adjustment module includes a timing monitoring unit and a dead-zone control unit. The timing monitoring unit monitors the time interval between two adjacent asynchronous trigger pulses. The dead-zone control unit compares the time interval with the lower limit of the physical response cycle of the underlying execution unit to output a judgment signal characterizing the saturation state of the adjustment when the time interval is less than or equal to the lower limit of the physical response cycle of the underlying execution unit. Based on the judgment signal, it performs a reverse expansion operation on the current control dead zone to set the physical response buffer. Within the physical response buffer, the adaptive adjustment module performs a mathematical time integral transformation on the real-time error signal and converts the accumulated integral result into a power control command based on the preset ramp rate limit of the underlying execution unit, which is then output to the underlying execution unit.

[0009] Preferably, the asynchronous triggering module is also used to block the generation of asynchronous triggering pulses when the state trajectory enters the resting window period and the absolute value of the slope is continuously lower than the preset slope triggering threshold; during the period when the asynchronous triggering pulse is in the blocked state, the adaptive adjustment module uses the residual compensation loop to perform zero-order maintenance correction on the cumulative execution residual of the underlying execution unit so that the state variable deviation of the energy coupling node is maintained within the preset value range.

[0010] Preferably, the status monitoring module is also used to inject a small disturbance signal with preset spectral characteristics during the process of issuing power control commands, and simultaneously capture the transient phase angle response induced by the small disturbance signal, so as to invert the dynamic equivalent impedance of the current operating topology of the energy coupling node in real time.

[0011] Preferably, the dead-zone control unit is also used to adjust the sensitivity weight of the preset slope trigger threshold based on the dynamic equivalent impedance, and the adjustment logic it follows is expressed by the following formula: ,in, The corrected preset slope trigger threshold. As the baseline trigger threshold, This is the real-time measured value of the dynamic equivalent impedance. This is the reference value for the system's rated impedance.

[0012] Preferably, the system also includes an environment sensing interface for receiving real-time temperature rise parameters of the underlying execution unit; and a dead zone control unit for determining whether the underlying execution unit has entered a nonlinear hysteresis state during the opening of the physical response buffer by comparing the real-time temperature rise parameters with the preset temperature rise threshold, and actively applying a positive offset correction to the preset slope trigger threshold to widen the error envelope.

[0013] Preferably, the adaptive adjustment module is also used to perform implicit topology probing by repeatedly using the pulse sequence of the power regulation command, to capture the evolution law of the external controlled environment, and thereby tighten the adjustment tolerance window corresponding to the high impedance connection environment.

[0014] Preferably, the dead zone control unit sets an expansion coefficient based on the limit ramp rate of the underlying execution unit when performing the reverse expansion operation, so that the width of the control dead zone increases linearly as the difference between the time interval and the lower limit of the physical response cycle decreases, until the upper limit of the physical output of the underlying execution unit is reached.

[0015] Preferably, the adaptive adjustment module is also used to obtain the aging factor of the underlying execution unit and limit the upper limit of the integral intensity of the mathematical time integral transformation according to the aging factor, so as to prevent the power control command from generating integral saturation when the characteristic response frequency of the underlying execution unit shifts.

[0016] Preferably, the power control command includes: a transient compensation component for sudden load changes within the sub-controlled system, and a steady-state adjustment component for state fluctuations in the main controlled system; an adaptive adjustment module is used to execute preset decoupling logic operations to achieve asynchronous output of the transient compensation component and the steady-state adjustment component in the time domain.

[0017] Preferably, the system also includes a security verification module, which verifies whether the single control amplitude exceeds the single output step limit of the underlying execution unit before the power control command is issued, and performs amplitude limiting processing if it exceeds the limit.

[0018] The beneficial effects of this invention are:

[0019] 1. In the micro-distribution grid dispatching, a time-domain decoupling mechanism based on the extreme ramp response constant of the execution node is constructed. Through the functional coupling of the logic decision layer, the control unit monitors the time interval between two adjacent effective triggering events in real time. When it is determined that the time interval is close to or less than the physical limit response time of the underlying execution hardware, the conventional control process based on the voltage fluctuation contraction trigger threshold is actively suspended. At this time, the logic decision layer forces the system to enter a resting window of a preset duration and uses a time integration algorithm to convert the high-frequency fragmented power deviation sequence generated within the window period into an equivalent low-frequency cumulative deviation value. After the resting period ends, it is converted into a single compensation command and issued in a centralized manner. This solves the time-domain mismatch contradiction between the ultra-high-speed sampling characteristics of the discrete control algorithm and the inherent physical characteristics of the underlying controlled object, intercepts redundant high-frequency commands that exceed the physical absorption limit of the hardware, and prevents integral divergence induced in the control loop due to the congestion of the adjustment channel. An adaptive adjustment architecture based on time-domain distribution density is constructed in the control logic layer to reserve the necessary response space for the physical action sequence of the underlying execution mechanism.

[0020] 2. Establish a logic self-correcting control flow based on execution residual feedback. After issuing a power compensation command, the scheduling unit synchronously extracts the second derivative deviation between the command setpoint and the actual power response value of the node to generate an execution residual sequence. When the accumulated residual exceeds the limit, it is determined that the underlying controlled object has entered a nonlinear hysteresis state with limited physical output. At this time, the logic judgment unit actively applies a positive offset correction to the dynamic trigger threshold to widen the error envelope. This scheme breaks the rigidity of the traditional closed-loop system's one-way control that forcibly tracks external fluctuation signals. Based on a feedback mechanism that spontaneously senses the downstream physical execution boundary, it effectively suppresses the invalid trigger pulses that are continuously generated under extreme conditions such as full load of equipment or hardware aging, and prevents the control domain from falling into unnecessary steady-state regulation oscillations.

[0021] 3. Design an active sensing path that embeds environmental impedance detection excitation within routine adjustment actions. During the normal power compensation command issuance process, the control system superimposes a small signal component with specific spectral characteristics, synchronously capturing the transient voltage phase angle and amplitude response characteristics induced by this excitation component. Based on this, the dynamic equivalent impedance change trend of the current operating topology of the grid connection point is inverted. This characteristic data is fed back to the trigger judgment logic, enabling the control system to dynamically calibrate the sensitivity weight of the trigger threshold to voltage fluctuations according to the strength of the current node topology. This implicit detection mechanism that reuses the control command pulse sequence allows the system to capture the implicit topology evolution law of the external network in real time without adding any independent measurement hardware. By tightening the adjustment tolerance window in advance under weak network conditions, the system avoids the risk of systemic over-limit caused by inappropriate large step size adjustment. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram illustrating the principle architecture and signal interaction of the micro-cooperative power grid dispatching system of the present invention;

[0024] Figure 2 This is a flowchart of the adaptive dead-zone control logic for physical execution boundary awareness in this invention. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0026] A distribution-micro-grid dispatching system with adaptive control capabilities includes:

[0027] The status monitoring module is used to acquire real-time data streams characterizing the operating status of energy coupling nodes between the main controlled system and the sub-controlled system, and to extract impedance characteristic parameters characterizing the strength properties of the energy coupling nodes.

[0028] The asynchronous triggering module is used to construct a state trajectory based on real-time data stream, and generate an asynchronous triggering pulse to activate the adaptive adjustment module when the absolute value of the slope of the state trajectory exceeds the preset slope triggering threshold.

[0029] The adaptive adjustment module includes a timing monitoring unit and a dead-zone control unit. The timing monitoring unit monitors the time interval between two adjacent asynchronous trigger pulses. The dead-zone control unit compares the time interval with the lower limit of the physical response cycle of the underlying execution unit to output a judgment signal characterizing the saturation state of the adjustment when the time interval is less than or equal to the lower limit of the physical response cycle of the underlying execution unit. Based on the judgment signal, it performs a reverse expansion operation on the current control dead zone to set the physical response buffer. Within the physical response buffer, the adaptive adjustment module performs a mathematical time integral transformation on the real-time error signal and converts the accumulated integral result into a power control command based on the preset ramp rate limit of the underlying execution unit, which is then output to the underlying execution unit.

[0030] Preferably, the asynchronous triggering module is also used to block the generation of asynchronous triggering pulses when the state trajectory enters the resting window period and the absolute value of the slope is continuously lower than the preset slope triggering threshold; during the period when the asynchronous triggering pulse is in the blocked state, the adaptive adjustment module uses the residual compensation loop to perform zero-order maintenance correction on the cumulative execution residual of the underlying execution unit so that the state variable deviation of the energy coupling node is maintained within the preset value range.

[0031] Preferably, the status monitoring module is also used to inject a small disturbance signal with preset spectral characteristics during the process of issuing power control commands, and simultaneously capture the transient phase angle response induced by the small disturbance signal, so as to invert the dynamic equivalent impedance of the current operating topology of the energy coupling node in real time.

[0032] Preferably, the dead-zone control unit is also used to adjust the sensitivity weight of the preset slope trigger threshold based on the dynamic equivalent impedance, and the adjustment logic it follows is expressed by the following formula: ,in, The corrected preset slope trigger threshold. As the baseline trigger threshold, This is the real-time measured value of the dynamic equivalent impedance. This is the reference value for the system's rated impedance.

[0033] Preferably, the system also includes an environment sensing interface for receiving real-time temperature rise parameters of the underlying execution unit; and a dead zone control unit for determining whether the underlying execution unit has entered a nonlinear hysteresis state during the opening of the physical response buffer by comparing the real-time temperature rise parameters with the preset temperature rise threshold, and actively applying a positive offset correction to the preset slope trigger threshold to widen the error envelope.

[0034] Preferably, the adaptive adjustment module is also used to perform implicit topology probing by repeatedly using the pulse sequence of the power regulation command, to capture the evolution law of the external controlled environment, and thereby tighten the adjustment tolerance window corresponding to the high impedance connection environment.

[0035] Preferably, the dead zone control unit sets an expansion coefficient based on the limit ramp rate of the underlying execution unit when performing the reverse expansion operation, so that the width of the control dead zone increases linearly as the difference between the time interval and the lower limit of the physical response cycle decreases, until the upper limit of the physical output of the underlying execution unit is reached.

[0036] Preferably, the adaptive adjustment module is also used to obtain the aging factor of the underlying execution unit and limit the upper limit of the integral intensity of the mathematical time integral transformation according to the aging factor, so as to prevent the power control command from generating integral saturation when the characteristic response frequency of the underlying execution unit shifts.

[0037] Preferably, the power control command includes: a transient compensation component for sudden load changes within the sub-controlled system, and a steady-state adjustment component for state fluctuations in the main controlled system; an adaptive adjustment module is used to execute preset decoupling logic operations to achieve asynchronous output of the transient compensation component and the steady-state adjustment component in the time domain.

[0038] Preferably, the system also includes a security verification module, which verifies whether the single control amplitude exceeds the single output step limit of the underlying execution unit before the power control command is issued, and performs amplitude limiting processing if it exceeds the limit.

[0039] Example 1: When the system faces high-frequency, non-periodic energy interaction fluctuations between the main controlled system and the sub-controlled system, this system implements closed-loop state variable regulation based on the aforementioned defined architecture. In the specific regional grid interconnection scenario with high-density access to distributed energy nodes, the energy storage inverter inside the controlled microgrid acts as the underlying execution unit. Its core power conversion component contains inherently large-capacity insulated-gate bipolar transistors. This type of hardware has a strict physical ramp rate limit when outputting power throughput scheduling pulses, that is, the lower limit of the single physical response period is fixed at 20ms. With the sudden jumps in external grid load and the distribution within the microgrid... The superposition of power output disturbances causes the state parameters at the energy coupling node to exhibit nonlinear high-frequency oscillations. This leads to a sharp contraction in the time interval of the power regulation pulse sequence output in fixed steps under the conventional control architecture, which then falls below the aforementioned 20ms physical response period lower limit. This cross-scale time-domain mismatch between the high-speed triggering characteristics based on discrete control algorithms and the rigid action delay of the underlying physical hardware results in the physical stacking and signal truncation of old instructions that have not yet been executed and newly arrived control instructions at the input port of the underlying execution unit, causing integral saturation within the control loop and low-frequency resonance at the energy coupling node.

[0040] To resolve the deep-seated contradiction between discrete control timing and continuous physical response, the system's internal state monitoring module continuously acquires real-time voltage and current data streams at energy coupling nodes using a 5kHz reference sampling frequency. Simultaneously, it extracts the dynamic equivalent impedance, characterizing the grid strength of the node, as an environmental sensing reference parameter. The asynchronous triggering module constructs a multi-dimensional state trajectory based on this real-time data stream. When the absolute value of the instantaneous slope of the state trajectory exceeds the preset slope trigger threshold initially configured by the system, an asynchronous trigger pulse is generated to activate the adaptive adjustment module. Before constructing the state trajectory, the system preprocesses the real-time voltage and current data streams acquired at the 5kHz sampling frequency: a median filtering algorithm with 5 sampling points is used to remove high-frequency glitches, and the overlap rate of the sliding data window is set to 80%. The slope extraction logic is as follows: the algebraic difference between the last two filtered data points within the current sliding window is calculated, and then divided by a fixed sampling period of 0.2ms. Only when the absolute value of the slope for three consecutive sampling periods penetrates the preset slope trigger threshold is it considered a valid wave. The adaptive adjustment module generates asynchronous trigger pulses. The timing monitoring unit inside the adaptive adjustment module uses a high-precision hardware counter to capture the time interval between two adjacent asynchronous trigger pulses and sends it to the dead zone control unit. When the numerical comparison logic in the dead zone control unit determines that the current time interval drops to 18ms and touches the 20ms physical response cycle lower limit critical zone, it outputs a judgment signal representing the adjustment saturation state. The system then stops the basic adjustment link based on the unidirectional contraction error tolerance of external state fluctuations and instead extracts the expansion coefficient based on the limit ramp rate of 0.5kW per millisecond calibrated by the underlying execution unit. It applies a reverse expansion operation to the current control dead zone, thereby forcibly setting a physical response buffer with a duration of 50ms within the control logic flow. During the period when the physical response buffer is kept open, the integral operation node inside the adaptive adjustment module performs mathematical time integral transformation on the real-time error signal continuously captured by the state monitoring module, which transforms the high-frequency fragmented error sequence that originally exceeded the digestion capacity of the underlying hardware into a single-dimensional energy accumulation integral result.

[0041] When the dead zone control unit performs the reverse expansion operation, the adaptive adjustment module extracts the preset ramp rate limit R and the lower limit of the physical response period from the underlying execution unit. As a baseline constraint, when the timing monitoring unit captures the time interval Δt decreasing to the critical threshold of 20ms, the dead zone control unit adjusts the proportional coefficient accordingly. The current control dead zone width W is widened, and the proportional coefficient is obtained by extracting the time interval Δt and the lower limit of the physical response period. The difference is determined by normalization, so that the width W increases linearly as the difference decreases. During the open period between physical response buffers, the adaptive adjustment module starts the hardware integral accumulator to perform mathematical time integral transformation on the real-time active power error signal ΔP(t) to generate the cumulative integral result E. After the timing program between physical response buffers terminates, the adaptive adjustment module, based on the inherent physical output upper limit of the underlying execution unit and the aforementioned ramp rate limit, converts the accumulated integral result into a power control command that matches the hardware throughput capability and outputs it centrally to the underlying execution unit. This reflects the autonomous perception and logical reconstruction of physical boundary conditions by the internal functional modules of the control system, and establishes a dynamic mapping relationship between the command issuance cycle and the physical execution boundary. In essence, this invention embeds a logic buffer with time perception capability in the control loop, and uses the time domain extension and mathematical equivalent reconstruction of the error morphology of the control logic layer to intercept high-frequency redundant control pulses in the logic judgment stage. Under the premise of maintaining the existing physical hardware topology unchanged, it suppresses the reverse overshoot of the state variable induced by the adjustment saturation of the underlying execution unit, and maintains the global adjustment convergence state of the controlled energy coupling node under complex aperiodic disturbance topology.

[0042] Example 2: A semi-physical microgrid power regulation verification platform with a rated power of 100kW was constructed. A grid simulation source with harmonic injection function was connected. Gaussian white noise with a signal-to-noise ratio of 20dB was actively superimposed at the output port of the grid simulation source, and a 50Hz power frequency interference harmonic was injected to simulate the electromagnetic environment of energy coupling nodes in a real industrial scenario. At the same time, the system's internal status monitoring module continuously acquired the raw voltage and current data stream of the node at a sampling frequency of 5kHz. Regarding the setting of the duration of the core parameter, namely the physical response buffer, the essence of the technical trade-off lies in balancing the real-time tracking accuracy of control commands with the underlying isolation. The minimum physical switching cycle of the insulated gate bipolar transistor (IGBT) hardware is specifically quantized by extracting the initial voltage drop slope caused by the step load, combining it with the preset ramp rate limit of 0.5kW per millisecond, and calculating the minimum time window function that allows the time integral energy to cover the load gap based on the signal aliasing boundary of the state variable under the sampling theorem. When the spectral bandwidth of the monitored signal is wide and the instantaneous slope is steep, this duration setting tends to take the lower limit to prevent the voltage depth from exceeding the limit. Based on this decision rule, when the system faces a sudden load unloading disturbance of 30kW under rated operating conditions, the calculated engineering setting value of the physical response buffer is 50ms.

[0043] To verify the technical effectiveness of this parameter setting and establish the reasonable boundaries of the reverse expansion operation mechanism, three sudden load unloading disturbance conditions with intensity gradients of 10kW, 30kW, and 50kW were set up in the experiment. Four comparative sample groups were established for the core disturbance scenario of 30kW intensity. Among them, control group 1 adopted a standard proportional-integral control architecture with dead zone control unit removed, control group 2 forced the physical response buffer interval to be 10ms, which is lower than the 20ms physical response period lower limit, and control group 3 set this interval to 100ms, which exceeds the normal response tolerance. The experimental group adopted the scheme of this invention with complete adaptive adjustment module and set the physical response buffer interval to 50ms according to the above calibration logic. After applying the composite test condition of 30kW load unloading and 20dB noise, the state monitoring module captured the instantaneous voltage surge of the energy coupling node. The absolute value of the slope of the state trajectory constructed by the asynchronous triggering module instantly penetrated the preset slope triggering threshold and generated dense asynchronous trigger pulses. The timing monitoring unit captured that the time interval between two adjacent pulses drastically contracted to 12.4ms. Under this state, the lack of control dead zone control resulted in a sudden increase in the voltage of the energy coupling node. In the first control group with zone constraints, the high-frequency control pulse directly impacted the underlying execution unit, causing the internal calculation program to saturate. Its output power exhibited a low-frequency physical oscillation with an amplitude of 35.2%. In the second control group, the 10ms setting interval failed to cover the inherent 20ms recovery period of the hardware, causing the underlying execution unit to directly trigger the thermal protection relay and physically shut down at 11.2ms. In the third control group, the ultra-long dead zone of 100ms caused the adaptive adjustment module to not issue any power action command in the initial stage, resulting in the maximum voltage deviation of the energy coupling node reaching 48.5V, which exceeded the safe operating range allowed by the conventional power grid. The dead zone control unit of the experimental group accurately determined that the time interval of 12.4ms was less than the lower limit of the physical response period of 20ms, output a judgment signal representing the adjustment saturation state, and performed a reverse expansion operation to lock the control input channel of the underlying execution unit and establish a 50ms physical response buffer. During this period, the adaptive adjustment module continuously received real-time error signals and performed mathematical time integral transformation, which converted the high-frequency redundant error sequence contaminated by electromagnetic noise within 50ms into a one-dimensional energy demand.

[0044] After the 50ms timing window expires, the adaptive adjustment module, based on a preset ramp rate limit of 0.5kW per millisecond, converts the one-dimensional energy demand into a single power control command of 24.8kW and outputs it. The underlying execution unit of the test group receives this command, and its output power transitions to a new steady state within 148.5ms. The final power overshoot measured by the system is 3.2%. Further gradient ergodic test data shows that when the sudden load intensity, which characterizes the severity of the disturbance, increases sequentially from 10kW to 50kW, the power overshoot of the test group stabilizes at 2.8%, 3.2%, and 4.1%, respectively. Furthermore, when the parameters within the physical response buffer are... As the value gradually increases from 10ms to 50ms, the recorded power overshoot exhibits a non-linear exponential decay. When the value in this range exceeds the physical inflection point of 62.5ms, the recorded voltage recovery time of the system deviates from the flat region and increases linearly and significantly. The above objective measurement data confirms that within the value range of 40ms to 60ms, the reverse expansion operation triggered by the time interval and the mathematical time integral transformation change the distribution density of the high-frequency discrete control command in the time domain, block the physical transmission path of high-frequency electromagnetic noise and sudden disturbances to the underlying execution hardware, and maintain the convergence state of the adaptive adjustment loop under wide-range fluctuation conditions.

[0045] Example 3: Facing the operating environment of energy coupling nodes in a microgrid caused by distributed power source access and superimposed with broadband electromagnetic interference, the state monitoring module acquires real-time voltage and current data streams representing the operating status of energy coupling nodes between the main controlled system and the sub-controlled system at a preset sampling frequency; the asynchronous triggering module delineates a sliding data window containing 10 consecutive sampling periods in the built-in memory, constructs the voltage and current sequences pressed into the window into a discrete two-dimensional state matrix, uses a first-order difference algorithm to extract the numerical change between two adjacent data points at the end of the state matrix, and defines the ratio of this change to a fixed sampling period as the absolute value of the instantaneous slope of the state trajectory. This data processing link converts simulated energy fluctuations into discrete digital slope features with a defined time step.

[0046] To address the physical phenomenon of the grid strength attributes of energy-coupled nodes evolving with the grid topology, the state monitoring module injects minute disturbance signals during the issuance of power regulation commands, simultaneously capturing the transient phase angle response induced by these signals, and calculating and extracting impedance characteristic parameters characterizing the strength attributes of energy-coupled nodes. The state monitoring module then inverts the dynamic equivalent impedance. At that time, a pseudo-random binary sequence signal with an amplitude of 3% to 5% of the system's rated power and a spectral bandwidth covering 10Hz to 500Hz is superimposed on the power control command as a disturbance source. The voltage response sequence v(t) and current response sequence i(t) of the energy coupling node are obtained using a synchronous trigger controller with a sampling frequency of 5kHz. The built-in discrete Fourier transform operator is called to extract the frequency domain feature components, and the complex ratio of the components at the same frequency is calculated to obtain the dynamic equivalent impedance. The disturbance source period length is set to 1023 bits. Before calculation, the background noise spectrum of the bus under no-load conditions is scanned to determine the reference noise floor. When the amplitude of the disturbance source-induced response signal exceeds the reference noise floor by 10dB, the calculation result is determined as the impedance characteristic parameter and input into the dead zone control unit to trigger the preset slope threshold. The anchoring is adjusted based on measurement data supported by objective signal-to-noise ratio. The logic comparison component inside the asynchronous trigger module obtains the reciprocal value of the impedance characteristic parameter and dynamically updates the preset slope trigger threshold based on the reciprocal value. When the absolute value of the slope extracted in the current sliding data window exceeds the preset slope trigger threshold, the asynchronous trigger module generates and sends a high-level asynchronous trigger pulse to the adaptive adjustment module to activate the adaptive adjustment module. The timing monitoring unit included in the adaptive adjustment module receives this pulse and monitors the time interval between two adjacent asynchronous trigger pulses. When the time interval is less than or equal to the lower limit of the physical response cycle of the underlying execution unit, the dead zone control unit compares the time interval with the lower limit of the physical response cycle to output a judgment signal characterizing the adjustment saturation state, and performs a reverse expansion operation on the current control dead zone based on the judgment signal to set the physical response buffer.

[0047] Within the physical response buffer, the adaptive adjustment module activates its built-in hardware integrator to perform a mathematical time-integral transformation on the real-time error signal. This transformation process follows the formula... Where E represents the cumulative integral result in joules. The initial trigger time for the reverse expansion operation is represented by T, which represents the fixed duration of the physical response buffer. ΔP(t) represents the real-time active power error signal continuously output by the state monitoring module. After the integral operation cycle ends, the adaptive adjustment module divides the accumulated integral result E by the duration T to convert it into a reference power adjustment amount. The product of the preset ramp rate limit of the underlying execution unit and the duration T is extracted as the upper limit of the physical constraint increment. The amplitude of the reference power adjustment amount is truncated using this upper limit. Finally, the truncated value is converted into a power control command and output to the underlying execution unit. This logical path is based on dimensional alignment and hardware capability boundary constraints. Through mathematical equivalent transformation algorithm, the high-frequency state deviation is reconstructed into a single control output that matches the nonlinear physical constraint limit of the actuator. This process essentially realizes the dynamic alignment of the discrete control sequence and the continuous physical process within the control algorithm, maintaining the convergence state of the energy coupling node under fluctuation disturbance.

[0048] Example 4: When the system faces objective conditions such as a new batch of underlying execution units connecting to energy coupling nodes or an overall topology change in the deployment environment, the system measures boundary parameters before entering the closed-loop regulation state. The state monitoring module sends a set of stepped active power step test pulse sequences to the underlying execution units in the no-load test state. The adaptive regulation module captures the transient power evolution trajectory of the response pulse sequence at the output port of the underlying execution unit with a sampling step size of 1μs. The system extracts the maximum first derivative value of the trajectory in the no-overshoot attenuation stage as a benchmark reference value, multiplies it by a derating factor of 0.9 to calculate the preset ramp rate limit. This measurement and calculation path avoids response delay fluctuations caused by hardware manufacturing discreteness or aging differences, and anchors the preset ramp rate limit within the objective power throughput range of the physical entity.

[0049] After establishing the physical boundaries, the state monitoring module injects a small perturbation signal covering the 10Hz to 50Hz frequency band into the energy coupling node, analyzes the phase feedback response excited by the small perturbation signal to extract the initial impedance characteristic parameters under steady state, and the asynchronous triggering module obtains the initial impedance characteristic parameters and applies them according to the formula. Calculate the scaling factor; where K represents the scaling factor. This represents the maximum permissible instantaneous power fluctuation amplitude set at the input terminal. This represents the extracted initial impedance characteristic parameters. In subsequent real-time operation cycles, the asynchronous trigger module calculates the product of the proportional mapping coefficient K and the reciprocal of the real-time extracted impedance characteristic parameters, and uses this product value to dynamically refresh the preset slope trigger threshold. The system then loads all calibrated values ​​and parameters and enters the adaptive control closed-loop operation state.

[0050] Example 5: When the system is deployed in a specific microgrid environment with periodic switching characteristics of multiple loads, and the state trajectory enters the resting window period, the system activates the discrete state deviation calibration procedure before formal grid connection to determine the preset numerical range boundary of the state variable deviation. Under the controlled isolation state of blocking the asynchronous trigger pulse generation channel, the adaptive adjustment module continuously injects the rated reference power into the underlying execution unit and maintains it for 10 standard power frequency cycles. The state monitoring module synchronously reads the natural steady-state voltage fluctuation sequence of the energy coupling node during this period. The system extracts the peak-to-peak value of the fluctuation sequence and multiplies it by the calibrated safety margin coefficient. The calculated voltage tolerance extreme value is defined as the upper and lower limit reference of the preset numerical range. This offline calibration process transforms the objective static withstand capability of the power grid into a definite control logic judgment boundary.

[0051] During the real-time closed-loop operation phase, when the asynchronous triggering module determines that the absolute value of the slope of the current state trajectory is continuously lower than the preset slope triggering threshold and maintains the blocking state of the asynchronous triggering pulse, the residual compensation loop built into the adaptive adjustment module starts the online error interception and correction program. The state monitoring module synchronously captures the historical power control command sequence issued in the previous active cycle and the actual physical output power sequence fed back by the underlying execution unit at a fixed sampling frequency. The residual compensation loop performs algebraic accumulation operation on the difference between the above two sets of sequences at the corresponding time nodes to generate the cumulative execution residual characterizing the hardware execution deviation. This loop performs zero-order hold correction on the cumulative execution residual of the underlying execution unit, extracts the final state value of the cumulative execution residual of the current calculation cycle, and applies the formula... This is converted into static compensation power; where, The static compensation power represents the zero-order hold correction output. This represents the pre-defined dimensionless steady-state gain coefficient. The final value of the cumulative execution residual is represented by the static compensation power. The adaptive adjustment module adds this static compensation power as a constant bias to the control port of the underlying execution unit and keeps the value locked until the new asynchronous trigger pulse unlocks the state. This data processing path uses the feedforward freezing mechanism of physical quantities to offset the static execution loss of the underlying hardware while keeping the state variable deviation of the energy coupling node within the preset value range.

[0052] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A distribution-micro-grid dispatching system with adaptive control capabilities, characterized in that, include: The status monitoring module is used to acquire real-time data streams characterizing the operating status of energy coupling nodes between the main controlled system and the sub-controlled system, and to extract impedance characteristic parameters characterizing the strength properties of the energy coupling nodes. The asynchronous triggering module is used to construct a state trajectory based on real-time data stream, and generate an asynchronous triggering pulse to activate the adaptive adjustment module when the absolute value of the slope of the state trajectory exceeds the preset slope triggering threshold. The adaptive adjustment module includes a timing monitoring unit and a dead-zone control unit; The timing monitoring unit is used to monitor the time interval between two adjacent asynchronous trigger pulses; the dead zone control unit is used to compare the time interval with the lower limit of the physical response cycle of the underlying execution unit when the time interval is less than or equal to the lower limit of the physical response cycle of the underlying execution unit to output a judgment signal characterizing the adjustment saturation state, and to perform a reverse expansion operation on the current control dead zone according to the judgment signal to set the physical response buffer. Within the physical response buffer, the adaptive adjustment module performs a mathematical time integral transformation on the real-time error signal, and based on the preset ramp rate limit of the underlying execution unit, converts the accumulated integral result into a power control command and outputs it to the underlying execution unit.

2. The distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The asynchronous triggering module is also used to block the generation of asynchronous triggering pulses when the state trajectory enters the resting window period and the absolute value of the slope is continuously lower than the preset slope triggering threshold. During the period when the asynchronous triggering pulse is blocked, the adaptive adjustment module uses the residual compensation loop to perform zero-order maintenance correction on the cumulative execution residual of the underlying execution unit so that the state variable deviation of the energy coupling node is maintained within the preset value range.

3. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The status monitoring module is also used to inject a small disturbance signal with preset spectral characteristics during the power control command issuance process, and simultaneously capture the transient phase angle response induced by the small disturbance signal, so as to invert the dynamic equivalent impedance of the current operating topology of the energy coupling node in real time.

4. A distribution micro-grid dispatching system with adaptive control capability according to claim 3, characterized in that, The dead-zone control unit is also used to adjust the sensitivity weight of the preset slope trigger threshold based on the dynamic equivalent impedance. The adjustment logic it follows is expressed by the following formula: ,in, The corrected preset slope trigger threshold. As the baseline trigger threshold, This is the real-time measured value of the dynamic equivalent impedance. This is the reference value for the system's rated impedance.

5. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The system also includes an environment sensing interface for receiving real-time temperature rise parameters from the underlying execution unit; and a dead zone control unit for determining whether the underlying execution unit has entered a nonlinear hysteresis state during the opening of the physical response buffer by comparing the real-time temperature rise parameters with the preset temperature rise threshold, and actively applying a positive offset correction to the preset slope trigger threshold to widen the error envelope.

6. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The adaptive adjustment module is also used to perform implicit topology probing by repeatedly using the pulse sequence of power control commands, to capture the evolution of the external controlled environment, and to tighten the adjustment tolerance window corresponding to the high impedance connection environment accordingly.

7. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, When performing reverse expansion operation, the dead zone control unit sets the expansion coefficient according to the limit ramp rate of the underlying execution unit, so that the width of the control dead zone increases linearly as the difference between the time interval and the lower limit of the physical response cycle decreases, until the upper limit of the physical output of the underlying execution unit is reached.

8. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The adaptive adjustment module is also used to obtain the aging factor of the underlying execution unit and limit the upper limit of the integral intensity of the mathematical time integral transform according to the aging factor, so as to prevent the power control command from generating integral saturation when the characteristic response frequency of the underlying execution unit shifts.

9. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The power regulation commands include: transient compensation components for sudden load changes within the sub-controlled system, and steady-state adjustment components for state fluctuations in the main controlled system; an adaptive adjustment module is used to execute preset decoupling logic operations to achieve asynchronous output of transient compensation components and steady-state adjustment components in the time domain.

10. A distribution micro-grid dispatching system with adaptive control capability according to claim 1, characterized in that, The system also includes a security verification module, which verifies whether the single control amplitude exceeds the single output step limit of the underlying execution unit before the power control command is issued, and performs amplitude limiting processing if it exceeds the limit.

Images