A micro-grid flexible load and energy storage collaborative control method and system

Abstract

A method and system for coordinated control of flexible loads and energy storage in a microgrid is disclosed. The method includes: acquiring the net power fluctuation signal of the target microgrid, physical line impedance parameters, real-time ambient temperature data and thermodynamic parameters, and the real-time operating voltage of the microgrid bus; converting the net power fluctuation signal into an active power modulation command when the amplitude of the net power fluctuation signal exceeds a preset active power intervention dead zone; predicting transient voltage offset; generating a feedforward reactive power compensation signal; controlling the flexible load to adjust its active power and injecting the feedforward reactive power compensation signal into the control loop of the energy storage inverter; generating a power recovery trajectory for the flexible load; and driving the flexible load to recover from the target active power corresponding to the active power modulation command to its initial operating power before participating in fluctuation mitigation when the amplitude of the net power fluctuation signal falls back to within the preset active power intervention dead zone. This application improves the efficiency of coordinated control of flexible loads and energy storage in a microgrid.

Description

Technical Field

[0001] This application relates to the field of collaborative control technology, specifically to a collaborative control method and system for flexible loads and energy storage in a microgrid. Background Technology

[0002] With a high proportion of renewable energy (such as photovoltaic and wind power) being connected to microgrids, the grid connection points of microgrids often face severe net power fluctuations due to the randomness of weather conditions and user electricity consumption behavior. In order to mitigate these power fluctuations, in addition to configuring expensive energy storage systems, fully exploring the regulation potential of "flexible loads" (such as temperature-controlled loads and industrial adjustable loads) within the microgrid and enabling them to participate in the grid's demand-side response has become an important technical approach to maintaining the real-time power balance of the microgrid.

[0003] In existing technologies, when power fluctuations at the grid connection point exceed limits, the system typically intervenes by directly issuing active power regulation commands to the flexible load. Simultaneously, the energy storage inverters within the microgrid passively output reactive power to support voltage based on the actual voltage dips in the local area. Once the grid fluctuations subside, the system directly cancels the intervention commands, and the flexible load autonomously resumes normal operation.

[0004] However, in the intervention phase, the active and reactive power regulation are separated in the standalone control method. Due to the impedance coupling characteristics of the microgrid, sudden changes in active power from flexible loads can cause transient voltage deviations on the bus, and the lagging passive reactive power support of the energy storage inverter cannot suppress this deviation instantaneously. In the withdrawal phase, load recovery is disconnected from the grid's steady-state state. After the intervention ends, the flexible loads will instantly restart at full power to make up for the accumulated energy deficit. This step-like independent recovery can easily cause secondary power surges to the microgrid, reducing the coordinated control efficiency of the microgrid's flexible loads and energy storage. Summary of the Invention

[0005] This application provides a method and system for coordinated control of flexible loads and energy storage in microgrids, which addresses the technical problems caused by the lack of coordination between flexible loads and energy storage when smoothing power fluctuations in microgrids, resulting in transient voltage shifts in active power regulation and secondary impacts when loads are removed. This improves the efficiency of coordinated control of flexible loads and energy storage in microgrids.

[0006] The first aspect of this application provides a method for coordinated control of flexible loads and energy storage in a microgrid, the method comprising: The system acquires the net power fluctuation signal of the target microgrid at the grid connection point, the physical line impedance parameters between the access node of the flexible load in the target microgrid and the microgrid bus, the real-time ambient temperature data and thermodynamic parameters of the flexible load, and the real-time operating voltage of the microgrid bus. The physical line impedance parameters include the equivalent line resistance and the equivalent line reactance. The target microgrid also includes an energy storage inverter connected in parallel to the microgrid bus. When the amplitude of the net power fluctuation signal exceeds the preset active power intervention dead zone, the net power fluctuation signal is converted into an active power modulation command for smoothing grid fluctuations. Based on the equivalent line resistance, the equivalent line reactance, and the real-time operating voltage, predict the transient voltage offset caused by executing the active power modulation command on the microgrid bus. A feedforward reactive power compensation signal is generated based on the transient voltage offset and the equivalent line reactance, and the feedforward reactive power compensation signal is used to offset the transient voltage offset. The active power modulation command controls the flexible load to adjust its active power, and within the synchronous timing or preset lead timing of controlling the flexible load, the feedforward reactive power compensation signal is injected into the control loop of the energy storage inverter to adjust the reactive power of the microgrid bus. The absolute temperature difference between the real-time ambient temperature data of the flexible load and the preset temperature threshold is determined as the current thermal inertia margin. Based on the current thermal inertia margin and the thermodynamic parameters, a power recovery trajectory for the flexible load is generated. This power recovery trajectory is used to suppress the secondary impact on the microgrid caused by the power step of the flexible load. When the amplitude of the net power fluctuation signal falls back to the preset active power intervention dead zone, based on the power recovery trajectory, the flexible load is driven to recover from the target active power corresponding to the active power modulation command to the initial operating power before participating in the fluctuation smoothing.

[0007] Optionally, based on the equivalent line resistance, the equivalent line reactance, and the real-time operating voltage, the transient voltage offset caused by executing the active power modulation command on the microgrid bus is predicted, specifically including: Monitor the step amplitude of the active power modulation command; If the command step amplitude is less than or equal to the preset transient stability margin, then the base voltage offset is calculated based on the resistive voltage drop characteristics of the active power modulation command and the equivalent line resistance, and the base voltage offset is determined as the transient voltage offset. If the step amplitude of the instruction is greater than the preset transient stability margin, the resistive voltage drop component mapped by the active power modulation instruction on the equivalent line resistance is extracted. The node power angle difference between the access node and the microgrid bus is determined based on the command step amplitude and the real-time operating voltage. Based on the node power angle difference, the active power modulation command, and the equivalent line reactance, an inductive voltage additional component is generated to characterize the nonlinear voltage deviation. The transient voltage offset is obtained by fusing the resistive voltage drop component with the inductive voltage additional component.

[0008] Optionally, based on the node power angle difference, the active power modulation command, and the equivalent line reactance, an inductive voltage additional component is generated to characterize the nonlinear voltage deviation, specifically including: Extract the target active power adjustment amount from the active power modulation command, and perform gain amplification processing on the target active power adjustment amount based on the equivalent line reactance to generate a transverse voltage drop product term; The ratio of the lateral voltage drop product term to the real-time operating voltage is determined as the lateral voltage drop component; The square of the transverse voltage drop component is calculated, and the ratio of the square to the real-time operating voltage is determined as the second-order voltage deviation term. The second-order voltage deviation term is used to characterize the second-order contraction feature of the voltage amplitude. Obtain the phase angle projection coefficient corresponding to the node power angle difference. The phase angle projection coefficient is used to characterize the degree of orthogonal deviation between the voltage phasor of the access node and the voltage phasor of the microgrid bus. Gain compensation is performed on the second-order voltage deviation term based on the phase angle projection coefficient to obtain the inductive voltage additional component.

[0009] Optionally, a feedforward reactive power compensation signal is generated based on the transient voltage offset and the equivalent line reactance, specifically including: The transient voltage offset is configured as the expected voltage recovery amplitude of the microgrid bus, and the ratio of the expected voltage recovery amplitude to the equivalent line reactance is determined as the feedforward reactive current reference. The feedforward reactive current reference is used to characterize the amount of reactive current required to offset the transient voltage offset. The feedforward reactive current reference is modulated and amplified based on the real-time operating voltage to obtain the feedforward reactive compensation signal.

[0010] Optionally, based on the current thermal inertia margin and the thermodynamic parameters, a power recovery trajectory for the flexible load is generated, specifically including: Extract the equivalent thermal resistance parameter and the equivalent heat capacity parameter from the thermodynamic parameters, and construct the basic thermal time constant of the flexible load based on the equivalent thermal resistance parameter and the equivalent heat capacity parameter; The current thermal inertia margin is configured as the recovery rate modulation factor, and the basic thermal time constant is dynamically corrected based on the recovery rate modulation factor to obtain the target power ramp-up time constant. Starting from the target active power corresponding to the active power modulation command and ending from the initial operating power, a first-order inertial response curve is generated based on the target power ramp-up time constant, and the first-order inertial response curve is determined as the power recovery trajectory.

[0011] Optionally, the basic thermal time constant is dynamically corrected based on the recovery rate modulation factor to obtain the target power ramp-up time constant, specifically including: The ratio of the recovery rate modulation factor to the preset limit temperature difference threshold is determined as the dynamic adjustment weight; Based on the dynamically adjusted weights, the basic thermal time constant is scaled proportionally to obtain the expected recovery time constant. Obtain the minimum physical response time constant of the flexible load, and determine the maximum value between the expected recovery time constant and the minimum physical response time constant as the target power ramp-up time constant.

[0012] Optionally, based on the power recovery trajectory, the flexible load is driven to recover from the target active power corresponding to the active power modulation command to its initial operating power before participating in fluctuation smoothing, specifically including: S1. Discretize the power recovery trajectory according to a preset control cycle and extract the dynamic power reference value of the current control cycle; S2. Obtain the real-time operating power of the flexible load, and determine the difference between the dynamic power reference value and the real-time operating power as the power tracking deviation; S3. Generate a bottom-level drive signal based on the power tracking deviation, and adjust the operating power of the flexible load according to the bottom-level drive signal; Repeat steps S1-S3 until the absolute deviation between the real-time operating power and the initial operating power is less than the preset steady-state tolerance, thus completing the power recovery of the flexible load.

[0013] Secondly, embodiments of this application provide a coordinated control system for flexible loads and energy storage in a microgrid. The coordinated control system includes one or more processors and a memory. The memory is coupled to the one or more processors and is used to store computer program code, which includes computer instructions. The one or more processors call the computer instructions to cause the coordinated control system for flexible loads and energy storage in the microgrid to perform the methods described in the first aspect and any possible implementation thereof.

[0014] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a microgrid flexible load and energy storage coordinated control system, cause the microgrid flexible load and energy storage coordinated control system to perform the method described in the first aspect and any possible implementation thereof.

[0015] Fourthly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a microgrid flexible load and energy storage coordinated control system, cause the microgrid flexible load and energy storage coordinated control system to execute the method described in the first aspect and any possible implementation thereof.

[0016] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: 1. By employing a technical solution that predicts transient voltage offset based on physical line impedance parameters and real-time operating voltage, and generates a feedforward reactive power compensation signal, injecting energy storage inverters within the synchronous or preset lead time sequence of controlling the active power regulation of flexible loads, and generating a power recovery trajectory based on the current thermal inertia margin to drive the flexible loads to restore their initial operating power, this application actively offsets the transient voltage offset caused by the execution of active power modulation commands during the intervention phase through the feedforward reactive power compensation signal, achieving time-series coordination between flexible load active power regulation and energy storage reactive power compensation. Simultaneously, during the exit phase, the power recovery trajectory is used to drive the flexible loads to smoothly transition, avoiding the power step caused by directly canceling the command. This effectively solves the technical problems in the prior art where the lack of coordinated control leads to transient voltage offset on the bus during the intervention phase and secondary power impact on the microgrid during the exit phase. Thus, it achieves the technical effects of ensuring transient stability of the microgrid bus voltage and suppressing secondary impacts when loads exit, significantly improving the coordinated control efficiency of flexible loads and energy storage in the microgrid.

[0017] 2. By monitoring the step amplitude of the active power modulation command and combining it with the preset transient stability margin, the prediction process of transient voltage offset is adaptively graded. When the command step amplitude is greater than the preset transient stability margin, not only is the resistive voltage drop component extracted, but also an inductive voltage additional component is generated by combining the node power angle difference. Specifically, when generating this inductive voltage additional component, the transverse voltage drop component is constructed by extracting the target active power regulation and the equivalent line reactance, and then the second-order voltage deviation term characterizing the secondary contraction of voltage amplitude is calculated. The gain compensation is performed using the phase angle projection coefficient corresponding to the node power angle difference. This process breaks through the limitations of traditional microgrid voltage assessment which relies on a single linear impedance voltage drop. It fully considers the phase angle orthogonal deviation and nonlinear voltage distortion caused by high-power step transmission when microgrids cope with severe power fluctuations. It effectively solves the technical problem in existing technologies where the deep coupling effect between line reactance and power angle difference under large disturbance conditions is ignored, leading to distortion in the prediction of transient voltage offset and resulting in over- or under-compensation of reactive power compensation for energy storage. This achieves the technical effect of quantifying the transient voltage offset of the microgrid bus with high accuracy under different power step amplitudes, providing an accurate data benchmark for the generation of feedforward reactive power compensation signals, and further consolidating the voltage transient stability support capability of microgrids under complex and severe fluctuation conditions.

[0018] 3. When generating the power recovery trajectory of the flexible load, a basic thermal time constant is constructed by extracting the equivalent thermal resistance and equivalent heat capacity parameters from the thermodynamic parameters. The current thermal inertia margin is configured as the recovery rate modulation factor. This basic thermal time constant is dynamically corrected to obtain the target power ramp-up time constant. Then, taking the target active power corresponding to the active power modulation command as the starting point and the initial operating power as the ending point, a first-order inertial response curve is generated based on the target power ramp-up time constant as the power recovery trajectory. This process deeply integrates the static physical and thermodynamic properties of the flexible load with the dynamic real-time environmental state, so that the power recovery rate is no longer a rigid fixed value, but can be adaptively adjusted according to the current thermal inertia state of the load. This smooth transition mechanism based on the first-order inertial response curve effectively solves the technical problem in existing technologies where flexible loads, when exiting the smoothing task, use simple and crude step recovery or fixed-rate ramping, failing to take into account both the real-time thermal state of the load and the grid's carrying capacity. This easily leads to secondary power surges in the microgrid or causes the load's thermal state to exceed limits. Furthermore, it enables the creation of a dynamically adaptive power recovery path tailored to the flexible load. While completely eliminating the secondary impact of power step surges on the microgrid, it maximizes the thermodynamic stability of the flexible load itself, significantly improving the power smooth transition capability and the precision of coordinated control of the microgrid during the load exit phase. Attached Figure Description

[0019] Figure 1This is a flowchart illustrating a method for coordinated control of flexible loads and energy storage in a microgrid, as described in this application. Figure 2 This is a schematic diagram of the process for predicting transient voltage offset in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a coordinated control system for flexible loads and energy storage in a microgrid, provided in an embodiment of this application.

[0020] Explanation of reference numerals in the attached drawings: 301, Central Processing Unit; 302, Read-Only Memory; 303, Random Access Memory; 304, Bus; 305, Input / Output Interface; 306, Input Section; 307, Output Section; 308, Storage Section; 309, Communication Section; 310, Driver; 311, Removable Media. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0022] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0023] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0024] Figure 1 This is a flowchart illustrating a method for coordinated control of flexible loads and energy storage in a microgrid, as described in this application.

[0025] Please see Figure 1 This application provides a method for coordinated control of flexible loads and energy storage in a microgrid. The method includes: S101. Acquire the net power fluctuation signal of the target microgrid at the grid connection point, the physical line impedance parameters between the access node of the flexible load in the target microgrid and the microgrid bus, the real-time ambient temperature data and thermodynamic parameters of the flexible load, and the real-time operating voltage of the microgrid bus. The physical line impedance parameters include equivalent line resistance and equivalent line reactance. The target microgrid also includes an energy storage inverter connected in parallel to the microgrid bus. As an independent power system achieving local energy autonomy and bidirectional power interaction, the target microgrid uses the microgrid bus as the core hub for energy aggregation and distribution in its physical topology. The target microgrid is connected to the external main grid through a grid connection point, which is a crucial gateway for monitoring the overall power surplus / deficit status of the target microgrid. Within the target microgrid, flexible loads are installed at specific access nodes. These access nodes are connected to the microgrid bus via transmission cables of a certain length and material properties. The inherent physical impedance parameters of these transmission cables constitute the electrical coupling bridge between the flexible loads and the microgrid bus. Simultaneously, the target microgrid is also equipped with energy storage inverters, which are directly connected in parallel to the microgrid bus, serving as a rapid response unit for voltage support and reactive power throughput. In terms of operational logic, when a sudden change in the output of distributed power sources within the target microgrid causes a power imbalance at the grid connection point, the flexible load is responsible for adjusting the active power through physical lines to smooth out the fluctuations. Meanwhile, the energy storage inverter connected in parallel to the microgrid bus is responsible for cooperating with the flexible load's actions and maintaining the voltage stability of the microgrid bus by injecting reactive power. Together, they form a spatial and temporal coordinated control closed loop relying on the microgrid bus.

[0026] After clarifying the physical architecture and logical relationships of the target microgrid, the primary prerequisite for achieving precise coordinated control of flexible loads and energy storage devices during the actual operation of the microgrid is to comprehensively perceive the electrical state and physical boundary conditions within the microgrid. Therefore, the core objective of step S101 is to construct a complete multi-dimensional information perception model that includes grid fluctuation states, line physical topology, load thermodynamic boundaries, and bus voltage references.

[0027] In practice, the first step is to collect bidirectional power flow data in real time using smart meters or microprocessor-based integrated protection devices installed at the connection point between the target microgrid and the main grid. The net power fluctuation signal at the grid connection point of the target microgrid is then extracted from this data. The reason for obtaining the net power fluctuation signal is that distributed power sources such as photovoltaics and wind power within the microgrid are greatly affected by the environment, easily causing an imbalance between the overall power generation and consumption of the microgrid. The net power fluctuation signal can directly reflect the degree of power impact that the current microgrid is causing on the external grid, thus serving as a trigger benchmark for subsequent judgments on whether to activate the intervention mechanism.

[0028] Simultaneously, the system needs to obtain the physical line impedance parameters between the access nodes of flexible loads within the target microgrid and the microgrid bus. These physical line impedance parameters specifically include equivalent line resistance and equivalent line reactance. In the underlying physical laws of AC microgrids, when active and reactive power are transmitted in the physical lines, a voltage drop inevitably occurs across the equivalent line resistance and equivalent line reactance. This is the "active power-voltage" coupling effect commonly found in microgrids. Obtaining accurate equivalent line resistance and equivalent line reactance by consulting the cable specification drawings from the microgrid's construction or by utilizing the online impedance identification algorithm of the microgrid's central controller is crucial for using mathematical models to predict the specific impact of power changes on voltage in subsequent steps.

[0029] Furthermore, for flexible loads involved in regulation (such as large central air conditioning systems and industrial temperature-controlled heating furnaces), the system collects real-time ambient temperature data of the flexible load through temperature sensors deployed in the operating area of ​​the flexible load. Simultaneously, it reads the thermodynamic parameters preset in the control panel or nameplate of the flexible load equipment. These thermodynamic parameters typically include, but are not limited to, the equipment's equivalent thermal resistance and equivalent heat capacity. The reason for acquiring real-time ambient temperature data and thermodynamic parameters is that flexible loads differ from ordinary lighting loads; they possess the physical characteristic of converting electrical energy into heat energy and storing it (i.e., thermal inertia). Mastering real-time ambient temperature data and thermodynamic parameters allows for an accurate assessment of the current thermodynamic state of the flexible load, providing boundary constraints for planning a smooth power recovery path and preventing over-regulation that could lead to uncontrolled ambient temperature.

[0030] Simultaneously, the system utilizes voltage transformers configured on the microgrid bus side to obtain the real-time operating voltage of the microgrid bus through high-frequency sampling. This real-time operating voltage reflects the current steady-state voltage level of the microgrid and serves as an indispensable initial reference for subsequent calculations of transient voltage offsets. Furthermore, the target microgrid also includes an energy storage inverter connected in parallel to the microgrid bus side. As a hardware carrier within the microgrid with rapid reactive power throughput capabilities, the energy storage inverter is always on standby, ready to receive subsequently generated compensation commands.

[0031] By comprehensively executing the above-mentioned multi-dimensional data acquisition actions, the information silos in traditional microgrid control, which "only consider power and not impedance" or "only adjust electrical and not thermal," can be completely broken. This lays a solid and accurate data foundation for the subsequent implementation of active and reactive power feedforward coordination and a smooth exit mechanism based on thermal inertia.

[0032] For example, suppose there's a microgrid in an industrial park containing rooftop solar panels and a central air conditioning system (flexible load). Around noon, a cloud obscures the solar panels, causing a sudden surge in power demand from the main grid at the smart meter at the grid connection point. The system then receives a net power fluctuation signal of +50kW. Simultaneously, the system retrieves the physical impedance parameters of the 200-meter cable between the central air conditioning access node and the park's busbar, determining the equivalent line resistance to be 0.15 ohms and the equivalent line reactance to be 0.08 ohms. Next, the temperature sensor in the workshop reports a real-time ambient temperature of 24℃, and the thermodynamic parameters of the air conditioner (e.g., heat capacity of 500kJ / ℃) are read. Finally, the voltage transformer at the busbar measures the current real-time operating voltage as a standard 380V. A 100kVA energy storage inverter is also connected in parallel within the park. By collecting all the above data, the microgrid central controller possesses all the initial conditions for deriving the next coordinated control strategy.

[0033] S102. When the amplitude of the net power fluctuation signal exceeds the preset active power intervention dead zone, the net power fluctuation signal is converted into an active power modulation command for smoothing grid fluctuations. In the daily operation of a microgrid, minor power disturbances constantly exist at the grid connection point due to slight environmental changes or random electricity consumption behavior of users. If the microgrid central controller responds to every minor disturbance and frequently dispatches flexible loads for power regulation, it will not only cause mechanical wear and lifespan reduction of the flexible load equipment, but also easily trigger frequent oscillations in the microgrid control system. To avoid equipment wear and system oscillations caused by frequent dispatching, step S102 introduces the concept of a preset active power intervention dead zone. The preset active power intervention dead zone originates from the dead zone nonlinear element in classical control theory. Essentially, the preset active power intervention dead zone is a power tolerance range that allows fluctuations around the zero power fluctuation baseline. The purpose of setting the preset active power intervention dead zone is to act as a digital filtering barrier, specifically used to shield normal background power noise in the microgrid that does not threaten the stable operation of the grid.

[0034] In practice, the microgrid central controller calculates the absolute value of the acquired net power fluctuation signal in real time, i.e., the amplitude of the net power fluctuation signal, and compares this amplitude with the boundary threshold of the preset active power intervention dead zone. When the amplitude of the net power fluctuation signal is within the preset active power intervention dead zone, the system determines that the microgrid is in a self-balancing steady-state range and takes no intervention action. Once the amplitude of the net power fluctuation signal exceeds the preset active power intervention dead zone, the system determines that the microgrid has encountered a substantial power imbalance event, and the system immediately initiates an active intervention mechanism. Subsequently, the microgrid central controller calls its internal power allocation algorithm, such as a proportional-integral control algorithm or a droop control strategy, and performs mathematical mapping and quantization transformation on the net power fluctuation signal exceeding the dead zone range according to a preset response ratio, ultimately generating an active power modulation command for smoothing grid fluctuations. The active power modulation command clearly specifies the specific active power value that flexible loads need to increase or decrease at the current moment.

[0035] Specifically, when generating active power modulation commands to smooth grid fluctuations, the microgrid central controller first extracts the excess power difference of the net power fluctuation signal exceeding the preset active power intervention dead zone boundary, and multiplies the excess power difference by a power response participation factor pre-allocated to flexible loads. The power response participation factor is a weighting coefficient set according to equipment capacity and regulation rate, and its purpose is to ensure reasonable power allocation under the coordination of multiple devices within the microgrid. After multiplication, the system obtains the theoretical active power regulation amount that the flexible load currently needs to undertake. Then, the system obtains the current actual operating power of the flexible load, and algebraically superimposes the theoretical active power regulation amount and the actual operating power to calculate the target active power that the flexible load should achieve after smoothing fluctuations. In order to prevent the command from exceeding the physical capacity limit of the equipment, the system also compares and limits the target active power with the upper and lower limits of the rated power on the nameplate of the flexible load. After the target active power is processed by safety limiting, it is finally encapsulated into data and formally output as an active power modulation command.

[0036] By setting a dead zone and performing command conversion, the implementation method effectively filters out invalid high-frequency disturbances in the microgrid, ensuring the service life of flexible load equipment and the control stability of the microgrid's underlying layer. On the other hand, when the microgrid truly faces power surges, it can quickly and accurately convert macroscopic grid fluctuations into microscopic equipment actions, providing a quantitative command basis for the subsequent precise execution of power regulation tasks by flexible loads.

[0037] For example, suppose the microgrid dispatch system has a preset active power intervention dead zone of ±10 kW. When several small office computers in the park are turned on, causing a net power fluctuation signal of ±3 kW at the grid connection point, the system ignores this since the amplitude of 3 kW does not exceed the preset active power intervention dead zone of ±10 kW, and the flexible load remains unchanged. When a large patch of dark clouds suddenly appears over the park, causing a sharp drop in photovoltaic power generation, the grid connection point instantly generates a net power fluctuation signal of ±60 kW. Since the amplitude of 60 kW far exceeds the preset active power intervention dead zone, the system immediately triggers the intervention mechanism. Through the control algorithm, the ±60 kW power gap is converted into an active power modulation command, such as generating an active power modulation command requiring the central air conditioning to reduce its operating power by 60 kW, thereby quickly filling the power gap of the microgrid.

[0038] S103 predicts the transient voltage offset caused by executing the active power modulation command on the microgrid bus based on the equivalent line resistance, the equivalent line reactance, and the real-time operating voltage. After the microgrid central controller generates an active power modulation command, if it directly drives flexible loads to execute power surges, the coupling effect of the equivalent line resistance and equivalent line reactance objectively existing in the physical lines will inevitably induce a significant voltage fluctuation on the microgrid bus. To proactively defend against voltage instability, the core objective of step S103 is to predict in advance the transient voltage offset on the microgrid bus caused by the execution of the active power modulation command, based on the equivalent line resistance, equivalent line reactance, and real-time operating voltage. However, in the underlying physical laws of actual AC microgrids, the distortion mechanism caused by power surges of different magnitudes on the bus voltage varies significantly: small power adjustments usually only manifest as a linear resistive voltage drop, while drastic power steps will trigger complex nonlinear reactance coupling and phase angle shift. To ensure the absolute accuracy of the prediction results and avoid errors caused by a single linear estimation, the process of predicting the transient voltage offset must be adaptively and hierarchically refined based on the actual adjustment amplitude of the active power modulation command. Figure 2 This is a flowchart illustrating the prediction of transient voltage offset in an embodiment of this application. The following is in conjunction with... Figure 2 Step S103 will be explained in detail.

[0039] S201. Monitor the step amplitude of the active power modulation command; Before executing specific voltage predictions, the microgrid central controller first needs to monitor the command step amplitude of the active power modulation command. The command step amplitude is a physical quantity characterizing the drastic instantaneous change in flexible load power; its specific value is equal to the absolute value of the difference between the target power required in the active power modulation command and the current actual operating power of the flexible load. The reason for monitoring the command step amplitude is that, in the underlying physical laws of AC microgrids, the distortion mechanism caused by power surges of different magnitudes on the bus voltage varies significantly. Obtaining the command step amplitude provides a quantitative basis for subsequently determining the appropriate accuracy of the voltage prediction model to use.

[0040] S202. If the command step amplitude is less than or equal to the preset transient stability margin, then based on the resistive voltage drop characteristics of the active power modulation command and the equivalent line resistance, the base voltage offset is calculated and the base voltage offset is determined as the transient voltage offset. After obtaining the command step amplitude, the microgrid central controller compares the command step amplitude with the preset transient stability margin. The preset transient stability margin is a linear approximate boundary threshold pre-set for the microgrid system, typically derived from short-circuit capacity assessments or historical operating experience data. Its purpose is to distinguish whether the microgrid is currently facing a normal small disturbance or an extreme large impact. If the command step amplitude is less than or equal to the preset transient stability margin, it indicates that the current power regulation action is relatively mild. Under small disturbance conditions, the nonlinear voltage distortion caused by line reactance is negligible. Therefore, the system directly calculates the base voltage offset based on the resistive voltage drop characteristics of the active power modulation command and the equivalent line resistance. The resistive voltage drop characteristic refers to the linear voltage drop law generated when pure active power flows through a purely resistive element. Specifically, the system multiplies the power value corresponding to the active power modulation command by the equivalent line resistance and then divides by the real-time operating voltage to obtain the base voltage offset. The calculation formula is expressed as follows:

[0041] In the formula, ΔU base P represents the base voltage offset. cmd R represents the power regulation amount corresponding to the active power modulation command. eq U represents the equivalent circuit resistance. realThis represents the real-time operating voltage. After calculation, the system directly determines the base voltage offset as the transient voltage offset. Employing a hierarchical processing method, redundant nonlinear calculation steps are eliminated to the greatest extent possible while ensuring prediction accuracy. This significantly saves the underlying computing power of the microgrid central controller and enables extremely rapid estimation of voltage offset under small disturbance conditions. For example, assuming a preset transient stability margin of 20 kW, when an active power modulation command requires the central air conditioning to reduce its operating power by 5 kW, since 5 kW is less than 20 kW, the system directly multiplies 5 kW by the equivalent line resistance and divides it by the real-time operating voltage, instantly obtaining a transient voltage offset of 2 volts.

[0042] S203. If the step amplitude of the instruction is greater than the preset transient stability margin, extract the resistive voltage drop component mapped by the active power modulation instruction on the equivalent line resistance. However, if the command step amplitude exceeds the preset transient stability margin, it indicates that the microgrid is experiencing a severe power surge. Under large disturbance conditions, not only will the equivalent line resistance generate a voltage drop, but the nonlinear distortion caused by the equivalent line reactance will also be amplified sharply. Continuing to use a single linear estimation at this time will inevitably lead to severe prediction distortion. Therefore, the system must activate a high-precision component decomposition prediction mechanism. First, the system extracts the resistive voltage drop component mapped by the active power modulation command onto the equivalent line resistance. The resistive voltage drop component is essentially still a linear voltage drop portion following the resistive voltage drop characteristics. Extracting the resistive voltage drop component is to extract a definite linear reference base from the complex voltage distortion. The specific calculation formula is as follows: In the formula, U drop_R This represents the resistive voltage drop component.

[0043] S204. Determine the node power angle difference between the access node and the microgrid bus based on the command step amplitude and the real-time operating voltage; After separating the resistive voltage drop component, in order to accurately quantify the nonlinear distortion caused by reactance, the system needs to determine the node power angle difference between the access node and the microgrid bus based on the command step amplitude and the real-time operating voltage. The node power angle difference originates from the classic power flow calculation theory of power systems and is used to characterize the spatial angle offset between the voltage phasors at both ends of the line (i.e., the access node and the microgrid bus) when active power is transmitted on a large scale in a spatial line. During high-power transmission, the node power angle difference widens rapidly, forcing reactive power to redistribute within the microgrid, leading to severe voltage drops. Specifically, to determine the node power angle difference, the system multiplies the command step amplitude by the equivalent line reactance and divides it by the square of the real-time operating voltage to estimate the current node power angle difference in radians. The calculation formula is expressed as: In the formula, δ represents the node power angle difference, and Pstep X represents the step size of the instruction. eq U represents the equivalent line reactance. real This represents the real-time operating voltage. Accurately determining the node power angle difference provides crucial phase offset parameters for subsequent deduction of the inductive voltage additional component unique to large disturbance conditions, completely compensating for the fatal flaw of traditional linear prediction models that blindly ignore phase angle deviation.

[0044] For example, when an active power modulation command requires an industrial heating furnace to instantly cut off 50 kW of power, and 50 kW far exceeds the preset transient stability margin of 20 kW, the system first calculates that the resistive voltage drop component caused by the resistor is 15 volts. Subsequently, the system uses 50 kW, equivalent line reactance, and a real-time operating voltage of 380 volts to calculate the node power angle difference between the access node and the busbar to be 0.15 radians, thus providing accurate data for the next step of calculating the nonlinear voltage deviation.

[0045] S205. Based on the node power angle difference, the active power modulation command, and the equivalent line reactance, generate an inductive voltage additional component to characterize the nonlinear voltage deviation. After accurately determining the node power angle difference, in order to thoroughly restore the true distortion of the microgrid bus voltage under large disturbance conditions, the system needs to generate an inductive voltage additional component to characterize the nonlinear voltage deviation based on the node power angle difference, active power modulation command, and equivalent line reactance. The inductive voltage additional component essentially reflects the second-order voltage amplitude contraction effect induced by drastic active power fluctuations on the physical line reactance. This nonlinear contraction effect is often directly ignored in traditional linear evaluation models, but it is the core reason for the severe distortion of voltage prediction under large disturbance conditions. Because the generation mechanism of this nonlinear deviation is extremely complex, deeply coupled with the transverse voltage drop transformation of active power on the reactance, the quantization of second-order nonlinear characteristics, and the projection offset of the phase angle space, it cannot be directly obtained through simple linear multiplication and division. Therefore, the microgrid central controller must break with convention and, following rigorous mathematical deduction logic, construct and generate this crucial inductive voltage additional component layer by layer through a refined process of progressively extracting the transverse voltage drop, calculating the second-order voltage deviation term, and combining it with the phase angle projection coefficient for dynamic gain compensation. Specifically, the deduction steps include the following: Extract the target active power adjustment amount from the active power modulation command, and perform gain amplification processing on the target active power adjustment amount based on the equivalent line reactance to generate a transverse voltage drop product term; The ratio of the lateral voltage drop product term to the real-time operating voltage is determined as the lateral voltage drop component; The square of the transverse voltage drop component is calculated, and the ratio of the square to the real-time operating voltage is determined as the second-order voltage deviation term. The second-order voltage deviation term is used to characterize the second-order contraction feature of the voltage amplitude. Obtain the phase angle projection coefficient corresponding to the node power angle difference. The phase angle projection coefficient is used to characterize the degree of orthogonal deviation between the voltage phasor of the access node and the voltage phasor of the microgrid bus. Gain compensation is performed on the second-order voltage deviation term based on the phase angle projection coefficient to obtain the inductive voltage additional component.

[0046] When analyzing the voltage distortion mechanism under large disturbance conditions, the microgrid central controller must accurately quantify the nonlinear contraction effect caused by physical line reactance. In specific implementation, the system first extracts the target active power regulation from the active power modulation command. In the phasor geometry of an AC power system, when active power flows through reactance elements, it does not produce a direct voltage drop in phase with the voltage as it would when flowing through a resistor; instead, it produces an orthogonal voltage drop perpendicular to the bus voltage phasor. To quantify this orthogonal voltage drop, the system performs gain amplification on the target active power regulation based on the equivalent line reactance, that is, multiplying the target active power regulation by the equivalent line reactance to generate a transverse voltage drop product. Subsequently, the system determines the transverse voltage drop component as the ratio of the transverse voltage drop product to the real-time operating voltage. The transverse voltage drop component reflects the vertical voltage deviation reference purely caused by the coupling between active power and reactance. The calculation formula is expressed as: In the formula, U trans P represents the transverse voltage drop component. cmd X represents the target active power adjustment amount. eq U represents the equivalent line reactance. real This represents the real-time operating voltage.

[0047] After obtaining the transverse voltage drop component, since it is spatially perpendicular to the bus voltage at a 90-degree angle, its influence on the final voltage amplitude is not a simple linear addition or subtraction, but rather follows a nonlinear geometric contraction law based on the Pythagorean theorem. To transform this nonlinear geometric contraction law into a calculable algebraic quantity, the system calculates the square of the transverse voltage drop component and determines the ratio of this square to the real-time operating voltage as the second-order voltage deviation term. This second-order voltage deviation term originates from the Taylor series expansion approximation of the voltage phasor geometry and is used to precisely characterize the secondary contraction characteristics of the voltage amplitude under high-power impacts. The calculation formula is expressed as: In the formula, ΔU 2nd This represents the second-order voltage deviation term.

[0048] Furthermore, during high-power transmission, a significant spatial phase angle offset occurs between the access node and the microgrid bus. To correct the projection error caused by this spatial phase angle offset, the system obtains the phase angle projection coefficient corresponding to the node power angle difference calculated in the preceding steps. The phase angle projection coefficient is typically the cosine function value of the node power angle difference, used to characterize the degree of orthogonal deviation between the access node voltage phasor and the microgrid bus voltage phasor. Finally, the system performs gain compensation on the second-order voltage deviation term based on the phase angle projection coefficient, that is, multiplying the second-order voltage deviation term by the phase angle projection coefficient to obtain the inductive voltage additional component. The calculation formula is expressed as: In the formula, U add_L K represents the inductive voltage component. proj This represents the phase angle projection coefficient.

[0049] Through the above steps, the technical bottleneck of simply ignoring or crudely linearly equipping reactance voltage drop in traditional microgrid control strategies has been broken. The full picture of the nonlinear voltage drop induced by the change in active power on the reactance has been perfectly restored, providing impeccable data support for the subsequent generation of absolutely accurate feedforward reactive power compensation signals.

[0050] For example, suppose the active power modulation command requires an instantaneous cutoff of 50 kW of active power, the equivalent line reactance is 0.08 ohms, and the real-time operating voltage is 380 volts. The system first multiplies 50 kW by 0.08 ohms and divides by 380 volts, obtaining a transverse voltage drop component of approximately 10.5 volts. Next, the system calculates the square of 10.5 volts (110.25) and divides by 380 volts, yielding a second-order voltage deviation term of approximately 0.29 volts. Assuming the phase angle projection coefficient obtained from the node power angle difference is 0.98, the system multiplies 0.29 volts by 0.98, ultimately obtaining an inductive voltage add-on component of 0.284 volts. Although the absolute value seems small, accurately compensating for this 0.284 volt nonlinear drop is often crucial to preventing voltage collapse of the entire microgrid system when it is in a vulnerable edge state.

[0051] S206. The resistive voltage drop component and the inductive voltage additional component are fused to obtain the transient voltage offset.

[0052] After extracting the resistive voltage sag component representing the linear base and the inductive voltage offset component representing nonlinear distortion, the microgrid central controller needs to perform a final data aggregation step to obtain a complete picture of the voltage sag when the microgrid bus experiences a high-power surge. This involves fusing the resistive voltage sag component and the inductive voltage offset component to obtain the final transient voltage offset. The reason for performing signal fusion is that the actual voltage sag is caused by the combined linear voltage drop of the resistor and the nonlinear voltage drop of the reactance in the physical circuit. Using either component alone cannot recreate the true physical phenomenon; only by combining the two components can an accurate target be provided for subsequent reactive power compensation.

[0053] In practice, the microgrid central controller inputs the resistive voltage sag component and the inductive voltage additional component into its internal signal fusion module. Signal fusion is a specialized concept derived from multi-source data processing theory, used to seamlessly integrate attenuation effects caused by different physical mechanisms on the same mathematical dimension. In the underlying logic of an AC microgrid, the mechanisms by which resistive heating and reactive magnetic field energy storage affect voltage are quite different. However, after the aforementioned scalarization and phase angle projection processing, the resistive voltage sag component and the inductive voltage additional component are mathematically unified and converted into a direct attenuation of the bus voltage amplitude. Therefore, the microgrid central controller directly calls the arithmetic logic unit to perform an algebraic addition operation on the resistive voltage sag component and the inductive voltage additional component. If the complex operating condition of bidirectional power flow in the microgrid (e.g., flexible loads changing from power consumption to reverse power supply) is considered, the system will also assign corresponding directional weights to the two components according to the sign of the active power modulation command before superimposing them. The sum of the superimposed output values ​​is then determined as the transient voltage offset.

[0054] By performing signal fusion operations, the system successfully reduced the complex "active power-voltage" nonlinear coupling process into a precise scalar value, completely making up for the problem of insufficient or excessive compensation caused by ignoring the nonlinear effect of reactance in traditional microgrid control, and providing a unique and absolutely reliable target for the subsequent energy storage inverter to accurately output reactive power.

[0055] For example, considering the extreme condition of an industrial heating furnace instantly cutting off 50 kW of power, the system has calculated that the resistive voltage drop caused by the resistance is 15 volts, and simultaneously deduced that the inductive voltage drop caused by the nonlinear contraction effect of the reactance is 0.284 volts. The microgrid central controller fuses the 15 volts and 0.284 volts, that is, performs a high-precision algebraic addition, and finally obtains a transient voltage offset of 15.284 volts. 15.284 volts is the actual voltage drop that the microgrid bus will face, and it is also the precise voltage gap that the subsequent energy storage inverter needs to strive to fill.

[0056] S104. Generate a feedforward reactive power compensation signal based on the transient voltage offset and the equivalent line reactance. The feedforward reactive power compensation signal is used to offset the transient voltage offset. After accurately predicting the transient voltage deviation on the microgrid bus caused by the active power modulation command executed by flexible loads, the microgrid central controller must take proactive defensive measures before voltage instability occurs to avoid substantial impact on sensitive electrical equipment within the microgrid due to voltage distortion. Since traditional closed-loop feedback control heavily relies on the actual occurrence of voltage deviations, it objectively suffers from an unavoidable response lag. Therefore, step S104 introduces a feedforward control mechanism. The core logic of the feedforward control mechanism is to use the predicted transient voltage deviation as a known condition and, by deeply integrating the inherent strong coupling physical law between reactive power and equivalent line reactance in the AC microgrid, directly deduce a feedforward reactive power compensation signal that can offset the transient voltage deviation. To accurately translate the macroscopic voltage drop gap into power commands that can be directly recognized and executed by the underlying hardware of the energy storage inverter, the microgrid central controller needs to break away from the conventional trial-and-error adjustment mode and execute a rigorous reverse physical algebra solution process. By gradually setting the voltage recovery target, extracting the current reference, and performing power-level modulation and amplification, an absolutely accurate feedforward reactive power compensation signal is finally generated. The specific solution steps include the following: The transient voltage offset is configured as the expected voltage recovery amplitude of the microgrid bus, and the ratio of the expected voltage recovery amplitude to the equivalent line reactance is determined as the feedforward reactive current reference. The feedforward reactive current reference is used to characterize the amount of reactive current required to offset the transient voltage offset. The feedforward reactive current reference is modulated and amplified based on the real-time operating voltage to obtain the feedforward reactive compensation signal.

[0057] After acquiring the transient voltage offset, the microgrid central controller needs to convert it into low-level physical commands that the energy storage inverter can directly execute. First, the microgrid central controller directly configures the transient voltage offset as the expected voltage recovery amplitude of the microgrid bus. The expected voltage recovery amplitude is a control theory-level target setting, and its setting logic is "fill in the gaps," serving to provide an absolutely precise voltage target for subsequent reverse engineering. In the physical laws of AC microgrids, the inductive reactive voltage drop of the lines is mainly generated by reactive current flowing through the line reactance. To offset the expected voltage recovery amplitude, the system must inject a corresponding reactive current into the microgrid bus. Therefore, the microgrid central controller calls its internal division logic to calculate the ratio of the expected voltage recovery amplitude to the equivalent line reactance, and determines the calculated ratio as the feedforward reactive current reference. The feedforward reactive current reference originates from a variation of Ohm's law in classical circuit theory, and its purpose is to accurately characterize the amount of reactive current that must be injected to offset the transient voltage offset. The calculation formula is expressed as follows: In the formula, I q_ref Represents the feedforward reactive current reference, ∆U offset X represents the transient voltage offset (i.e., the expected voltage recovery amplitude). eq This represents the equivalent line reactance. Through ratio calculation, the system successfully transforms the macroscopic voltage drop problem into a microscopic current compensation requirement.

[0058] However, the upper-level power control loop of an energy storage inverter typically cannot directly recognize simple current commands; the inverter needs to receive standard power stage signals. Therefore, after obtaining the feedforward reactive current reference, the microgrid central controller needs to modulate and amplify it based on the real-time operating voltage. Modulation and amplification is a specialized operation derived from power electronic converter technology. Its principle is based on the physical law that AC power equals the product of voltage and current. Its purpose is to map and amplify the single-dimensional current reference signal into a two-dimensional power control signal, based on the actual voltage level of the current grid. In practice, the microgrid central controller multiplies the feedforward reactive current reference with the real-time operating voltage; the resulting product is the final generated feedforward reactive power compensation signal. The calculation formula is expressed as: , in the formula, Q ff U represents the feedforward reactive power compensation signal. realThis represents the real-time operating voltage. The feedforward reactive power compensation signal is a control command with a clearly defined reactive power value. Through modulation and amplification processing, the system generates a reactive power compensation command that perfectly matches the current operating state of the microgrid. This ensures that once the energy storage inverter receives the feedforward reactive power compensation signal, it can instantly output just the right amount of reactive power, "lifting" the microgrid bus voltage to the rated level before a substantial voltage drop occurs, thus eliminating the transient impact on the microgrid caused by sudden changes in flexible load power.

[0059] For example, following the aforementioned derivation of a transient voltage offset of 15.284 volts, the microgrid central controller first sets 15.284 volts as the expected voltage recovery amplitude. Assuming the microgrid's equivalent line reactance is 0.08 ohms, the system divides 15.284 volts by 0.08 ohms, calculating the feedforward reactive current reference to be 191.05 amperes. 191.05 amperes represents the additional 191.05 amperes of reactive current that must flow through the line to fill the voltage gap. Subsequently, the system obtains the current real-time operating voltage of the microgrid bus as 380 volts. The system multiplies 191.05 amperes by 380 volts, performs modulation and amplification processing, and finally obtains a feedforward reactive power compensation signal of 72599 VAR (approximately 72.6 kVAR). 72.6 kVAR is the amount of reactive power that the energy storage inverter needs to precisely inject into the microgrid bus at the same instant the flexible load operates.

[0060] S105. Control the flexible load to adjust the active power according to the active power modulation command, and inject the feedforward reactive power compensation signal into the control loop of the energy storage inverter within the synchronous timing or preset advance timing of the control of the flexible load, so as to adjust the reactive power of the microgrid bus. After generating an absolutely accurate feedforward reactive power compensation signal, the microgrid central controller enters the substantive physical execution phase. If commands are simply issued arbitrarily, the physical response speed of flexible loads, which typically contain mechanical components or thermodynamic inertia, is slow, while the energy storage inverter, as a pure power electronic device, has an extremely fast response speed. This results in a significant time lag between the flexible load and the energy storage inverter. If the reactive power compensation action of the energy storage inverter lags behind the active power surge of the flexible load, the microgrid bus will still suffer from transient voltage drops; conversely, if the compensation is premature and uncontrolled, it will trigger instantaneous overvoltage on the bus. To completely eliminate the time misalignment risk caused by the difference in equipment response speed, the microgrid central controller must implement strict timing coordination control.

[0061] In practice, the microgrid central controller first controls the flexible load to adjust its active power according to the active power modulation command, that is, it sends power increase / decrease signals to the underlying drivers of the flexible load. Simultaneously, the microgrid central controller initiates a high-precision time synchronization protocol, injecting a feedforward reactive power compensation signal into the control loop of the energy storage inverter within the synchronization sequence or preset lead sequence for controlling the flexible load. Synchronization sequence is a time alignment concept, used to precisely calculate communication delays and equipment physical action times so that the instant the energy storage inverter outputs reactive power and the instant the flexible load actually changes its active power coincide perfectly on the physical time axis. The preset lead sequence originates from the power system transient stability defense theory, and its purpose is to trigger the energy storage inverter a few milliseconds in advance, pre-establishing a reactive power voltage support barrier on the microgrid bus, given an anticipated high-power surge. The control loop refers to the closed-loop feedback algorithm structure within the energy storage inverter used to regulate voltage and current. In pursuit of the ultimate response speed, the microgrid central controller directly bypasses the slow-responding outer voltage loop inside the energy storage inverter, and uses the feedforward reactive power compensation signal as a feedforward quantity, directly superimposing it into the extremely fast-responding inner current loop or pulse width modulation generator.

[0062] By implementing rigorous timing coordination and underlying loop injection, the energy storage inverter can precisely release calculated reactive power at the same instant, or even a fraction of an instant, when voltage distortion is caused by flexible loads. The reactive power of the microgrid bus is dynamically balanced instantaneously, and the transient voltage deviation that would have inevitably occurred is perfectly canceled out in its infancy, achieving a seamless and smooth voltage transition for the microgrid when experiencing severe power fluctuations.

[0063] For example, suppose the microgrid central controller issues an active power modulation command to an industrial heating furnace to cut off 50 kW of active power. It is known that the industrial heating furnace requires a 50-millisecond response time from receiving the command to physical disconnection, while the energy storage inverter only needs 5 milliseconds to output reactive power. To achieve preset lead-time control, the microgrid central controller is set to establish voltage support 2 milliseconds in advance. Therefore, 43 milliseconds after issuing the command to the industrial heating furnace, the microgrid central controller precisely injects a 72.6 kVar feedforward reactive power compensation signal into the control loop of the energy storage inverter. When the energy storage inverter outputs exactly 72.6 kVar reactive power at 48 milliseconds, the microgrid bus voltage is supported in advance; then, immediately at 50 milliseconds, the 50 kW power from the industrial heating furnace is instantly cut off, and the 15.284 volt voltage drop that would have occurred is compensated by the pre-injected reactive power, ensuring the microgrid bus voltage remains stable.

[0064] S106. The absolute temperature difference between the real-time ambient temperature data of the flexible load and the preset temperature threshold is determined as the current thermal inertia margin. After the microgrid central controller successfully coordinated with the energy storage inverter to mitigate the transient voltage drop caused by sudden changes in active power, the flexible load is now in a controlled adjustment phase, deviating from its normal operating state. Since flexible loads (such as central air conditioning, industrial heating furnaces, or cold storage facilities) are essentially thermodynamic devices, changes in power inevitably lead to a gradual drift in the internal or surrounding temperature. To prevent excessive temperature drift from harming user comfort or disrupting industrial production processes, and to provide precise physical boundary conditions for the subsequent smooth exit of the flexible load from the controlled state, the microgrid central controller must quantitatively assess the current physical buffer state of the flexible load in real time.

[0065] In practice, the microgrid central controller first collects real-time ambient temperature data of the flexible loads using temperature sensors deployed in the flexible load's operating area. Simultaneously, the central controller retrieves a preset temperature threshold from its internal memory. This preset temperature threshold is a limit boundary value derived from user settings or process safety specifications, used to define the highest or lowest tolerable temperature limit for the flexible load when participating in grid regulation. After acquiring the real-time ambient temperature data and the preset temperature threshold, the central controller calls its internal arithmetic logic unit to subtract the real-time ambient temperature data from the preset temperature threshold and extracts the absolute value of the result, i.e., calculating the absolute temperature difference. Subsequently, the central controller directly determines the calculated absolute temperature difference as the current thermal inertia margin.

[0066] Current thermal inertia margin is a cross-disciplinary, custom-defined concept that deeply integrates thermodynamics and power system control theory. It originates from the physical inertia phenomenon in thermodynamics where temperature changes lag behind changes in heat input. It is used to quantify how much temperature buffer space a flexible load still has that can be maintained or managed by the microgrid before approaching the temperature red line. The principle is that the larger the absolute temperature difference, the farther the current temperature is from the danger boundary, and the more sufficient the thermal inertia margin of the flexible load. This allows for a slower, smoother strategy when restoring power. Conversely, the smaller the absolute temperature difference, the closer the temperature is to exceeding the limit, the more critical the thermal inertia margin, and the more urgent the subsequent recovery strategy must be.

[0067] By performing the above-mentioned calculation and margin conversion of absolute temperature difference, the microgrid central controller successfully transformed the purely physical temperature state of the flexible load into control parameters that the microgrid control algorithm can directly read and participate in mathematical calculations. This lays a solid data foundation for the subsequent customized power recovery trajectory and completely avoids the problem of blind disconnection or blind recovery caused by ignoring the physical state of equipment in traditional microgrid control.

[0068] For example, suppose the flexible load participating in the microgrid power regulation is a large cold storage facility, with a set maximum tolerable temperature (preset temperature threshold) of -18 degrees Celsius. After the cold storage facility reduces its operating power to smooth grid fluctuations for a period of time, the real-time ambient temperature data transmitted by the temperature sensor is -21 degrees Celsius. The microgrid central controller subtracts -18 degrees Celsius from -21 degrees Celsius and takes the absolute value, obtaining an absolute temperature difference of 3 degrees Celsius. Therefore, the microgrid central controller determines 3 degrees Celsius as the current thermal inertia margin. 3 degrees Celsius clearly tells the microgrid control system that the cold storage facility currently has sufficient temperature buffer space and is temporarily in an absolutely safe and controlled state.

[0069] S107. Based on the current thermal inertia margin and the thermodynamic parameters, generate the power recovery trajectory of the flexible load. The power recovery trajectory is used to suppress the secondary impact of the microgrid caused by the power step of the flexible load. After quantifying the current thermal inertia margin, the microgrid central controller needs to plan a safe exit path for flexible loads in advance. When the net power fluctuations at the microgrid's grid connection point gradually subside, directly issuing a hard command to the flexible load to instantly restore it to its initial operating power would inevitably disrupt the fragile balance newly established by the microgrid, causing severe secondary shocks to the microgrid bus and even triggering new voltage oscillations and frequency instability. To completely suppress the secondary shocks to the microgrid caused by power jumps in flexible loads, the microgrid central controller must abandon the traditional "one-size-fits-all" instantaneous reset strategy and instead tailor a smooth power recovery trajectory for the flexible load. The power recovery trajectory is a mathematical curve that defines the slow, continuous change of active power over time, designed to guide the flexible load to gradually release its regulating capacity like a "soft landing."

[0070] However, generating a perfect power recovery trajectory cannot be blindly set without considering physical reality; it must deeply integrate the current temperature buffer state of the flexible load and the inherent physical properties of the equipment. Therefore, the microgrid central controller needs to deeply analyze the underlying heat conduction model of the flexible load based on the current thermal inertia margin and thermodynamic parameters. By extracting thermal resistance and heat capacity parameters, a basic time scale is established, and the current thermal inertia margin is cleverly used as a dynamic factor for adjusting the speed. Ultimately, a first-order inertial response curve that perfectly matches the current physical state of the equipment is derived. This process includes the following steps: Extract the equivalent thermal resistance parameter and the equivalent heat capacity parameter from the thermodynamic parameters, and construct the basic thermal time constant of the flexible load based on the equivalent thermal resistance parameter and the equivalent heat capacity parameter; The current thermal inertia margin is configured as the recovery rate modulation factor, and the basic thermal time constant is dynamically corrected based on the recovery rate modulation factor to obtain the target power ramp-up time constant. Starting from the target active power corresponding to the active power modulation command and ending from the initial operating power, a first-order inertial response curve is generated based on the target power ramp-up time constant, and the first-order inertial response curve is determined as the power recovery trajectory.

[0071] After clarifying the current temperature buffer state of the flexible load, the microgrid central controller needs to tailor a smooth power recovery trajectory for the flexible load. To ensure that the power recovery trajectory does not trigger a secondary impact on the microgrid bus and fully complies with the physical bearing capacity of the flexible load itself, the microgrid central controller must deeply analyze the underlying heat conduction model of the flexible load and perform dynamic simulations in conjunction with the current thermal inertia margin.

[0072] In practical implementation, the microgrid central controller first extracts the equivalent thermal resistance and equivalent heat capacity parameters from the thermodynamic parameters. These parameters are derived from lumped parameter models in engineering thermodynamics. The equivalent thermal resistance parameter characterizes the resistance to heat dissipation from the flexible load to the external environment, while the equivalent heat capacity parameter characterizes the ability of the flexible load material itself to absorb and store heat. The microgrid central controller then uses its internal multiplication logic to multiply the equivalent thermal resistance and equivalent heat capacity parameters, constructing the fundamental thermal time constant of the flexible load based on the product. The fundamental thermal time constant is an inherent property value in a purely physical dimension, used to establish the basic time scale required for the flexible load to undergo a natural exponential decay or rise in temperature without any external intervention. The calculation formula is expressed as: In the formula, τ base R represents the fundamental thermal time constant. th C represents the equivalent thermal resistance parameter. th This represents the equivalent heat capacity parameter.

[0073] After successfully constructing a fundamental thermal time constant reflecting the static physical properties of flexible loads and configuring the current thermal inertia margin as the recovery rate modulation factor, the microgrid central controller faces the technical challenge of deeply integrating dynamic operating states with static physical parameters. Relying solely on the fundamental thermal time constant for power recovery often leads to either excessively rapid recovery causing secondary grid impacts due to ignoring the urgency of the current temperature, or excessively slow recovery impairing the user's energy experience. To completely resolve these contradictions, the microgrid central controller must perform refined dynamic correction on the fundamental thermal time constant. The specific dynamic correction derivation steps include: determining the ratio of the recovery rate modulation factor to a preset limit temperature difference threshold as the dynamic adjustment weight; scaling the fundamental thermal time constant proportionally based on the dynamic adjustment weight to obtain the expected recovery time constant; obtaining the minimum physical response time constant of the flexible load; and determining the maximum value between the expected recovery time constant and the minimum physical response time constant as the target power ramp-up time constant.

[0074] The microgrid central controller directly configures the current thermal inertia margin obtained in the aforementioned steps as the recovery rate modulation factor. Subsequently, the microgrid central controller divides the recovery rate modulation factor by a preset extreme temperature difference threshold, and determines the resulting ratio as the dynamic adjustment weight. The preset extreme temperature difference threshold refers to the maximum theoretical temperature difference span between the flexible load's normal operating temperature and the absolutely unacceptable dangerous temperature. The dynamic adjustment weight is a dimensionless proportionality coefficient between zero and one, used to quantitatively assess the percentage of the current remaining temperature buffer space within the total allowable space. The calculation formula is expressed as: , in the formula, W dyn M represents the dynamically adjusted weight. th ∆T represents the recovery rate modulation factor (i.e., the current thermal inertia margin). limit This represents the preset limit temperature difference threshold.

[0075] After obtaining the dynamic adjustment weights, the microgrid central controller performs proportional scaling on the base thermal time constant based on the values ​​of the dynamic adjustment weights. Specifically, the microgrid central controller calls its internal multiplication logic to multiply the dynamic adjustment weights by the base thermal time constant, and determines the calculated product as the expected recovery time constant. The expected recovery time constant is a theoretically derived value, its physical meaning being a quantitative representation of the time scale that the system expects flexible loads to complete power recovery, considering only the current remaining temperature buffer space. The calculation formula is expressed as: , in the formula, τ exp W represents the expected recovery time constant. dyn Represents the dynamically adjusted weight, τ base The fundamental thermal time constant represents the inherent thermodynamic buffering inertia of the flexible load, which is obtained by multiplying the previously extracted equipment equivalent thermal resistance parameter and equivalent heat capacity parameter.

[0076] However, since the expected recovery time constant is a theoretical value based on the current thermodynamic state, under certain extreme conditions (e.g., when the temperature margin is extremely small or the dynamic adjustment weight is close to zero), the calculated expected recovery time constant may be very short. If this short time constant is used to drive the equipment, it is highly likely to far exceed the safe response speed of the underlying hardware (such as mechanical valves, compressor motor chassis, or power electronic drives), which could easily lead to hardware overload damage or mechanical fatigue fracture, and may also cause the control at the microgrid's underlying layer to lose lock-up. Therefore, this embodiment introduces a rigid hardware protection line: the microgrid central controller obtains the minimum physical response time constant of the flexible load from the equipment parameter library, either factory-calibrated or engineering-set. Subsequently, the microgrid central controller calls an extreme value comparison algorithm to compare the expected recovery time constant with the minimum physical response time constant bidirectionally, extracts the maximum value between the two, and finally locks and determines it as the target power ramp-up time constant. This is a classic "lower limit cutoff to prevent over-limit" control logic, ensuring that no matter how quickly external scheduling commands or internal thermodynamic calculations require the equipment to recover, the actual operating rate of the flexible load will never exceed the physical limits that its underlying hardware can safely withstand. The calculation formula is expressed as: , in the formula, τ target τ represents the target power ramp-up time constant. min This represents the minimum physical response time constant, and max() represents the function to take the maximum value.

[0077] Suppose the flexible load participating in smoothing grid fluctuations is a large industrial cold storage facility, whose basic thermal time constant is known to be 200 seconds. In the preceding steps, it is assumed that due to the long adjustment time of the cold storage facility, its internal temperature has approached the set danger threshold, resulting in a calculated dynamic adjustment weight of only 0.15. The microgrid central controller multiplies 0.15 by 200 seconds using a proportional scaling factor, yielding an expected recovery time constant of 30 seconds. This means that theoretically, the system should "urge" the cold storage facility to quickly restore full power within 30 seconds to prevent the temperature from exceeding the limit. Simultaneously, the control system reads that the minimum physical response time constant of the core compressor unit of the cold storage facility is 45 seconds (i.e., to prevent inter-turn short circuits or mechanical surges in the motor, the compressor must undergo a hardware buffer extreme value of at least 45 seconds to smoothly transition from its current state to full load). At this point, the microgrid central controller compares the theoretically calculated 30 seconds with the physical baseline of 45 seconds and extracts the maximum value between the two, ultimately determining the target power ramp-up time constant to be 45 seconds. Under this operating condition, the control system successfully intercepted the aggressive 30-second theoretical recovery request, forcing the cold storage to generate a subsequent power recovery trajectory based on a 45-second timescale. Conversely, if the current temperature buffer of the cold storage is extremely generous, and the dynamic adjustment weight is calculated to be 0.9, then the expected recovery constant is 180 seconds (0.9 × 200 seconds). In this case, comparing 180 seconds with the baseline 45 seconds and taking the maximum value, the target power ramp-up time constant becomes 180 seconds. The system will utilize 180 seconds for flexible recovery, minimizing secondary impacts on the power grid. Through the above logic, while ensuring microgrid power smoothing and responding to thermodynamic requirements, the hardware immune protection of the underlying equipment is perfectly triggered.

[0078] After successfully deriving the target power ramp-up time constant that balances equipment safety and grid demand, the microgrid central controller needs to finally define the specific mathematical form of flexible load power recovery. If a simple linear ramp-up (i.e., uniform recovery) strategy is adopted, abrupt changes in the rate of power change will still occur at the start and end of the recovery action. This discontinuity at the derivative level can still easily trigger minute-frequency oscillations within the microgrid. To achieve the ultimate smoothness, the microgrid central controller must introduce a mathematical model that fully conforms to the laws of physical energy release in nature, thereby completely eliminating any form of power abrupt change.

[0079] In practice, the microgrid central controller first defines the physical boundary conditions of the recovery process. Starting with the target active power corresponding to the active power modulation command, the central controller represents the current controlled suppression or rise state of the flexible load. Simultaneously, it uses the initial operating power before participating in fluctuation mitigation as the endpoint, representing the normal operating state that the flexible load needs to return to. Subsequently, the central controller calls its internal function generator, substituting the starting point, endpoint, and the target power ramp-up time constant obtained in the preceding steps into the underlying algorithm to generate a first-order inertial response curve.

[0080] The first-order inertial response curve is a professional concept derived from classical automatic control principles. It is a mathematical trajectory exhibiting a natural exponential smooth transition characteristic, used to completely eliminate right-angle step edges caused by sudden power changes. The underlying principle of the first-order inertial response curve lies in utilizing the negative exponential decay characteristic of the natural logarithm base. This allows the flexible load to exhibit a large rate of change in the initial recovery phase, enabling it to quickly escape the controlled state. However, in the final stage near the initial operating power, the rate of power change gradually decays and approaches zero, thus achieving a seamless transition back to normal operation. The calculation formula is expressed as:

[0081] In the formula, P(t) represents the power recovery trajectory that varies with time. initial P represents the initial operating power (i.e., the final operating power). target τ represents the target active power (i.e., the starting point). target t represents the target power ramp-up time constant, and t represents the elapsed time of the recovery process.

[0082] Finally, the microgrid central controller directly determines the generated first-order inertial response curve as the power recovery trajectory. By generating the first-order inertial response curve, the system completely abandons the simple and crude step reset method, endowing flexible loads with intelligent recovery capabilities for "soft landing," and fundamentally eliminating the secondary impact on the microgrid that may be caused when flexible loads exit the controlled state on a large scale.

[0083] For example, suppose a flexible load (such as a large cold storage facility) has a target active power of 50 kW (starting point) when participating in regulation, and its normal initial operating power is 100 kW (ending point). After the aforementioned dynamic correction calculation, the target power ramp-up time constant is 40 seconds. The microgrid central controller substitutes 50 kW, 100 kW, and 40 seconds into the exponential function model to generate a power recovery trajectory. The generated power recovery trajectory indicates that the cold storage facility's power is 50 kW at second 0, and at second 40, the power smoothly recovers to approximately 81.6 kW (i.e., recovering 63.2% of the total difference of 50 kW), and then gradually approaches 100 kW in subsequent times. The cold storage facility will slowly ramp up strictly according to the generated exponential rise curve, and the microgrid bus will not experience any abrupt power surges during this period, achieving a perfect steady-state transition.

[0084] S108. When the amplitude of the net power fluctuation signal falls back to the preset active power intervention dead zone, based on the power recovery trajectory, the flexible load is driven to recover from the target active power corresponding to the active power modulation command to the initial operating power before participating in the fluctuation smoothing.

[0085] When the amplitude of the net power fluctuation signal at the microgrid's grid connection point falls back to the preset active power intervention dead zone, it signifies that the large disturbance in the external power grid has been completely quelled, and the microgrid system has re-established a preliminary active power balance. The preset active power intervention dead zone is a range derived from the steady-state control theory of power systems, used to define the power fluctuation boundary that the microgrid can tolerate based on its own conventional regulation capabilities. With the external disturbance eliminated, the flexible load has completed its emergency mission of assisting in smoothing grid fluctuations. To ensure the normal production and energy needs of users, the microgrid central controller must drive the flexible load from the target active power under controlled conditions to safely and smoothly recover to the initial operating power before participating in fluctuation smoothing. However, the power recovery trajectory generated by the aforementioned steps is mathematically a continuous time function curve, while the underlying digital controller of the flexible load can only recognize and execute discrete digital instructions. Simultaneously, during the long power ramp-up process, internal mechanical friction, thermodynamic losses, and minor disturbances in the power grid can all cause the actual power of the flexible load to deviate from the theoretically set trajectory. To eliminate identification obstacles in digital control systems and accumulated errors during physical execution, the microgrid central controller cannot adopt an open-loop, blindly distributed approach. Instead, it must establish a rigorous closed-loop tracking control mechanism. The microgrid central controller needs to slice the continuous mathematical curves on the time axis, extracting the precise target within each tiny time step, and continuously comparing and correcting it against the actual physical state of the flexible load. By continuously generating underlying drive signals, it guides the actual power of the flexible load to perfectly match the pre-planned power recovery trajectory. The specific closed-loop tracking drive steps include: S1. Discretize the power recovery trajectory according to a preset control cycle and extract the dynamic power reference value of the current control cycle; S2. Obtain the real-time operating power of the flexible load, and determine the difference between the dynamic power reference value and the real-time operating power as the power tracking deviation; S3. Generate a bottom-level drive signal based on the power tracking deviation, and adjust the operating power of the flexible load according to the bottom-level drive signal; Repeat steps S1-S3 until the absolute deviation between the real-time operating power and the initial operating power is less than the preset steady-state tolerance, thus completing the power recovery of the flexible load.

[0086] After acquiring the continuous power recovery trajectory, the microgrid central controller faces the physical limitation that digital control systems cannot directly execute continuous mathematical functions. To transform the macroscopic mathematical curve into microscopic instructions recognizable by the underlying hardware, the microgrid central controller must discretize the power recovery trajectory according to a preset control cycle. The preset control cycle is a time interval parameter derived from digital control theory, specifying the fixed time required for the microprocessor to execute one algorithm loop. Discretization sampling is a fundamental operation in signal processing, extracting an instantaneous value every preset control cycle along a continuous time axis. By performing discretization sampling, the microgrid central controller successfully extracts the dynamic power reference value for the current control cycle. The dynamic power reference value represents the power target that the flexible load must absolutely adhere to within the current small time segment.

[0087] Having defined the current instantaneous power target, to prevent uncontrollable deviations in flexible loads during physical execution due to mechanical friction, thermodynamic losses, or grid voltage fluctuations, the microgrid central controller introduces a rigorous closed-loop feedback mechanism. The microgrid central controller acquires the real-time operating power of the flexible load through a high-precision power sensor. Subsequently, the microgrid central controller calls its internal subtraction logic to subtract the dynamic power reference value from the real-time operating power, and the calculated difference is determined as the power tracking deviation. The power tracking deviation is a classic closed-loop control parameter used to accurately quantify the error distance between the current actual physical state of the flexible load and its theoretically planned trajectory. The calculation formula is expressed as: , in the formula, E p (k) represents the power tracking deviation in the current control cycle, P ref (k) represents the dynamic power reference value, P real (k) represents the real-time operating power.

[0088] After acquiring the power tracking deviation, the microgrid central controller generates a low-level drive signal based on the deviation. This low-level drive signal is typically a pulse-width modulated waveform or an analog voltage command, used to directly control the opening of power electronic switches or mechanical valves within the flexible load. The microgrid central controller adjusts the operating power of the flexible load according to the low-level drive signal, forcibly pulling the deviated physical equipment back onto the correct path.

[0089] However, power recovery is a lengthy process lasting tens of seconds or even minutes, far too long for a single adjustment. Therefore, the microgrid central controller continuously cycles through the steps of extracting reference values, calculating deviations, and issuing drive signals. In each cycle, the microgrid central controller monitors the absolute deviation between the real-time operating power and the initial operating power before participating in fluctuation mitigation. The microgrid central controller compares this absolute deviation with a preset steady-state tolerance. This preset steady-state tolerance is an engineering boundary value derived from the control system's steady-state error analysis, used to define an allowable range of small errors to prevent high-frequency oscillations near the endpoint. When the absolute deviation is less than the preset steady-state tolerance, it indicates that the flexible load has approached and stabilized at its normal operating state. The microgrid central controller then terminates the cycle, officially declaring the power recovery of the flexible load complete. By constructing a rigorous closed-loop tracking drive mechanism, the system eliminates accumulated errors during the physical execution process.

[0090] For example: Assume the microgrid central controller has a preset control cycle of 10 milliseconds, and the flexible load needs to smoothly recover from 50 kW to 100 kW. At a certain 10-millisecond instant during the recovery process, the system extracts the current dynamic power reference value of 60.5 kW through discretized sampling. At this time, the sensor measures the real-time operating power of the flexible load as 60.2 kW. The system subtracts 60.2 kW from 60.5 kW, resulting in a power tracking deviation of +0.3 kW. The system immediately generates a low-level drive signal based on the 0.3 kW deviation, slightly increasing the input current of the flexible load. The system repeats the above error correction process every 10 milliseconds. When the real-time operating power of the flexible load reaches 99.95 kW, the system calculates the absolute deviation from the initial operating power of 100 kW to be 0.05 kW. Assuming the preset steady-state tolerance is set to 0.1 kW, since 0.05 kW is less than 0.1 kW, the system determines that the recovery task is successfully completed, stops dynamic tracking, and the flexible load smoothly returns to normal operation.

[0091] Please see Figure 3 This is a schematic diagram of the structure of a coordinated control system for flexible loads and energy storage in a microgrid, as described in this application.

[0092] It should be noted that, Figure 3The structure of the coordinated control system for flexible loads and energy storage in a microgrid shown is merely an example and should not impose any limitations on the functionality and scope of application of the embodiments of the present invention.

[0093] like Figure 3 As shown, a coordinated control system for flexible loads and energy storage in a microgrid includes a central processing unit 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory 302 or a program loaded from a storage section 308 into a random access memory 303, such as executing the methods described in the above embodiments. The random access memory 303 also stores various programs and data required for system operation. The central processing unit 301, the read-only memory 302, and the random access memory 303 are interconnected via a bus 304. An input / output interface 305 is also connected to the bus 304.

[0094] The following components are connected to the input / output interface 305: an input section 306 including audio input devices, push-button switches, etc.; an output section 307 including an LCD display, audio output devices, indicator lights, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the input / output interface 305 as needed. A removable medium 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 310 as needed so that computer programs read from it can be installed into the storage section 308 as needed.

[0095] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit 301, it performs the various functions defined in the present invention.

[0096] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, flash memory, optical fiber, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0097] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0098] Specifically, the microgrid flexible load and energy storage coordinated control system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the microgrid flexible load and energy storage coordinated control method provided in the above embodiment.

[0099] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the microgrid flexible load and energy storage coordinated control system described in the above embodiments; or it may exist independently and not incorporated into the microgrid flexible load and energy storage coordinated control system. The storage medium carries one or more computer programs, which, when executed by a processor of the microgrid flexible load and energy storage coordinated control system, cause the microgrid flexible load and energy storage coordinated control system to implement the microgrid flexible load and energy storage coordinated control method provided in the above embodiments.

Claims

1. A method for coordinated control of flexible loads and energy storage in a microgrid, characterized in that, The method includes: The system acquires the net power fluctuation signal of the target microgrid at the grid connection point, the physical line impedance parameters between the access node of the flexible load in the target microgrid and the microgrid bus, the real-time ambient temperature data and thermodynamic parameters of the flexible load, and the real-time operating voltage of the microgrid bus. The physical line impedance parameters include the equivalent line resistance and the equivalent line reactance. The target microgrid also includes an energy storage inverter connected in parallel to the microgrid bus. When the amplitude of the net power fluctuation signal exceeds the preset active power intervention dead zone, the net power fluctuation signal is converted into an active power modulation command for smoothing grid fluctuations. Based on the equivalent line resistance, the equivalent line reactance, and the real-time operating voltage, predict the transient voltage offset caused by executing the active power modulation command on the microgrid bus. A feedforward reactive power compensation signal is generated based on the transient voltage offset and the equivalent line reactance, and the feedforward reactive power compensation signal is used to offset the transient voltage offset. The active power modulation command controls the flexible load to adjust its active power, and within the synchronous timing or preset lead timing of controlling the flexible load, the feedforward reactive power compensation signal is injected into the control loop of the energy storage inverter to adjust the reactive power of the microgrid bus. The absolute temperature difference between the real-time ambient temperature data of the flexible load and the preset temperature threshold is determined as the current thermal inertia margin. Based on the current thermal inertia margin and the thermodynamic parameters, a power recovery trajectory for the flexible load is generated. This power recovery trajectory is used to suppress the secondary impact on the microgrid caused by the power step of the flexible load. When the amplitude of the net power fluctuation signal falls back to the preset active power intervention dead zone, based on the power recovery trajectory, the flexible load is driven to recover from the target active power corresponding to the active power modulation command to the initial operating power before participating in the fluctuation smoothing.

2. The method according to claim 1, characterized in that, The prediction of the transient voltage offset on the microgrid bus caused by the execution of the active power modulation command, based on the equivalent line resistance, the equivalent line reactance, and the real-time operating voltage, specifically includes: Monitor the step amplitude of the active power modulation command; If the command step amplitude is less than or equal to the preset transient stability margin, then the base voltage offset is calculated based on the resistive voltage drop characteristics of the active power modulation command and the equivalent line resistance, and the base voltage offset is determined as the transient voltage offset. If the step amplitude of the instruction is greater than the preset transient stability margin, the resistive voltage drop component mapped by the active power modulation instruction on the equivalent line resistance is extracted. The node power angle difference between the access node and the microgrid bus is determined based on the command step amplitude and the real-time operating voltage. Based on the node power angle difference, the active power modulation command, and the equivalent line reactance, an inductive voltage additional component is generated to characterize the nonlinear voltage deviation. The transient voltage offset is obtained by fusing the resistive voltage drop component with the inductive voltage additional component.

3. The method according to claim 2, characterized in that, The generation of an inductive voltage additional component to characterize the nonlinear voltage deviation, based on the node power angle difference, the active power modulation command, and the equivalent line reactance, specifically includes: Extract the target active power adjustment amount from the active power modulation command, and perform gain amplification processing on the target active power adjustment amount based on the equivalent line reactance to generate a transverse voltage drop product term; The ratio of the lateral voltage drop product term to the real-time operating voltage is determined as the lateral voltage drop component; The square of the transverse voltage drop component is calculated, and the ratio of the square to the real-time operating voltage is determined as the second-order voltage deviation term. The second-order voltage deviation term is used to characterize the second-order contraction feature of the voltage amplitude. Obtain the phase angle projection coefficient corresponding to the node power angle difference. The phase angle projection coefficient is used to characterize the degree of orthogonal deviation between the voltage phasor of the access node and the voltage phasor of the microgrid bus. Gain compensation is performed on the second-order voltage deviation term based on the phase angle projection coefficient to obtain the inductive voltage additional component.

4. The method according to claim 1, characterized in that, The step of generating a feedforward reactive power compensation signal based on the transient voltage offset and the equivalent line reactance specifically includes: The transient voltage offset is configured as the expected voltage recovery amplitude of the microgrid bus, and the ratio of the expected voltage recovery amplitude to the equivalent line reactance is determined as the feedforward reactive current reference. The feedforward reactive current reference is used to characterize the amount of reactive current required to offset the transient voltage offset. The feedforward reactive current reference is modulated and amplified based on the real-time operating voltage to obtain the feedforward reactive compensation signal.

5. The method according to claim 1, characterized in that, The generation of the power recovery trajectory of the flexible load based on the current thermal inertia margin and the thermodynamic parameters specifically includes: Extract the equivalent thermal resistance parameter and the equivalent heat capacity parameter from the thermodynamic parameters, and construct the basic thermal time constant of the flexible load based on the equivalent thermal resistance parameter and the equivalent heat capacity parameter; The current thermal inertia margin is configured as the recovery rate modulation factor, and the basic thermal time constant is dynamically corrected based on the recovery rate modulation factor to obtain the target power ramp-up time constant. Starting from the target active power corresponding to the active power modulation command and ending from the initial operating power, a first-order inertial response curve is generated based on the target power ramp-up time constant, and the first-order inertial response curve is determined as the power recovery trajectory.

6. The method according to claim 5, characterized in that, The step of dynamically correcting the basic thermal time constant based on the recovery rate modulation factor to obtain the target power ramp-up time constant specifically includes: The ratio of the recovery rate modulation factor to the preset limit temperature difference threshold is determined as the dynamic adjustment weight; Based on the dynamically adjusted weights, the basic thermal time constant is scaled proportionally to obtain the expected recovery time constant. Obtain the minimum physical response time constant of the flexible load, and determine the maximum value between the expected recovery time constant and the minimum physical response time constant as the target power ramp-up time constant.

7. The method according to claim 1, characterized in that, The step of driving the flexible load to recover from the target active power corresponding to the active power modulation command to the initial operating power before participating in fluctuation smoothing, based on the power recovery trajectory, specifically includes: S1. Discretize the power recovery trajectory according to a preset control cycle and extract the dynamic power reference value of the current control cycle; S2. Obtain the real-time operating power of the flexible load, and determine the difference between the dynamic power reference value and the real-time operating power as the power tracking deviation; S3. Generate a bottom-level drive signal based on the power tracking deviation, and adjust the operating power of the flexible load according to the bottom-level drive signal; Repeat steps S1-S3 until the absolute deviation between the real-time operating power and the initial operating power is less than the preset steady-state tolerance, thus completing the power recovery of the flexible load.

8. A coordinated control system for flexible loads and energy storage in a microgrid, characterized in that, The microgrid flexible load and energy storage coordinated control system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the microgrid flexible load and energy storage coordinated control system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the microgrid flexible load and energy storage co-control system, the microgrid flexible load and energy storage co-control system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is running on the microgrid flexible load and energy storage co-control system, the microgrid flexible load and energy storage co-control system performs the method as described in any one of claims 1-7.

Images