CN-122026543-A - Multimode coordinated optical storage charging and discharging micro-grid operation control system

CN122026543ACN 122026543 ACN122026543 ACN 122026543ACN-122026543-A

Abstract

The invention relates to a multimode coordinated optical storage charging and discharging micro-grid operation control system, which relates to the technical field of new energy micro-grids and comprises a multi-source sensing layer, an energy management optimizing layer, a coordinated control executing layer and a seamless switching control module; the invention combines the improved model predictive control algorithm with the historical data training, has the accurate pre-judging capability on the multi-source energy trend, adopts the virtual synchronous machine technology to be matched with the feedforward compensation algorithm under high-frequency sampling, so that the energy storage converter has the characteristics of physical inertial support and extremely-fast response, can capture and counteract the transient load fluctuation generated by the access of a charging pile in a second level, effectively stabilizes the bus voltage fluctuation, and simultaneously realizes the phase active compensation and the current accurate injection of switching transient through the deep cooperation of grid-connected protection and voltage frequency pre-synchronization, effectively inhibits the voltage drop and transient impact at the switching moment, and further solves the problem of switching stiffness.

Inventors

WANG HAIDONG
DU LIANGWEI
WANG XIAOZHONG
LIU MINGGUO
HOU RUFENG
LIU HONGSHAN
XIE WEI

Assignees

中和汇能(山东)电气科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260205

Claims (6)

1. The multimode coordinated optical storage charging and discharging micro-grid operation control system is characterized by comprising a multi-source sensing layer, an energy management optimizing layer, a coordinated control executing layer and a seamless switching control module, wherein: The multisource sensing layer is used for synchronously collecting current, voltage and power vector information of the photovoltaic inverter, the energy storage converter, the charging pile and the public connection point in millisecond level, integrating the current, voltage and power vector information into a micro-grid state data set and sending the micro-grid state data set to the energy management optimizing layer; the energy management optimization layer builds a power distribution model based on an improved model predictive control algorithm, acquires a plurality of sets of historical data to perform optimization training on the power distribution model, and inputs a micro-grid state data set to output an optimal power distribution scheme to the coordination control execution layer; the coordination control execution layer is used for acquiring an optimal power distribution scheme and analyzing and generating a scheme execution instruction, a virtual synchronous machine technology is adopted in the process of executing the scheme execution instruction, so that the energy storage converter presents the external characteristics of the synchronous generator, real-time load fluctuation data are acquired, millisecond-level power hedging of load fluctuation is realized by combining a feedforward compensation algorithm, and the scheme execution process is completed; The seamless switching control module comprises a grid-connected protection unit and a voltage frequency presynchronization unit, wherein the grid-connected protection unit is used for carrying out presynchronization calculation according to power distribution parameters in an optimal power distribution scheme when the system is perceived to be in grid-connected and island mode switching, so as to obtain phase compensation parameters and transient suppression parameters, and sending the phase compensation parameters and the transient suppression parameters to a coordination control execution layer; The voltage frequency presynchronization unit is used for tracking the phase of the large power grid in real time, carrying out presynchronization calculation according to power distribution parameters when phase fluctuation is sensed, obtaining a current compensation value, and sending the current compensation value to the coordination control execution layer to realize voltage drop inhibition through injection compensation current in switching transient state.
2. The multimode coordinated optical storage and discharge micro-grid operation control system according to claim 1, wherein the specific process of constructing a power distribution model to output an optimal power distribution scheme is as follows: S101, establishing a micro-grid discrete state space model, and constructing a state equation which takes energy storage system electricity quantity, bus voltage deviation and PCC point power deviation as state variables x (k) and power instructions as control variables u (k) according to a micro-grid state data set acquired by a multi-source sensing layer: ; Wherein alpha is a preset proportionality coefficient, and Dd (k) represents a predicted disturbance vector of photovoltaic output and load fluctuation; S102, acquiring a plurality of groups of historical operation data sets, wherein the historical operation data sets comprise photovoltaic power curves under different weather and charging load curves in different seasons, a weight coefficient matrix of an MPC controller is optimized by adopting a reinforcement learning algorithm, and the weight coefficient matrix comprises a state weighting matrix Q and a control weighting matrix R, and is specific: The optimization target is that the total running cost of the micro-grid under the history working condition is minimized, and the power fluctuation variance of the grid-connected point is minimized; Training output, namely obtaining an optimal controller parameter set aiming at different working conditions; S103, constructing a multi-constraint objective function Solving the following quadratic programming problem in each sampling period: ; wherein i=1, 2 the term "N", N is the amount of data sampled and, The constraint conditions are as follows: power balance constraint, namely the algebraic sum of the power of each branch is zero; the energy storage safety constraint is that Cmin is not more than SOC and not more than Cmax, SOC is the electric quantity of the energy storage system, cmax is the maximum energy storage electric quantity, cmin is the minimum energy storage electric quantity; The hedging margin constraint is that the energy storage system reserves rated power with preset duty ratio in real time for millisecond transient response; And S104, adopting a rolling optimization strategy, sending an optimization sequence u (k|k) as a current optimal power distribution scheme to a coordination control execution layer in each calculation period, and entering the next prediction period by combining feedback information perceived in real time.
3. The multimode coordinated optical storage, charging and discharging micro-grid operation control system according to claim 1, wherein the specific process of obtaining the optimal power allocation scheme and analyzing the generation scheme execution instruction is as follows: S201, acquiring an optimal power distribution scheme, wherein the optimal power distribution scheme is specifically an optimal power vector P of a photovoltaic, energy storage, charging pile and a power grid side in a future period; S202, analyzing an optimal power vector P into a current instruction under a synchronous coordinate system, wherein the current instruction comprises an active component and a reactive component; And S203, taking the optimal power vector P as given reference power of the virtual synchronous machine, calculating the internal potential phase angle and frequency of the energy storage converter according to the preset system inertia and damping coefficient, and integrating the internal potential phase angle and frequency as scheme execution instructions.
4. The multimode coordinated optical storage, charging and discharging micro-grid operation control system according to claim 1, wherein the specific process of realizing millisecond-level power hedging of load fluctuation by combining a feedforward compensation algorithm is as follows: S301, acquiring an optimal power distribution scheme, wherein the optimal power distribution scheme comprises an energy storage converter steady-state output reference Pd and a feedforward compensation sensitivity coefficient Kf; s302, optimizing and training a power distribution model based on a plurality of groups of historical data, obtaining a characteristic matrix about the load change rate and the energy storage response intensity by learning the influence of load mutation in the historical data on the frequency and the voltage of the micro-grid, and extracting Kf which is most suitable for the current environment from the characteristic matrix according to a real-time micro-grid state data set; S303, in the process of executing the instruction by the execution scheme, monitoring the load side power fluctuation quantity delta Pl in real time through a high-speed sensing link, introducing the load side power fluctuation quantity delta Pl into a control loop by adopting a feedforward control strategy, and calculating an instantaneous hedging compensation instruction Pc: Pc=Kf*ΔPl; S304, vector superposition is carried out on the steady-state output reference Pd of the energy storage converter and the instantaneous hedging compensation command Pc, and a final control vector command Ut is obtained so as to realize millisecond-level power hedging of load fluctuation.
5. The multimode coordinated optical storage, charging and discharging micro-grid operation control system according to claim 1, wherein the specific process of obtaining the phase compensation parameter and the transient suppression parameter is as follows: S401, acquiring a large power grid phase θG acquired by a voltage frequency presynchronization unit in real time, and a virtual internal potential phase θv of a virtual synchronous machine in a coordination control execution layer, and calculating an instantaneous phase difference: Δθ=θG-θv; S402, acquiring power distribution parameters in an optimal power distribution scheme to generate phase compensation parameters, wherein the phase compensation parameters are used for correcting phase angle instructions of a virtual synchronous machine, and in a preset period before triggering a switching criterion, the instantaneous phase difference is forcedly converged to zero through a proportional integral regulator so as to ensure that a potential vector in a micro-grid at the moment of switching is completely overlapped with a large grid voltage vector; S405, acquiring target power Pa distributed to an energy storage converter in an optimal power distribution scheme and real-time grid-connected power Pb fed back by a sensing layer, and calculating a power jump predicted value, wherein the calculated power jump predicted value is used for indicating a power gap which needs to be immediately taken over by an off-grid instantaneous energy storage system; s406, based on the calculated power jump predicted value, calculating to obtain a transient suppression parameter Ktr through a preset sensitivity mapping table, wherein the transient suppression parameter comprises a virtual damping correction coefficient and a transient voltage feedforward value, and sending the transient suppression parameter to a coordination control execution layer for temporarily increasing the damping coefficient of the virtual synchronous machine at the switching moment so as to suppress frequency oscillation caused by a power gap.
6. The multimode coordinated optical storage, charging and discharging micro-grid operation control system according to claim 1, wherein the specific process of obtaining the current compensation value is as follows: S501, acquiring high-frequency voltage vector information of a public connection point through a multi-source sensing layer, further utilizing synchronous reference coordinate system transformation to decompose a voltage vector into d-axis voltage ud and q-axis voltage uq, and simultaneously adopting an improved phase-locked loop to extract the frequency fg and the phase thetag of a large power grid in real time; s502, triggering a presynchronization compensation mechanism when any value of d-axis voltage ud and q-axis voltage uq exceeds a preset steady-state threshold value; S503, obtaining an optimal power distribution scheme output by an energy management optimization layer, wherein the optimal power distribution scheme comprises current operation datum points Pr and Qr of an energy storage converter and a current margin coefficient Km of the converter; S504, calculating the maximum current space available for transient compensation according to the rated current Imax of the current transformer and the current operation reference: ; s505, obtaining the voltage drop depth Deltaud and the phase fluctuation rate Deltaθ/dt, and calculating a compensation current vector required to be injected, wherein the compensation current vector comprises a reactive compensation component and an active compensation component, and the method specifically comprises the following steps: reactive compensation component wq=gu(s) × (Ub- Δud), where = Gu(s) is the voltage support gain function and Ub is the nominal voltage; active compensation component wd=gp(s) × (fg-fu), where fg is the large grid frequency, =gp(s) is the frequency support gain function, fu is the standard frequency; s506, performing priority amplitude limiting processing on the calculated reactive compensation component and the active compensation component according to the maximum current space, preferentially ensuring reactive compensation to inhibit voltage sag, and finally sending a synthesized current compensation value to a coordination control execution layer.

Description

Multimode coordinated optical storage charging and discharging micro-grid operation control system Technical Field The invention relates to the technical field of new energy micro-grids, in particular to a multimode coordinated optical storage charging and discharging micro-grid operation control system. Background Along with the development of photovoltaic, energy storage and electric automobile charging facilities, an optical storage charging and discharging integrated power station becomes an important application mode. The prior art solutions generally have the following disadvantages: 1. Extensive energy management-lack of sophisticated real-time logic for power coordination between photovoltaic, energy storage PCS, direct current charging piles (DC/DC modules), may cause overcharging, overdischarging, or inefficiency of the energy storage battery. 2. And when the commercial power is disconnected, the system black start logic is complex, the system is usually restarted after all the power is stopped, the power interruption risk of a key load is high, the automation degree of the grid-connected recovery process is low, and the system is dependent on manual intervention. 3. The protection and coordination are insufficient, namely, an integrated strategy of demand control and backflow prevention based on total gateway power is lacking in grid-connected operation, grouping control and key load guarantee strategy of an energy storage converter (PCS) are imperfect in off-grid or mode switching, and due to huge time constants and response characteristic differences of all subsystems (such as millisecond-level response of power electronic devices and minute-level scheduling of chemical batteries), power oscillation easily occurs in the existing control strategy when facing extreme weather fluctuation or large-scale vehicle access, and in off-grid mode, the system lacks voltage and frequency support of a large power grid, and the control system is difficult to realize millisecond-level power seamless real-time hedging. In view of the technical drawbacks described above, solutions are now proposed. Disclosure of Invention The invention aims to combine the historical data training with the improved model predictive control algorithm, has the accurate pre-judging capability on the multi-source energy trend, adopts the virtual synchronous machine technology to match with the feedforward compensation algorithm under the high-frequency sampling, so that the energy storage converter has the physical inertial support and the extremely-fast response characteristic, can capture and counteract the instantaneous load fluctuation generated by the connection of a charging pile and the like in the second level, effectively stabilizes the busbar voltage fluctuation, and simultaneously realizes the phase active compensation and the current accurate injection of switching transients through the deep cooperation of grid-connected protection and voltage frequency pre-synchronization, effectively inhibits the voltage drop and the transient impact at the switching moments, and further solves the problem of switching stiffness. In order to achieve the aim, the invention adopts the following technical scheme that the multimode coordinated optical storage charging and discharging micro-grid operation control system comprises a multisource sensing layer, an energy management optimizing layer, a coordinated control executing layer and a seamless switching control module, wherein: The multisource sensing layer is used for synchronously collecting current, voltage and power vector information of the photovoltaic inverter, the energy storage converter, the charging pile and the public connection point in millisecond level, integrating the current, voltage and power vector information into a micro-grid state data set and sending the micro-grid state data set to the energy management optimizing layer; the energy management optimization layer builds a power distribution model based on an improved model predictive control algorithm, acquires a plurality of sets of historical data to perform optimization training on the power distribution model, and inputs a micro-grid state data set to output an optimal power distribution scheme to the coordination control execution layer; the coordination control execution layer is used for acquiring an optimal power distribution scheme and analyzing and generating a scheme execution instruction, a virtual synchronous machine technology is adopted in the process of executing the scheme execution instruction, so that the energy storage converter presents the external characteristics of the synchronous generator, real-time load fluctuation data are acquired, millisecond-level power hedging of load fluctuation is realized by combining a feedforward compensation algorithm, and the scheme execution process is completed; The seamless switching control module comprises a grid-connected protection unit and a voltage f