Search

CN-121983990-A - Voltage stabilization control method under island operation of multi-energy system

CN121983990ACN 121983990 ACN121983990 ACN 121983990ACN-121983990-A

Abstract

The application relates to a voltage stability control method under island operation of a multi-energy system, which comprises the steps of generating a voltage predicted value sequence in a target time period, calculating a dynamic response matching index based on the voltage predicted value sequence and combining response time of an energy storage system and a conventional unit, responding to a compensation mechanism that the dynamic response matching index exceeds a preset threshold, calculating a compensation capacity evaluation factor, generating a power control instruction through a feedforward compensation channel and a feedback compensation channel, embedding an upper optimization target into a lower real-time control process through a mapping mode from a cost function to a reward function, correcting a mapping relation according to execution feedback to form a closed-loop optimization operation record, and determining a voltage stability control scheme under island operation of the multi-energy system based on the closed-loop optimization operation record. Therefore, the voltage stability control problem caused by the fluctuation of new energy and the response time difference of heterogeneous energy when the multi-energy system operates in an island is solved.

Inventors

  • MA DONGJIN

Assignees

  • 国家能源集团科学技术研究院有限公司

Dates

Publication Date
20260505
Application Date
20260123

Claims (10)

  1. 1. The voltage stability control method under the island operation of the multi-energy system is characterized by comprising the following steps of: acquiring historical voltage data, new energy output data and meteorological data of the multi-energy system in an island running state, and performing time sequence feature extraction and feature fusion to generate a voltage predicted value sequence in a target time period; Based on the voltage predicted value sequence, calculating a dynamic response matching index by combining the response time of the energy storage system and the response time of the conventional unit, wherein the dynamic response matching index is used for representing the response mismatch degree of the heterogeneous energy unit; triggering a voltage compensation mechanism comprising a feed-forward compensation path and a feedback compensation path in response to the dynamic response matching index exceeding a preset threshold; After the compensation mechanism is triggered, calculating a compensation capability assessment factor based on the current charge state, the available output power and the preset compensation power of the energy storage system, and selecting an execution path according to the compensation capability assessment factor, wherein the energy storage independent compensation path is executed in response to the compensation capability assessment factor meeting a preset independent compensation condition, and a linkage signal is generated to start a multi-energy unit cooperative compensation path in response to the compensation capability assessment factor not meeting the preset independent compensation condition; generating a pre-dispatching instruction through the feedforward compensation path and generating a supplementary adjustment instruction through the feedback compensation path under the execution path so as to obtain a power control instruction according to the pre-dispatching instruction and the supplementary adjustment instruction; Based on the power control instruction, under a layered collaborative multi-target rolling optimization framework, generating a reference power instruction sequence by taking the running cost and the voltage deviation as upper optimization targets, and outputting a final power instruction by combining the real-time running state of the equipment; Embedding the upper layer optimization target into a lower layer control process through the mapping from a cost function to a reward function, and correcting a mapping relation according to the execution feedback of the final power instruction to form a closed-loop optimization operation record; and adjusting the output ratio of each energy unit in the multi-energy system based on the closed-loop optimized operation record so as to determine a voltage stability control scheme.
  2. 2. The method of claim 1, wherein said adjusting the output ratio of each energy unit in the multi-energy system to determine a voltage stabilization control scheme comprises: Determining the current state of the multi-energy system according to the compensation capability assessment factor, and determining an initial energy output distribution action in response to an adjustment instruction associated with the current state and a linkage signal; Acquiring voltage predicted value sequence data corresponding to the initial energy output distribution action, and locally correcting the output proportion of the multi-energy system to obtain adjusted distribution data under the condition that the voltage predicted value sequence data meets a preset unstable range; Determining fluctuation conditions of voltage according to the adjusted distribution data, and correcting deviation data of the multi-energy system based on the fluctuation conditions to generate corrected voltage control parameters; based on the corrected voltage control parameters, acquiring real-time feedback signals, and classifying the feedback signals by using a pre-established support vector machine model to obtain classified control priorities; Acquiring resource scheduling information associated with the classified control priority, and rearranging the resource scheduling information to determine an optimized execution plan in response to the condition that the resource scheduling information meets preset display uneven allocation conditions; and acquiring running state data of different time periods based on the optimized execution plan, and sorting the running state data of different time periods to generate the voltage stability control scheme.
  3. 3. The method of claim 1, wherein the obtaining historical voltage data, new energy output data and meteorological data of the multi-energy system in the island operation state, performing time sequence feature extraction and feature fusion to generate a voltage predicted value sequence in a target time period, comprises: Based on the historical voltage data, the new energy output data and the meteorological data, an initial data set of the multi-energy system is constructed, and the initial data set is input into a preset multi-mode time sequence feature fusion prediction network to obtain a comprehensive data record of the multi-energy system; based on the comprehensive data record, extracting time sequence characteristics of the historical voltage data and the new energy output data to determine a local space-time characteristic set of the multi-energy system; And carrying out feature fusion processing on the local space-time feature set and the weighting information in the meteorological data to obtain a fused feature vector, and outputting a voltage predicted value sequence in the target time period based on the local space-time feature and the fused feature vector.
  4. 4. The method of claim 3, further comprising, after obtaining historical voltage data, new energy output data, and meteorological data of the multi-energy system in the island operation state, performing time sequence feature extraction and feature fusion to generate a voltage predicted value sequence in the target time period: Re-weighting the fused feature vector to obtain an adjusted predicted sequence in response to the condition that the fluctuation range of the voltage predicted value sequence in the target time period exceeds a preset fluctuation range; Analyzing the change trend of the voltage in the target time period based on the adjusted prediction sequence to determine the deviation range of the voltage prediction; Based on the deviation range of the voltage prediction and the real-time updating condition of the meteorological data, generating a corresponding prediction correction parameter, and generating a final voltage prediction result according to the prediction correction parameter.
  5. 5. The method of claim 1, wherein the triggering a voltage compensation mechanism comprising a feed-forward compensation path and a feedback compensation path in response to the dynamic response matching index exceeding a preset threshold comprises: Determining the dynamic response matching index according to the voltage predicted value sequence; Generating a trigger signal in response to the dynamic response matching index being greater than the preset threshold, and determining a priority class of the trigger signal; And acquiring the activation condition of the compensation mechanism according to the priority classification of the trigger signal, and triggering the compensation mechanism in response to the compensation mechanism meeting the activation condition.
  6. 6. The method of claim 5, further comprising, after triggering a voltage compensation mechanism comprising a feed-forward compensation path and a feedback compensation path in response to the dynamic response matching index exceeding a preset threshold: Generating the activation instruction according to the compensation mechanism, acquiring real-time state data of the energy storage system according to the activation instruction, and determining at least one execution parameter of the compensation mechanism based on the real-time state data and the response time of the thermal power generating unit; Adjusting the output power of the energy storage system according to the at least one execution parameter to obtain an adjusted system matching state; And under the condition that the adjusted system matching state reaches a preset response balance condition, acquiring an updated value of the dynamic response matching index, and determining a final system operation stability record according to the updated value of the dynamic response matching index.
  7. 7. The method of claim 1, wherein generating a pre-scheduling instruction via the feed-forward compensation path and generating a supplemental adjustment instruction via the feedback compensation path under the execution path to derive a power control instruction from the pre-scheduling instruction and the supplemental adjustment instruction comprises: acquiring current running state data of the energy storage system, and activating a bimodal mechanism of the energy storage system under the condition that the current running state data and a preset voltage range standard meet preset deviation; Based on the bimodal mechanism, carrying out parameter configuration on the feedforward compensation path by adopting a pre-established support vector machine model to generate the pre-dispatching instruction, and transmitting the pre-dispatching instruction to the energy storage system to determine an initial voltage adjustment direction; based on the initial voltage adjustment direction, acquiring real-time response data of the energy storage system, and processing the real-time voltage deviation according to the real-time response data and monitoring information of a feedback compensation path to generate the supplementary adjustment instruction.
  8. 8. The method of claim 7, further comprising, after generating a pre-scheduling command via the feed-forward compensation path and generating a supplemental adjustment command via the feedback compensation path to derive a power control command from the pre-scheduling command and the supplemental adjustment command: acquiring an output state of the energy storage system under the condition that the supplemental adjustment instruction meets a preset voltage balance condition, and adjusting control parameters of the feedback compensation path according to the output state to determine at least one voltage correction action; Acquiring updated operation data of the energy storage system based on the at least one voltage correction action, and acquiring an operation log of the bimodal mechanism based on the updated operation data, the feedforward compensation path and the feedback compensation path under the condition that the voltage meets a preset standard condition; And recording response time and deviation correction data of the energy storage system according to the operation log, and determining a final system operation state according to the response time and the deviation correction data.
  9. 9. The method of claim 1, wherein generating a reference power command sequence for an upper layer optimization objective with an operation cost and a voltage deviation under a hierarchical collaborative multi-objective rolling optimization framework, and outputting a final power command in combination with a real-time operation state of the device, comprises: Processing the target rolling period data by using a layered collaborative multi-target rolling optimization framework to obtain initial power demand distribution; Based on the initial power demand distribution, a generation rule of a power instruction sequence is constructed, and a power instruction sequence of a global reference is determined in response to the predicted demand corresponding to the generation rule meeting a preset demand threshold; based on the global reference power instruction sequence, acquiring dynamic update information of equipment data, and responding to the condition that equipment corresponding to the dynamic update information meets a preset available state condition, and generating a feedback result of the real-time state of the equipment; based on a feedback result of the real-time state of the device, acquiring an operation limiting condition of the device data, and locally correcting the power instruction sequence to determine an adjusted power instruction sequence in response to the condition that the device data does not meet the operation limiting condition; And based on the adjusted power instruction sequence, acquiring load distribution data of the equipment in a target rolling period, and generating the final power instruction in response to the load distribution data meeting the preset distribution balance condition.
  10. 10. The method according to claim 1, wherein the embedding the upper layer optimization objective into a lower layer control process through a mapping from a cost function to a bonus function, and correcting a mapping relation according to execution feedback of the final power instruction, forming a closed-loop optimization operation record, includes: Processing cost function data by using a cost function conversion interface to map the cost function data into an input form of the lower layer rewarding function, and acquiring rewarding function definition corresponding to the input form to determine an initialized collaborative mapping result; based on the initialized collaborative mapping result, acquiring real-time state execution data of the energy storage system, and locally adjusting the lower layer reward function in response to the threshold value of the real-time state execution data being smaller than a preset execution threshold value to obtain adjusted reward function data; acquiring deviation information of the energy storage system based on the adjusted reward function data, and classifying the deviation information through a pre-established support vector machine model to determine a priority sequence of strategy updating under the condition that the deviation information exceeds a preset range; based on the priority sequence updated by the strategy, acquiring a parameter set of an upper-layer optimization target, and carrying out local correction on the parameter set to obtain corrected parameter configuration data in response to the condition that the parameter set and the current execution state do not meet a preset matching condition; Acquiring resource scheduling information of the energy storage system based on the corrected parameter configuration data, and carrying out reallocation processing on the resource scheduling information to determine an optimized resource scheduling scheme; And acquiring target data of the upper layer optimization target and the lower layer optimization target based on the optimized resource scheduling scheme, and processing the target data to generate the final closed-loop optimization operation record.

Description

Voltage stabilization control method under island operation of multi-energy system Technical Field The application relates to the technical field of automatic control of power systems, in particular to a voltage stability control method under island operation of a multi-energy system. Background Along with the rapid development of new energy power generation technology, the multi-energy system is widely applied to scenes such as micro-grids in industrial parks and remote areas by the characteristics of cleanness and high efficiency. The operation modes of the multi-energy system are divided into a grid-connected mode and an island mode, wherein in the grid-connected mode, the system can maintain stable voltage and stable frequency through power support of a large power grid, in the island mode, the system is separated from independent operation of the large power grid, energy balance is maintained only by means of internal distributed power supplies, energy storage devices and load side regulation and control, and the difficulty in voltage stability control is remarkably improved. In the related art, for voltage control of an island system, a simple trigger mechanism based on a fixed threshold is generally adopted, or a module such as timing prediction, feedforward-feedback control, hierarchical optimization and the like is singly used. For example, one common approach is to make adjustments via a PID controller based on real-time voltage deviations, and another approach is to pre-schedule the energy storage system by predicting future short-term voltages. However, in the related art, response matching is misaligned, namely, because of the order-of-magnitude difference between the response time of an energy storage system (millisecond response) and the response time of a thermal power generating unit (second/minute response), the existing static threshold method cannot dynamically adapt to the difference, so that the triggering of a compensation instruction is premature or delayed, and the control instantaneity is affected. The module cooperation is insufficient, the modules such as prediction, compensation and optimization are often designed independently, effective connection is lacking, data flow is not smooth, closed loop control for coping with complex fluctuation is difficult to form, the self-adaptive capacity is weak, when the new energy output is subjected to severe fluctuation and complex meteorological conditions, the controller for fixing parameters and the optimization target cannot be adjusted in a self-adaptive mode, the control effect is reduced, the system stability is difficult to guarantee, and improvement is needed. Disclosure of Invention The application provides a voltage stability control method under island operation of a multi-energy system, which aims to solve the problems of how to realize accurate short-term voltage prediction, how to design a self-adaptive dynamic response matching mechanism to accurately trigger compensation, how to quickly offset voltage deviation through cooperative compensation of feedforward and feedback, how to perform layered cooperative optimization according to the real-time state of the system, finally ensuring the voltage stability of the system and the like in the related technology. The embodiment of the first aspect of the application provides a voltage stability control method under island operation of a multi-energy system, which comprises the steps of acquiring historical voltage data, new energy output data and meteorological data of the multi-energy system under island operation state, carrying out time sequence feature extraction and feature fusion to generate a voltage prediction value sequence in a target time period, calculating a dynamic response matching index for representing response mismatch degree of a heterogeneous energy source unit by combining response time of an energy storage system with response time of a conventional unit based on the voltage prediction value sequence, triggering a voltage compensation mechanism comprising a feedforward compensation path and a feedback compensation path in response to the dynamic response matching index exceeding a preset threshold, calculating a compensation capability evaluation factor based on current charge state, available output power and preset compensation power of the energy storage system after the compensation mechanism is triggered, and selecting an execution path according to the compensation capability evaluation factor, wherein the compensation capability evaluation factor meets preset independent compensation conditions, executing the energy storage independent compensation path in response to the compensation capability evaluation factor does not meet the preset independent compensation conditions, generating a compensation signal by a coordinated power control unit under the control command of the preset independent compensation command, generating a compensation signal by a