CN-122026399-A - Thermal power-energy storage-super capacity cooperative frequency modulation system and method based on multi-terminal flexible direct current

CN122026399ACN 122026399 ACN122026399 ACN 122026399ACN-122026399-A

Abstract

The invention discloses a thermal power-energy storage-super capacity cooperative frequency modulation system and method based on multi-terminal flexible direct current, and relates to the technical field of control of power systems. The method comprises the steps of collecting system frequency deviation signals of an alternating current power grid connected with a multi-terminal flexible direct current power transmission system, decomposing the system frequency deviation signals based on a preset multi-characteristic time scale, generating initial power reference instructions corresponding to all resources, carrying out forward prediction on equipment states after the initial power reference instructions are executed, generating power correction instructions and power redistribution suggestion vectors if the prediction states violate preset physical constraints, carrying out dynamic matching in a simulated continuous bidirectional auction market based on the corrected power demand instructions and adjustment capacity quotations generated by all resource twinning models, determining optimal power instructions finally issued to all physical resource actuators, and further controlling corresponding power output by a thermal power unit, an energy storage converter and a super-capacity converter, so that the cooperative adjustment of the power grid frequency is realized.

Inventors

ZHAO YUANYOU
YAN YU
WANG FEI
YANG DONGSHENG
MIN QIANG
TAN JIN
ZHANG JIANSHENG
LI MENG
LIU YONGJIANG
YANG SHUANG

Assignees

大唐贵州发耳发电有限公司

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (10)

1. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current is characterized by comprising the following steps of: s1, collecting system frequency deviation signals of an alternating current power grid connected with a multi-terminal flexible direct current power transmission system in real time; S2, inputting the system frequency deviation signal into a virtual physiological filter bank, decomposing the system frequency deviation signal into an ultrafast component corresponding to the super capacity resource, a fast component corresponding to the energy storage resource and a slow component corresponding to the thermal power resource based on a plurality of preset characteristic time scales, and generating initial power reference instructions corresponding to the resources; s3, inputting the initial power reference instruction into a digital twin constraint pre-evaluation model corresponding to each resource, wherein the digital twin constraint pre-evaluation model carries out forward prediction on the state of equipment after the initial power reference instruction is executed, and if the predicted state violates a preset physical constraint, a power correction instruction and a power redistribution suggestion vector are generated; s4, dynamically matching in a simulated continuous bidirectional auction market based on the power demand instruction corrected by the digital twin constraint pre-evaluation model and the adjustment capability quotation generated in real time by each resource twin model according to the current running state and the economic index of the resource twin model so as to determine an optimized power instruction finally issued to each physical resource executor; And S5, issuing the optimized power instruction through a converter station controller of the multi-terminal flexible direct current system to control the thermal power unit, the energy storage converter and the super-capacity converter to output corresponding power, so as to realize cooperative adjustment of the power grid frequency.
2. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current according to claim 1, wherein in step S2, the virtual physiological filter bank adopts a group of Gamma-tone filters with different center frequencies and bandwidths, and the impulse response function is defined as: , wherein, In order to be the center frequency of the response characteristic, For the resource type index to be used, Respectively corresponding to super capacity, energy storage and thermal power resource types; Is in combination with A proportional bandwidth parameter for achieving a natural separation and resonance mapping of the frequency deviation signal on a plurality of time scales of milliseconds, seconds, minutes; Is a normalization constant; For the filter order, take ; Is a phase offset.
3. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current according to claim 1, wherein in step S3, the digital twin constraint pre-evaluation model comprises: The super-capacity twin model has core constraints including terminal voltage, internal temperature and maximum allowable current; the core constraint of the energy storage twin model comprises a charge state, a health state, a charge-discharge multiplying power and a temperature; the core constraint of the thermal power twin model comprises climbing rate, minimum technical output and main steam pressure; the power correction instruction is based on constraint violations The dynamic generation is carried out, and the calculation mode is as follows: , wherein, For the power correction amount, And In order to adjust the coefficient of the power supply, Is an initial power reference instruction, the constraint violation degree Quantization is performed by the following formula: , is the first The constraint weight coefficients of the individual state parameters, Is the first Class resource of The predicted value of the individual state parameter(s), 、 Respectively the first Class resource of The minimum and maximum values of the individual state parameters, In order to operate at the maximum value, Is the first The total number of critical constraints for the class resource.
4. The thermal power-energy storage-super capacity collaborative frequency modulation method based on multi-terminal flexible direct current according to claim 3, wherein the power redistribution advice vector is dynamically calculated according to the characteristic time constant matching degree and real-time available capacity between resources, in particular through a redistribution matrix The implementation is defined as: , wherein, Representing slave resources Attenuated power advice allocation to resources Is used in the ratio of (a), 、 Is a characteristic time constant of the resource and, Is a resource Is used to determine the real-time available capacity of the system, In order to adjust the coefficient of the degree of matching, Is a natural constant.
5. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current as claimed in claim 1, wherein in step S4, the adjustment capability quotation From the basic marginal cost Cost of equipment loss And operating point offset cost The composition is as follows: Wherein the equipment costs Calculating a life-time decay gradient predicted based on the digital twin model, the operating point offset cost And has positive correlation with the current state of charge or the degree of deviation from the optimum value of the operating condition.
6. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current according to claim 5, wherein in step S4, the dynamic clearing process of the continuous bidirectional auction market is described by a price dynamic equation, which is defined as: , wherein, Is a time scale band The price of the product is clear in the market, In order to converge the rate coefficient, For the total power demand of the time scale band, Is a resource At the position of The following feed function.
7. The thermal power-energy storage-super capacity cooperative frequency modulation method based on multi-terminal flexible direct current according to any one of claims 1 to 6, wherein the method further comprises: carrying out online identification and updating on parameters of the digital twin constraint pre-evaluation model by using a recursive least square method; based on historical frequency modulation performance indexes, the gain coefficient of the virtual physiological filter bank and key parameters of the auction market are periodically optimized through a gradient descent method so as to minimize a comprehensive objective function comprising frequency deviation, adjustment cost and equipment loss.
8. A multi-terminal flexible dc based thermal power-energy storage-super capacitor co-modulation system for implementing the method of any one of claims 1-7, comprising: the signal acquisition and processing unit is used for acquiring a system frequency deviation signal of an alternating current power grid connected with the multi-terminal flexible direct current power transmission system and the running state of the multi-terminal flexible direct current system in real time; A cooperative fm controller comprising: The virtual physiological resonance decomposition module is used for decomposing the system frequency deviation signal into an ultrafast component corresponding to the super capacity resource, a fast component corresponding to the energy storage resource and a slow component corresponding to the thermal power resource based on a plurality of preset characteristic time scales, and generating an initial power reference instruction corresponding to each resource; The digital twin sentinel pre-evaluation module is used for receiving the initial power reference instruction, carrying out forward prediction on the equipment state after the initial power reference instruction is executed, and generating a power correction instruction and a power redistribution suggestion vector if the predicted state is in violation of a preset physical constraint; The market bidding dynamic allocation module is used for dynamically matching in a simulated continuous bidirectional auction market based on the power demand instruction corrected by the digital twin constraint pre-evaluation model and the adjustment capability quotation generated in real time by each resource twin model according to the current running state and the economic index of the resource twin model so as to determine the optimized power instruction finally issued to each physical resource executor; And the execution unit comprises a thermal power unit regulator, an energy storage converter controller and a super-capacity converter controller and is used for receiving and executing the optimized power instruction.
9. The thermal power-energy storage-super capacity cooperative frequency modulation system based on multi-terminal flexible direct current according to claim 8, wherein the cooperative frequency modulation controller is connected with a station-level controller of the multi-terminal flexible direct current system through a communication network, and the station-level controller receives the optimized power instruction and converts the optimized power instruction into a corresponding converter station modulated wave signal or a direct power control instruction.
10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements a thermal power-energy storage-super capacity collaborative frequency modulation method based on multi-terminal flexible direct current according to any of claims 1-7.

Description

Thermal power-energy storage-super capacity cooperative frequency modulation system and method based on multi-terminal flexible direct current Technical Field The invention relates to the technical field of power system control, in particular to a thermal power-energy storage-super capacity cooperative frequency modulation system and method based on multi-terminal flexible direct current. Background With the promotion of the 'double carbon' target, high-proportion renewable energy sources represented by wind power and photovoltaics are accessed into a power grid in a large scale. The strong intermittence and randomness of these power supplies, and their low inertia, weak damping characteristics brought by the grid connection of the power electronics, lead to an increasingly severe problem of frequency stabilization of the grid. The traditional power system mainly depends on the rotational inertia of a synchronous generator set (such as thermal power and hydroelectric power) and primary frequency modulation standby maintenance frequency to be stable, but the response speed (in the order of seconds to tens of seconds) is difficult to completely adapt to rapid frequency change caused by new energy fluctuation, and frequent deep adjustment can aggravate the abrasion of the generator set and reduce the economic operation efficiency. To address the challenges described above, the introduction of fast frequency response resources based on power electronics has become a necessary choice. The battery energy storage system has second-level power response capability, the super capacitor can provide millisecond-level instantaneous power support, and the super capacitor and the thermal power unit with large capacity and sustainable regulation property are combined to form an ideal frequency support system covering all time scales from millisecond to minute. The multi-terminal flexible direct current transmission system is used as a new generation direct current transmission technology, has the advantages of flexibility and controllability, capability of supplying power to a passive network, capability of rapidly and independently controlling active/reactive power and the like, and provides a high-efficiency and controllable aggregation and power exchange platform for various geographically dispersed heterogeneous frequency modulation resources. However, how to realize efficient coordination and optimal coordination of three dynamic characteristics, namely thermal power, energy storage and super capacitor, and economic attribute and physical constraint disparate resources through a multi-terminal flexible direct current system is a core technical problem to be solved in the current field. At present, in the aspect of control architecture and signal distribution, the existing technical scheme is mainly based on power distribution of fixed proportion or simple droop coefficient. And generating a total power instruction by measuring alternating current frequency deviation, and distributing the total power instruction to connected resources according to a preset fixed proportion. The method fails to fully consider inherent time response characteristics (fastest super capacity, inferior energy storage and slowest thermal power) of different resources, so that the resources with the fastest response can bear slow variable components unsuitable for energy capacity, and the resources with slow response cannot effectively inhibit initial frequency dip, so that the resource characteristics and frequency modulation requirements are mismatched in time scale, and the overall frequency modulation efficiency cannot be optimized. Secondly, at the level of resource management and constraint processing, existing schemes typically employ "post-correction" or "hard clipping" strategies. The stored energy charge state is monitored in real time, and the power is forcedly withdrawn or reversed when the stored energy charge state is out of limit. The above mode belongs to passive protection, and secondary loss or impact of frequency modulation power can be caused during constraint triggering, so that secondary oscillation of frequency can be caused. In the collaborative optimization level, the existing method is mainly focused on optimizing the technical performance, the collaborative control strategy only aims at minimizing the frequency recovery speed and deviation, the loss cost of frequent actions to the life of the thermal power generating unit and the energy storage cycle times is ignored, the overall frequency modulation economy is poor, and the marketized and sustainable operation mode is difficult to support. Therefore, a brand new cooperative frequency modulation method is needed to be invented, so that core problems of time scale matching, operation constraint prospective avoidance, technical economy multi-objective cooperation and the like can be fundamentally solved, the potential of multi-terminal flexible direct current system fo