CN-121997564-A - Parallel multi-rate simulation method based on time constant network-splitting decoupling

CN121997564ACN 121997564 ACN121997564 ACN 121997564ACN-121997564-A

Abstract

The invention discloses a parallel multi-rate simulation method based on time constant network splitting decoupling, which divides networks according to different time constants of electric equipment of an electric power system, and enables a sub-network to independently solve state variables of the sub-network by extracting and interactively decoupling sub-network simulation information such as a sub-network node admittance matrix and a historical current source column vector, and simultaneously constructs a parallel multi-rate simulation framework aiming at the problem of data interaction opportunity of each sub-network after decoupling, wherein the different time constant networks adopt different time rates for parallel simulation, namely, the sub-networks with different time constants parallelly promote simulation processes in different simulation steps. According to the invention, through two angles of dimension reduction and parallel multi-rate through the network, the calculation complexity is reduced, unnecessary calculation cost is reduced, the calculation resources are fully utilized, and the simulation efficiency of electromagnetic transient state is improved on the premise of ensuring the simulation precision.

Inventors

FENG MOKE
LI YANJUN
WANG HUIZE
XU JIANZHONG
SIMA WENXIA
YANG MING
YUAN TAO
SUN POTAO

Assignees

重庆大学
华北电力大学

Dates

Publication Date: 20260508
Application Date: 20251230

Claims (10)

1. The parallel multi-rate simulation method based on time constant network separation decoupling is characterized by comprising the following steps of: step 1, dividing the different electric equipment of the power system into a plurality of sub-networks; step 2, extracting network information from each sub-network, and performing interactive decoupling of the network information; mutual decoupling is realized through extraction and interaction of simulation information; step 3, designing subnet data interaction time for enabling each subnet to solve in parallel, and constructing a parallel multi-rate simulation framework; step 4, step conversion of different time rate sub-network state variables is carried out; And 5, completing parallel multi-rate electromagnetic transient simulation of the power system by using a parallel multi-rate simulation framework.
2. The parallel multi-rate simulation method based on time constant network separation decoupling according to claim 1, wherein in step 1, different electrical devices of the power system are separated according to the time constant.
3. The parallel multi-rate simulation method based on time constant network splitting decoupling according to claim 2 is characterized in that the formed subnetwork is divided into a fast-changing subnetwork and a slow-changing subnetwork, wherein in the fast-changing subnetwork, the power electronic switching frequency is more than 1kHz, the transient process duration time and the time constant are less than 1ms, the switching frequency is less than 1kHz, and the transient process duration time and the time constant are more than 1ms; The subnets with the transient duration time being greater than the preset time threshold are fast-changing subnets, and the subnets with the transient duration time being less than or equal to the preset time threshold are slow-changing subnets.
4. The parallel multi-rate simulation method based on time constant network splitting decoupling of claim 1, wherein the network information comprises a subnet node admittance matrix and a historical current source column vector.
5. The parallel multi-rate simulation method based on time constant network splitting decoupling according to claim 1, wherein in step 2, each subnet extracts network information to a connection node between subnets through a matrix compression technology, and then broadcasting and matching of the subnet information are performed to realize the interactive decoupling of the network information.
6. The parallel multi-rate simulation method based on time constant network separation decoupling according to claim 1, wherein the step of designing the subnet data interaction time is to endow different simulation step sizes for the fast-changing subnet and the slow-changing subnet so that the solutions of the subnets are performed in parallel.
7. The parallel multi-rate simulation method based on time constant network separation decoupling of claim 6, wherein the simulation step size of the slow-varying sub-network is larger than the simulation step size of the fast-varying sub-network.
8. The parallel multi-rate simulation method based on time constant network splitting decoupling according to claim 1, wherein in step 4, step conversion of different time rate subnet net variables is performed by performing step conversion on node admittance matrix elements extracted from a slow-change subnet and historical current source column vector elements, so that node admittance matrix elements of the slow-change subnet and the fast-change subnet are unified with the step of the historical current source column vector elements; the node admittance matrix element conversion is to perform explicit conversion on element conductance according to step length; Step conversion is carried out on the historical current source column vector elements, namely, the changed conductance is brought into a historical current source solving formula, and a historical current source value after step conversion is obtained.
9. The parallel multi-rate simulation method based on time constant network separation decoupling according to claim 1, wherein in step 5), when parallel multi-rate electromagnetic transient simulation of the power system is completed, after each step of solution is finished, simulation time is accumulated in step length, and fast and slow subnetwork states are synchronously solved at fixed time rate.
10. The parallel multi-rate simulation method based on time constant network separation decoupling according to claim 9, wherein when the simulation time of the fast-change sub-network and the slow-change sub-network reaches the preset simulation time, the simulation is ended, and the parallel multi-rate simulation of the power system is completed.

Description

Parallel multi-rate simulation method based on time constant network-splitting decoupling Technical Field The invention relates to the technical field of power simulation, in particular to a parallel multi-rate simulation method based on time constant network-splitting decoupling. Background Under the national strategy of a novel power system, large-scale wind power and photovoltaic new energy are connected into a power grid, and due to the fluctuation characteristic of the new energy, the electric energy generated by the new energy is generally converted by power electronic equipment and then is integrated into the power grid, and a power electronic converter does not have rotor inertia in a traditional generator set, so that great challenges are brought to the stability of the system frequency. Meanwhile, because of different control strategies of the power electronic equipment, massive power electronic equipment is connected into mutual coupling and interaction effects, complex electromagnetic transient processes such as harmonic oscillation and the like are very easy to occur, and great influence is caused on the safety of the electrical equipment and the stability of a system. The electromagnetic transient simulation of the power system is an important means for knowing complex transient processes such as system harmonic oscillation. However, it is not easy to perform electromagnetic transient simulation on a large number of power electronic devices, and if the high frequency component of the power electronic devices is concerned, the high frequency change of the switching tube in the power electronic devices means that if the transient high frequency component needs to be simulated in a very small step length, the continuously changing switching state makes each time step length need to invert the high-dimensional matrix, which brings great stress to modern computing devices, a large amount of computing resources need to be piled up for real-time simulation, and the simulation time is unacceptably long for offline simulation. In order to improve the simulation efficiency, the main method adopted in engineering is to independently solve the subnetwork after decoupling the system so as to reduce the calculation complexity. Meanwhile, each sub-network after decoupling can be simulated in parallel relatively independently, so that serial delay in a calculation process is eliminated, and the calculation efficiency is improved in multiple. However, the time constant difference of different areas including the power electronic network is larger, the dynamic characteristic difference of each sub-network cannot be reflected only by using a single-rate simulation means, unnecessary calculation cost is increased by adopting small-step simulation to partial sub-networks with larger time constants, so that a multi-rate simulation flow is designed on the basis of a decoupling method in engineering, sub-networks are simulated at different time rates, simulation step sizes are selected according to actual requirements, and calculation resources are fully utilized. Specifically, there are examples of parallel multi-rate algorithms applied to transmission line network separation, delay insertion, multi-region Thevenin equivalent methods, and the like. However, the parallel multi-rate method has the problems of numerical oscillation, poor stability, simulation precision loss and the like caused by switching, and because the area with larger time constant difference still exists in the decoupled network, the area with small time constant still migrates in the subnetwork uses small step-size simulation, and the advantages of the parallel multi-rate are difficult to fully develop. The reason for these problems is that the parallel multi-rate method is a modification of the existing network-splitting decoupling scheme, which cannot be fully adapted to the parallel multi-rate algorithm. Disclosure of Invention The invention aims to provide a parallel multi-rate simulation method based on time constant network splitting decoupling, which comprises the following steps: step 1, dividing the different electric equipment of the power system into a plurality of sub-networks; step 2, extracting network information from each sub-network, and performing interactive decoupling of the network information; mutual decoupling is realized through extraction and interaction of simulation information; step 3, designing subnet data interaction time for enabling each subnet to solve in parallel, and constructing a parallel multi-rate simulation framework; step 4, step conversion of different time rate sub-network state variables is carried out; And 5, completing parallel multi-rate electromagnetic transient simulation of the power system by using a parallel multi-rate simulation framework. In step 1, different electric devices of the electric power system are separated into networks according to time constant. Further, the formed subnetworks a