A low-frequency band multi-sine signal design method for electric power system state space model identification

Abstract

The invention discloses a low-frequency-band multi-sine signal design method for electric power system state space model identification. The method comprises the following steps of determining each harmonic component amplitude, each harmonic component frequency and a sampling length of a low-frequency-band multi-sine signal x (t); Then, defining a wave crest factor of the low-frequency-band multi-sine signal x (t); converting a wave crest factor minimization problem into a wave crest factor minimization problem; gradually increasing the p value of the lp norm, and solving a column vector pp composed of the harmonic wave phases (please see the formula in the specification), so as to minimize the lp norm of the low-frequency band multi-sine signal x (t); And finally, solving the problem by adopting an algorithm, wherein p is set as 4, 8, 16, 32, 64, 128, 256, 512 ... in sequence. For each set p value, adopting a Gauss Newton method in combination with Levenberg-Marquardt algorithm to solve a column vector pp, so that the lp norm of x (t) is minimum; And when p is set to be 512, the approximation requirement can be met, so that the low-frequency-band multi-sine signal x (t) meeting the requirement is solved. The input signal designed by the method disclosed by the invention simultaneously meets the time domain waveform amplitude limiting requirement and the frequency domain energyconcentration requirement.

Description

technical field [0001] The invention relates to the cross technical field of power system identification and signal processing, in particular to a design method for low-frequency multi-sine signals used for power system state space model identification. technical background [0002] With the development of the interconnected power grid, the problem of low-frequency oscillation in the power system in the frequency range of 0.1 to 2.5 Hz is prominent, which needs to be solved by installing a power system low-frequency oscillation controller. The design of power system low-frequency oscillation controller depends on the power system state-space model. In the actual power system engineering design, the problem of state-space model identification needs to be solved first. In order to ensure the safe and stable operation of the power system during the identification process, practical projects usually use small disturbance signals to excite the power system, then collect the power...

AI Technical Summary

Problems solved by technology

The energy of the two types of signals is large in the 0.1 to 2.5 Hz frequency band concerned by low-frequency oscillation, but there is a certain amount of energy outside the concerned frequency band, which leads to insufficient concentration of the input signal energy when the state space model identification of the power system is performed, and the system outputs a response signal The signal-to-noise ratio is reduced, which affects the accuracy of power system state-space model identification

Images