Focus on the problem that Compressive Sensing (CS) measurement can not measure the electrical energy of pseudo random dynamic test signal accurately, in this paper, the spectral sparseness of pseudo random dynamic test signal is firstly analyzed, and dynamic test signal satisfies the measurement condition of compressive sensing is proved. Secondly, the system stable state optimization method is used to establish the deterministic compressive sensing measurement matrix, which meets the RIP (Restricted Isometry Property) condition. Finally, a new compressive sensing measurement methed for the electrical energy of pseudo random dynamic electric energy is proposed. The experimental result shows that the theoretical error of the compressive sensing measurement is superior over traditional sample power electrical energy measurement, and it can measure the pseudo random dynamic electrical energy accurately.