An error measurement method for electricity meters based on dynamic test signal modeling with random measurements.

Addressing the challenge of evaluating electricity meter measurement errors caused by nonlinear high-power dynamic loads in smart grids, as well as the issues of long cycle periods and high time consumption associated with traditional m-sequence test signals, this paper proposes an indirect error measurement method based on dynamic test signal modeling using random measurements. Firstly, compressive sensing theory is introduced to construct a structured Orthogonal Pseudo-Random Measurement (OPRM) matrix, generating a dynamic test signal that balances randomness and compactness. This achieves a significant dimensionality reduction of the test sequence while accurately preserving the stochastic fluctuation characteristics of actual dynamic loads. Secondly, a "Run-length Likelihood Function" for dynamic electrical energy is innovatively defined. Leveraging a high-precision synchronous gating control mechanism to eliminate time-domain random truncation effects, a rigorous mapping model is established for tracing dynamic reference energy back to the steady-state reference value. Experimental verification demonstrates that the OPRM model's capability to reflect dynamic errors is highly consistent with that of the traditional m-sequence. However, the single-test duration is drastically reduced from 197 to 49 min, achieving a 75% reduction in time cost. Concurrently, the system's test repeatability is as low as 0.0003%, and the expanded uncertainty is strictly constrained at 0.1442% (inclusion factor k = 2), pushing the comprehensive measurement accuracy of dynamic metering to a new high of better than 0.15%. This work provides core technical support for the agile and high-precision calibration of massive smart electricity meters.