论文部分内容阅读
实时准确的无线链路质量预测是保证智能电网厂、站区域通信链路可靠性的必要信息。无线链路质量信噪比时间序列所表现出的非线性和非平稳随机性的叠加是影响预测准确性的主要因素。为此,提出一种通信链路可靠性置信区间预测方法,通过对无线链路质量的信噪比序列近似解耦处理,将其分解为非线性序列和非平稳随机序列,采用小波神经网络建立信噪比非线性序列和非平稳随机方差序列的预测模型,并用预测结果计算通信链路可靠性置信区间上、下界。最后,在实际的智能电网环境中验证了所提出的算法和结果。
Real-time and accurate wireless link quality prediction is necessary to ensure the reliability of communication links between smart grid plants and stations. The superposition of nonlinear and nonstationary randomness exhibited by the SNR sequence of wireless link quality is the main factor affecting the prediction accuracy. Therefore, this paper proposes a confidence interval prediction method for reliability of communication links. Through the approximate decoupling of signal-to-noise ratio of wireless link quality, this method is decomposed into nonlinear sequence and non-stationary random sequence, Signal-to-noise ratio non-linear sequence and non-stationary random variance sequence prediction model, and use the prediction results to calculate the upper and lower bounds of the reliability of the communication link. Finally, the proposed algorithm and results are verified in a real smart grid environment.