基于等值信息交换的分布式抗差估计算法

提出了一种基于等值信息交换的分布式抗差状态估计算法。各子系统通过等值计算将自身量测信息浓缩成等值信息,协调层收集各子系统的等值信息计算出边界状态量进而实现分布式状态估计。此外,在分布式算法基础上实现了分布式抗差估计。采用等价权原理将指数型目标函数抗差估计方法转换成变权重的加权最小二乘估计,并基于不动点迭代的方法进行求解。在求解过程中,等值信息随着权重值的变化而不断更新,子系统得以综合全系统信息进行抗差估计。最后,构造了多子系统算例和含不良数据的算例对算法进行测试。测试结果表明分布式抗差估计算法具有很高的计算精度和很好的抗差性能。

A distributed robust estimation algorithm based on equivalent information exchange

A distributed robust state estimation algorithm based on equivalent information exchange is proposed. Each subsystem condenses its own measurement information into equivalent information through equivalent calculation, and the coordinator collects the equivalent information of each subsystem to calculate the boundary state variables, and then realizes the distributed state estimation. In addition, distributed robust estimation is achieved based on the distributed algorithm. Based on the principle of equivalent weight, the robust estimation method with exponential objective function is transformed into weighted least square estimation with variable weight, and the solution is arrived at based on the fixed point iteration method. In the process of solving, the equivalent information is updated with the change of the weight, and the subsystem can synthesize the whole system information for robust estimation. Finally, a multi subsystem example and an example with bad data are constructed to test the algorithm. The test results show that the distributed robust estimation algorithm has high calculation accuracy and good robustness. This work is supported by the National Natural Science Foundation of China (No. 51777069).