基于模糊熵和非对称贴近度的安全预警研究

针对道床漏泄严重导致道砟电阻和分路电阻发生变化、继而引起"红光带"与分路不良故障的问题,以ZPW-2000 A轨道电路为例,建立预警指标体系。采用模糊模式识别法对轨道电路状态进行预警。首先,计算各预警指标的隶属度函数和隶属度矩阵,利用模糊熵理论确定各预警指标的权重;其次,采用非对称贴近度原则对预警决策进行识别,以判决系统所属安全等级;最后,以长沙电务段新开铺站为例进行实例验证。结果表明,基于模糊非对称贴近度的轨道电路分路不良预警是可行的,其对轨道电路的安全等级划分更精细,故障预警更精确。

On the safety warning to be made based on the fuzzy entropy and asymmetric proximity

This paper is to probe a new approach to the safety warning delivery based on the fuzzy entropy and asymmetric proximity because of the red-light-strap and poor shunting problem likely to be caused by the leakage of the ballast for ZPW-2000 A track circuits.Just for the said purpose,we have made an exploration and put forward a decision to assess its safety warning system based on the fuzzy asymmetric closeness principle.In so doing,we have first of all analyzed the system features of ZPW-2000 A track circuit,and established an equivalent model for linking the four-terminal networks of the ZPW-2000 A track circuit.And,then,we have worked out and chosen the appropriate warning indicators constituting the warning system and settled down the warning limit boundaries of the area under the early warning.It is also on the aforementioned basis that we have settled down the warning indicators and the theory of fuzzy entropy to calculate the distribution weights of each particular warning indicator.In addition,we have collected the real-time data information of all the early warning indicators in the field,and uploaded the data concerned into the computer of the malfunction warning system.And,after that,we have applied the fuzzy pattern recognition method to processing the data to make necessary decisions intelligently and comprehensively according to the actual need so as to define and determine the membership function of each early warning indicator and the fuzzy comprehensive vectors.On the above mentioned basis,it would be possible for us to work out the asymmetric closeness in accordance with each corresponding early warning indicator via the principle of asymmetric closeness degree.Thus,the appropriate handling measures have been taken to judge the system fault level state of the system and deliver the corresponding warning signals for the electrical service operators to use.The assessment results we have managed to gain show that it is possible to use the asymmetric closeness principle to judge the data of the fuzzy pattern recognition,whose processing steps can be categorized into more accurate data,such as into the fault status levels for the track circuits to describe the status of the track circuit and warning of the fault levels according to the warning signals to be processed.Moreover,the assessment results we have done prove to be well in accord with the actual status-in-situ.Thus,the present research can be expected to provide necessary conditions for the track circuit working safety and efficiency.