过失误差

一种改进的同步侦破和识别过失误差的方法

An Improved Method for Detecting and Identifying Gross Errors

介绍了一种识别、校正稳态过程过失误差的混合方法，该方法基于节点检测法和整体检测法的相关调整量。

A combinatory method for detecting , identifying gross errors in steady state process is described in this paper .

采用移动中值滤波器（MM）消除过失误差，然后应用小波变换滤除随机误差。

In this paper , moving median filter is used to remove gross errors and wavelet transformation is used to eliminate random errors .

首先对实际生产中的原始数据，经过过失误差剔除及滤波处理后得到一套训练数据和校验数据组成训练样本，然后采用BP神经网络进行训练，得到组分纯度的非参数模型。

First by filter process of practical primal data , the training samples are gained . Then the nonparametric model on component purity is developed by use of BP neural network .

针对未知概率分布、含过失误差的小样本数据识别问题，提出了基于GM（1,1）模型残差和残差平方和的判别方法；

Aiming at solving the distinguishing problem of gross error for the small samples whose distribution is unknown , a new distinguishing algorithm based on the residual error of the GM ( 1,1 ) model is proposed .

DRS软件可以连续判断生产装置操作是否平稳并对处于稳态下的测量数据进行校正，还可以对测量数据中含有的过失误差进行正确的侦破和定位。

DRS is capable of detecting the steady state of operation continuously , reconciling data under steady state , and detecting and identifying the gross errors in the measurements .

但由于仪表测量的现场数据大都存在随机误差，甚至过失误差，使得DCS系统采集到的数据难以真实反映工艺过程，无法直接用于决策与指导中。

However , most of the field data measurement instrument with random errors , even gross error , making the data collected by DCS system is difficult to truly reflect the process , can not be directly used for decision-making and guidance .

介绍了一种用于过失误差侦破和校正改进的MT-NT算法。

An improved MT-NT combined method used in gross error detection and data reconciliation in process data is described .

基于时间冗余度侦破过失误差的修正系数法

Correction coefficient method for gross error detection based on temporal redundancy

一种基于函数型连接神经网络的过失误差检测方法

An Unpremeditated Error Detecting Method Based on the Functional Connected Neuro-Network

中国历史探空资料中的一些过失误差及偏差问题

Some gross errors and biases in china 's historical radiosonde data

基于灰色模型残差的过失误差判别方法

Distinguishing Algorithm of Gross Error Based on Residual Error of Grey Model

用递阶神经网络进行多过失误差检测的方法

Detecting Multiple Simultaneous Gross Errors via Hierarchical Artificial Neural Networks

用于含过失误差数据稳态检测的改进滤波法

Improved filtering method for steady state tests with measurements containing gross errors

判断观测数据过失误差的方法

A method for judging gross error of observating data

原油常压蒸馏塔稳态过程的数据协调与过失误差检测

Data coordination and lapse error detection for steady-state process of crude atmospheric tower

测量仪表过失误差的单结点识别方法

Approach to gross error identification of instruments based on residual of single node

数据协调与过失误差侦破的鲁棒估计同步方法

Simultaneous gross error detection and data reconciliation based on the robust estimation principle

过程工业测量数据中过失误差的侦破与校正

Gross error detection and data reconciliation in process industries

采用两步矩阵投影与广义似然比法相结合的算法进行数据协调和过失误差的侦破、识别。

Generalized likelihood ratio method ( GLR ) was used to detect gross errors .

主要研究内容包括过失误差侦破、数据校正与参数估计等。

The main contents include gross errors detection , data reconciliation and parameter estimation .

稳态系统的过失误差识别

Gross Error Identification for Steady State System

本文首先对传统的数据协调方法和过失误差侦破方法进行了研究。

In this thesis , traditional data reconciliation and gross errors detection methods are studied firstly .

过失误差称异常值，处理时应剔除。

Unexpected errors which are , in fact , abnormalities should be eliminated in data processing .

在计量测试领域中，离群值就是含粗差即粗大误差或过失误差的测量值。

The outliers are the measured values including gross errors or fault errors in the metrology field .

实例研究表明，这种方法能够有效地提高多过失误差共存时的侦破能力。

The results show that the new method can improve the detecting power of the muhiple gross errors detection greatly .

该方法无需量测信息的先验知识，能够准确地获得实时导航信息，并且在小噪声扰动时处于稳定，对于过失误差有较强的鲁棒特性。

The method has no need for the transcendental information of measured data and has good real-time characteristic , stability and robustness .

按误差特性分为三类即随机误差、系统误差和过失误差。

According to their specific characteristics , errors can be classified into three categories & random error , system error and unexpected error .

按照数学处理方法的不同，仿真系统中的误差又可以分为系统误差、随机误差和过失误差三类。

Owing to different methods in mathematical process , the errors in stimulation system consist of system error , random error and defect error .

用该方法对某炼油厂的气体分离装置的丙烯浓度软测量建模数据进行过失误差侦破，取得了很好的效果。

The method was used to detect gross errors in modeling data for a propylene concentration soft sensor in a refinery factory and good results were obtained .