Multivariate data / Mathematical and statistical background / Exploring multivariate data graphically / Exploratory factor analysis / Correspondence analysis / Multidimensional scaling / Cluster analysis / Latent class analysis / Discriminant function analysis / Multivariate dependencies / Classical multivariate inference / Longitudinal data / Problems of measurement error / Confirmatory factor analysis and structural equation models.

Applied Multivariate Data Analysis

This paper surveys and discusses the application of data-derived soft-sensing techniques in biological wastewater treatment plants. Emphasis is given to an extensive overview of the current status and to the specific challenges and potential that allow for an effective application of these soft-sensors in full-scale scenarios. The soft-sensors presented in the case studies have been found to be effective and inexpensive technologies for extracting and modelling relevant process information directly from the process and laboratory data routinely acquired in biological wastewater treatment facilities. The extracted information is in the form of timely analysis of hard-to-measure primary process variables and process diagnostics that characterize the operation of the plants and their instrumentation. The information is invaluable for an effective utilization of advanced control and optimization strategies. We review data-derived soft-sensors proposed for biological wastewater treatment.Increased amount of measured process data has made data-driven modelling attractive.A general guideline for the data-derived soft-sensor development is presented.Artificial intelligence and multivariate statistical methods are popular in soft-sensor design.Popularity of the on-line prediction applications has increased during recent years.

Review: Data-derived soft-sensors for biological wastewater treatment plants: An overview

Level 1 reliability methods have been internationally accepted as the basis for development of the new generation of geotechnical design codes. A key requirement of this design approach is the identification and quantification of uncertainties associated with the geotechnical design under consideration. This paper presents four load test databases from South Africa for driven piles in noncohesive soils (29 tests), bored piles in noncohesive soils (33 tests), driven piles in cohesive soils (59 tests), and bored piles in cohesive soils (53 tests). The capacity model factor is defined as the ratio of the interpreted capacity (Chin-Davisson approach) and the predicted capacity (static pile design formula). The uncertainty in the capacity model factor is modeled as a lognormal random variable. The model factor statistics reported in this study are required for reliability-based ultimate limit state design. The uncertainty in the load-settlement behavior is characterized by fitting measured load-settlement data to a hyperbolic equation and then normalizing the hyperbolic curve with the interpreted capacity. The resulting exercise reduces uncertainties in a set of nonlinear continuous curves to uncertainties in two hyperbolic curve-fitting parameters. This approach is practical and grounded realistically on the load test database with minimal assumptions. The hyperbolic parameter statistics reported in this study are required for reliability-based serviceability limit state design.

Characterization of Model Uncertainty in the Static Pile Design Formula

Assessment of model uncertainties for structural resistance

The purpose of STATISTICS FOR BUSINESS AND ECONOMICS is to give students, primarily those in the fields of business, management and economics, a conceptual introduction to the field of statistics and its many applications. The text is applications oriented and written with the needs of the non-mathematician in mind. The mathematical pre-requisite is knowledge of algebra.

Applications of data analysis and statistical methodology are an integral part of the organization and presentation of the material in the text. The discussion and development of each technique are presented in an application setting, with the statistical results providing insights to problem solution and decision making.

Although the book is applications oriented, care has been taken to provide sound methodological development and to use notation that is generally accepted for the topic being covered. Hence, students will find that this text provides good preparation for the study of more advanced statistical material. A revised and updated bibliography to guide further study is included as an appendix.

商务与经济统计 = Statistics for Business and Economics

Evaluating water quality data for outliers is a good quality control/quality assessment procedure whether the data are used for monitoring or for modeling. Often water quality data are correlated, e.g., carbonaceous biochemical oxygen demand (CBOD) has some correlation with N H3 . Univariate methods for identifying outliers do not consider the correlation between variables and may identify too many data points as outliers or miss observations which have extreme ratios between variables, e.g., a raw wastewater sample with relatively low CBOD but high N H3 . Testing for outliers using multivariate methods such as the Mahalanobis distance, Jackknife distance, p -values, or Hadi’s automatically incorporates the correlation or covariance between variables and is fundamentally more correct. Such multivariate methods can better identify potential outliers and avoid eliminating valid data.

R. B. Robinson

Papers

Identifying outliers in correlated water quality data