Michela Mulas

On the topological modeling and analysis of industrial process data using the SOM

In this paper, we overview and discuss the implementation of topology-based approaches to modeling and analyzing industrial process data. Emphasis is given to the representation of the data obtained with the self-organizing map (SOM). The methods are used in visualizing process measurements and extracting relevant information by exploiting the topological structure of the observations. Benefits of the SOM with industrial data are presented for a set of process measurements measured in an industrial gas treatment plant. The practical goal is to identify significant operational modes and most sensitive process variables before developing an alternative control strategy. The results confirmed that the SOM-based approach is capable of providing valuable information and offers possibilities for direct application to other process monitoring tasks.

Wastewater treatment modelling in practice: a collaborative discussion of the state of the art

Three consulting teams conducted independent modelling projects for three different wastewater treatment plants ranging in size from approximately 113,800 m(3)/d (30 mgd) to 530,000 m(3)/d (140 mgd), in different parts of the world (USA and Finland). The plants have different treatment objectives ranging from nitrification and partial denitrification (nitrate plus nitrite <8.7 mg/L) to enhanced nutrient removal (total nitrogen <3 mg/L, total phosphorus <0.3 mg/L). Commonly-used models were applied in the case studies, including ASM3 (using the GPS-X simulator), New General (using GPS-X), Dold (using BioWin), and a variation of the Dold model methanol degradation capabilities (NGmeth within GPS-X). The authors compare and contrast the modelling approaches taken, including calibration and validation approaches, sensitivity analyses, and the application of results to full-scale studies, designs and operations. Despite several differences between the approaches, there are many similarities which are discussed in light of the IWA draft uniform protocol for activated sludge modelling. The authors also discuss current modelling limitations and offer suggestions to improve the state of the art.

On the topological analysis of industrial process data using the SOM

Abstract In this paper, we overview and discuss the implementation of some topological approaches to modeling and analyzing industrial process data. The discussed methods are used in visualizing process measurements and extracting information by exploiting the metric structure of the observations. Emphasis is given to modeling with the Self-Organizing Map (SOM). The SOM is a standard method for dimensionality reduction and vector quantization equipped with many displays for visualization. Some of the possibilities of the SOM with process data are discussed by exploring measurements from a full-scale gas treatment plant where the goal is to identify important operational modes and sensitive process variables before developing an alternative control strategy.

Data derived analysis and inference for an industrial deethanizer

Abstract Abstract In this paper, we present an application of data derived approaches for analyzing and monitoring an industrial deethanizer column. The discussed methods are used in visualizing process measurements, extracting operational information and designing an estimation model. Emphasis is given to the modeling of the data obtained with standard paradigms like the Self-Organizing Map (SOM) and the Multi-Layer Perceptron (MLP). The SOM and the MLP are classic methods for nonlinear dimensionality reduction and nonlinear function estimation widely adopted in process systems engineering; here, the effectiveness of these data derived techniques is validated on a full-scale application where the goal is to identify significant operational modes and most sensitive process variables before developing an alternative control scheme.

Optimized Control Structure for a Wastewater Treatment Benchmark

Abstract In this paper, we define and implement the design of an optimized control structure for the activated sludge process given as COST/IWA benchmark simulation model No.1. Emphasis is given to the identification of controlled variables that contribute to minimize economic costs while the effluent requirements are met. This is achieved considering the self-optimizing procedure as reference method for the controlled variables selection. The proposed optimal control strategy consists of multivariable PID loops which manipulate the airflow rate in the aerobic basins, the nitrate and sludge recirculation flows and the waste sludge flow proportionally to the influent flow such that the overall cost function is minimized. Dynamic simulations validate the resulting optimized controller structure, showing that minimal costs can be achieved.