Software sensors

Constrained Recursive Input Estimation of Blending and Mixing Systems

Blending, mixing processes are often supported by advanced process control systems to maximise margins from available component and heat streams. Since these model-based solutions require accurate and reliable data, in weakly instrumented processes, the unknown inlet concentrations and temperatures should be estimated based on the measured outflows. This work presents a method for the reliable estimation of multiple input variables of process units. The key idea is that the input estimation problem is formulated as a constrained recursive estimation task. The applicability of the method is illustrated based on a benchmark model of a blending system. The performance of the method is compared to the moving window and Kalman Filter based solutions. The results show the superior performance of the proposed method and confirm that the apriori knowledge-based constraints improve the robustness of the estimates.

Abonyi J.: „Constrained Recursive Input Estimation of Blending and Mixing Systems”, Chemical Engineering Transactions, 76, 727-732

Operating regime model based multi-objective sensor placement for data reconciliation

Although the number of sensors in chemical production plants is increasing thanks to the IoT revolution, it is still a crucial problem what to measure and how to place the sensors as such the resulted sensor network be robust and cost-effectively provide the required information. This problem is especially relevant in flexible multi-purpose, multi-product production plants when there are significant differences among the operating regions. The present work aims the development of a sensor placement methodology that utilizes the advantages of local linear models. Realizing the often conflicting nature of the key objectives of sensor placement, the problem is formulated as a multi-objective optimization task taking into consideration the cost, estimation accuracy, observability and fault detection performance of the designed networks and simultaneously seeking for the optimal solutions under multiple operating regimes. The effectiveness of the Non-dominated Sorting Genetic Algorithm-II (NSGA-II)-based solution of the defined problem is demonstrated through benchmark examples.

Dörgő Gy., Haragovics M., Abonyi J.: "Operating regime model based multi-objective sensor placement for data reconciliation", 29th European Symposium on Computer Aided Process Engineering, Netherlands, Eindhoven, 2019 June 16-19.

Soft Sensors Special Issue

Post date: Aug 30, 2018 8:1:00 PM


We are editing a special issue related to software sensors.

Please forward this link to researchers potentially interested insubmitting a paper.

Slide to promote special issue in conference

http://www.mdpi.com/journal/sensors/special_issues/Soft_Sensors

Software sensor for activity-time monitoring and fault detection in production lines

Industry 4.0-based human-in-the-loop cyber-physical production systems are transforming the industrial workforce to accommodate the ever-increasing variability of production. Real-time operator support and performance monitoring require accurate information on the activities of operators. The problem with tracing hundreds of activity times is critical due to the enormous variability and complexity of products. A software-sensor-based activity-time and performance measurement system to handle this problem is proposed.

To ensure a real-time connection between operator performance and varying product complexity, fixture sensors and an indoor positioning system (IPS) were designed and this multi sensor data merged with product-relevant information.

The proposed model-based performance monitoring system tracks the recursively estimated parameters of the activity-time estimation model. As the estimation problem can be ill-conditioned and poor raw sensor data can result in unrealistic parameter estimates, constraints were introduced into the parameter-estimation algorithm to increase the robustness of the software sensor.

The applicability of the proposed methodology is demonstrated on a well-documented benchmark problem of a wire harness manufacturing process. The fully reproducible and realistic simulation study confirms that the indoor positioning system-based integration of primary sensor signals and product-relevant information can be efficiently utilized in terms of the constrained recursive estimation of the operator activity.

Multivariate Statistical and Computational Intelligence Techniques for Quality Monitoring of Production Systems

The ISO 9001:2008 quality management standard states that organizations shall plan and implement monitoring, measurement, analysis and improvement processes to demonstrate conformity to product requirements. According to the standard, detailed analysis of data is required for this purpose. The analysis of data should also provide information related to characteristics and trends of processes and products, including opportunities for preventive action. The preliminary aim of this chapter is to show how intelligent techniques can be used to design data–driven tools that are able to support the organization to continuously improve the effectiveness of their production according to the Plan—Do—Check—Act (PDCA) methodology. The chapter focuses on the application of data mining and multivariate statistical tools for process monitoring and quality control. Classical multivariate tools such as PLS and PCA are presented along with their nonlinear variants. Special attention is given to software sensors used to estimate product quality. Practical application examples taken from chemical and oil and gas industries illustrate the applicability of the discussed techniques.

Kulcsár, T., Farsang, B., Németh, S., & Abonyi, J. (2016). Multivariate statistical and computational intelligence techniques for quality monitoring of production systems. In Intelligent Decision Making in Quality Management (pp. 237-263). Springer, Cham.

Data-driven Development and Maintenance of Soft-Sensors

Product quality related process variables have significant role in advanced process control (APC). Online analyzers and software sensors can provide accurate and timely information for APC systems. In this paper we give an overview of data based soft-sensor development. We show that soft-sensor models of APC require maintenance and demonstrate that statistical quality control (SQC) techniques can be effectively used to automatize the related fault detection tasks.

János Abonyi, Barbara Farsang and Tibor Kulcsar: Data-driven Development and Maintenance of Soft-Sensors, IEEE 12th International Symposium on Applied Machine Intelligense and Informatics, Herlany, Slovakia, 2014, 239-244.

Statistical Process Control Based Performance Evaluation Of On-Line Analysers

On-line analyzers can provide accurate and timely information for process control and monitoring. Statistical Process Control (SPC) techniques can be effectively utilized to support the development and maintenance of these tools. The D6299- 10 ASTM standard details how on-line analyzers should be validated. The applicability of this standard is demonstrated through the analysis of industrial data collected from an on-line gas chromatograph. The results confirm that automatized SPC can effectively improve the reliability of advanced process control systems.

Tibor Kulcsár, János Abonyi, Statistical Process ControlBased Performance Evaluation Of On-Line Analysers, Hungarian Journal Of Industry And Chemistry, Veszprém, 41, 77-82, 2013.

Semi-mechanistic Models for State-Estimation – Soft Sensor for Polymer Melt Index Prediction

Nonlinear state estimation is a useful approach to the monitoring of industrial (polymerization) processes. This paper investigates how this approach can be followed to the development of a soft sensor of the product quality (melt index). The bottleneck of the successful application of advanced state estimation algorithms is the identification of models that can accurately describe the process. This paper presents a semi-mechanistic modeling approach where neural networks describe the unknown phenomena of the system that cannot be formulated by prior knowledge based differential equations. Since in the presented semi-mechanistic model structure the neural network is a part of a nonlinear algebraic-differential equation set, there are no available direct input-output data to train the weights of the network. To handle this problem in this paper a simple, yet practically useful spline-smoothing based technique has been used. The results show that the developed semi-mechanistic model can be efficiently used for on-line state estimation.

Effects of Catalyst Activity Profiles on the Operating Conditions of an Industrial Polymerization Reactor

The aim of this paper is to analyze how different catalyst activity profiles influence the operating strategies of industrial polymerization reactors. Since the catalyst activity profiles may change considerably in the range of the residence-time of the reactor, the properties of the produced polymer powder are the averages of ones produced by catalyst particles with various activities. Hence, the residence-time distribution of the polymer phase has an effect on the product properties as well. Based on these phenomena a new method proposed to transform information from laboratory reactor into a form which can be used to estimate the operation condition of industrial reactor. The whole approach is demonstrated in case of the production of high-density polyethylene.

S. Németh, B. Feil, P. Árva, J. Abonyi, Effects of catalyst activity profiles on the operating conditions of an industrial polymerization reactor, Polymer-plastics technology and engineering, pp. 1301-1306. (2006)