SKOLTECH Project: Information-computation technologies for the analysis and provision of identifiability and controllability of energy pipeline systems, 2015-2017

Project Summary

The main problem dealt with in the project arises due to the absence of common methods practically applicable for quantitative analysis and justification of the required degree of identifiability and controllability of energy pipeline systems (for heat, water, oil, gas supply, etc.). This problem is a limiting factor in the development and introduction of the advanced concepts of technological control. Potentially such concepts have to provide an optimal control of pipeline systems (PLS) through the adaptation of the decision-making processes to changes both in the external conditions and in the internal state of a control object.

Identifiability is one of system properties, that determines the restoration possibility of its mathematical model accurate to the parameters from the results of measurements and observations. Controllability is a capability of system to maintain the admissible operating conditions of pipeline systems by using the control tools.

For the problems of identifiability analysis it is planned in the project to generalize the criteria and algorithms for estimating the possibility of determining actual parameters and characteristics of the pipeline system elements, calculating the conditions based on measurement data, and adjusting the structure (type) of mathematical models. To ensure identifiability it is suggested that the theory and methods developed at the Melentiev Energy Systems Institute of SB RAS for the pipeline system identification be updated on the basis of new capabilities of information-measurement systems and methods for active identification and monitoring of pipeline conditions in real time. This in turn will make it possible to check the admissibility of conditions, the quality of supply, and make timely decisions on control in normal and emergency situations.

It is also planned to develop a new approach to the pipeline system controllability analysis that will be based on comprehensive consideration of stochastic external impacts, probabilistic flow distribution and uncertainty of internal pipeline system parameters. The control problems are planned to be solved by the enhancement of the system of original approaches and methods of structural and parametric optimization of conditions that was developed at the Melentiev Energy Systems Institute. It is also planned to make this system practically applicable.

Project description.

Identifiability of pipeline systems. A set of the pipeline system identifiability problems can be represented by two main classes 1) the problems of analysis, consisting in establishing the possibility of the pipeline system identification under specified conditions and 2) the problems of synthesis, consisting in finding the conditions which will ensure this possibility.

For the analysis of identifiability it is planned in the project to comprehensively consider the main factors that affect this characteristic to develop and systematize the criteria and methods to quantitatively assess the possibility of 1) determining the actual parameters and characteristics of the pipeline system elements; 2) calculating the operating conditions based on the measurements, 3) specifying the structure (form) of the mathematical models and 4) providing the adequacy of the models.

For the methods of ensuring the identifiability it is planned to generalize and update the methods of pipeline system identification in terms of the capabilities of modern information and measurement systems. The methods include: 1) the methods to optimize the allocation of measuring devices to identify the parameters and operating conditions of the pipeline system; 2) the methods to actively identify the pipeline system, which make it possible to extract most of the information on pipeline system characteristics and reduce the laboriousness of experiments, 3) the methods to identify the pipeline systems operation to monitor their behavior in real time.

Pipeline System Controllability. Currently there are no formally rigorous methods for the pipeline system controllability analysis which are also suitable for practical application. Therefore, it is planned 1) to develop a new probabilistic approach to the quantitative assessment of the pipeline system controllability, considering the uncertainty factors of the external impacts and internal parameters of pipeline systems; 2) devise and study the methods of simulation modeling of the pipeline system operation, considering the factors of uncertainty and delay in the internal responses for the analysis of pipeline system controllability.

Modeling of pipeline system operating conditions. The advent of new equipment including that for control and regulation, as well as the emerging new methods of control, and the factors of active behavior of consumers require the improvement of the existing models of flow distribution in the pipeline system, which represent the basis for stating and solving the entire set of problems considered in the project. In this respect the project suggests:1) developing the models and methods for the flow distribution calculation which could be independent of concrete relationships for the laws of working medium flow along certain components of the pipeline systems, 2) updating the models and methods for flow distribution calculation for the case of nontraditional types of automatic regulators (of flow rate, pressure, etc.); 3) devising and updating the probabilistic models and methods to calculate hydraulic and thermal–hydraulic conditions.

Pipeline system operation optimization. The main requirements for the operating conditions of pipeline systems are related not only to the provision of services for consumers but also to the satisfaction of a set of economic and technological limitations. Here it is planned: 1) to develop the efficient methods for optimizing the hydraulic conditions of pipeline systems according to the economic criteria in order to search for the optimal controls of continuous and discrete nature; 2) to investigate the problems and develop the methods for optimizing the operation according to the technological criteria related to the minimization of the control laboriousness; 3) to develop and test the technique of multi-level (hierarchical) optimization of large pipeline systems.

Water supply system of the city of Irkutsk owned by MUE Vodokanal has been chosen as a pilot project to test the developed methodology.
x
x