Sawant, Parantapa
E-Mail-Adresse
Geburtsdatum
Suchergebnisse
Low-Cost Monitoring thermischer Solaranlagen mit IoT-Sensor und maschinellem Lernen
Das Projekt behandelte die kritische Notwendigkeit, die Überwachung von solarthermischen Anlagen in Bezug auf einfache Implementierung, Zuverlässigkeit und Kosteneffizienz zu verbessern. Dies wurde erreicht durch minimalen Hardwareaufwand in Form eines einzigen LoRaWAN Sensor zur Messung der Vorlauftemperatur, einer cloudbasierten Datenbank mit automatischer Einbindung von Wetterdaten der nächstgelegenen Wetterstation und eines regelbasierten Algorithmus (RBA) zur automatisierten Datenanalyse und Generation von Fehlermeldungen. Darüber hinaus waren alle Software-Tools Open Source. Die Grundlagen wurden bereits in einem früheren Projekt, «LoCoSol», gelegt. Ziel des aktuellen Projekts war es, die Genauigkeit der Fehlererkennung zu erhöhen, den Solarertrag zu schätzen und den Bewertungsprozess in einem benutzerfreundlichen und robusten Rahmen vollständig zu automatisieren. Die quantitative und qualitative Analyse der Ergebnisse bestätigte, dass die angestrebte Genauigkeit für die verschiedenen Key Performance Indikatoren (KPIs) erreicht wurde, z.B. 98% für den Pumpenbetrieb und 93% für die Stagnationserkennung. Zusätzlich wurden vier neue KPIs für die Erkennung von Wärmeverlusten durch Nachtauskühlung mit einer zufriedenstellenden Genauigkeit von mindestens 86% implementiert. Der neue RBA schätzt auch den zu erwartenden Solarertrag für jede Anlage ausreichend ab. Gegenüber dem im Vorgängerprojekt entwickelten hybriden Algorithmus wurden mit dem RBA bereits gleichwertige, beziehungsweise für einige KPIs sogar höhere Genauigkeiten erreicht. Im Hinblick auf das Software-Framework wurde ein automatisiertes Test-Setup erstellt, welches die Zuverlässigkeit des Codes erhöht, und die RBA wurde in die Datenbankumgebung des Unternehmens implementiert. Dies erhöhte die Robustheit, da die Anzahl der beteiligten Softwaretools im Vergleich zum Vorgängerprojekt reduziert wurde. Die erfolgreiche Implementierung eines Überwachungssystems zur Erkennung von Betriebsfehlern oder Abweichungen vom geplanten Solarertrag aufgrund von Konstruktions- und Installationsfehlern an 468 Anlagen (in einem relativ kurzen Zeitraum) hat die Skalierbarkeit und Kosteneffizienz dieses Frameworks bewiesen. Im Vergleich zu einem typischen drahtgebundenen Datenerfassungssystem mit TCP/IP-Kommunikationsprotokoll und mehreren Sensoren zur Überwachung einer Anlage sind geringere Kosten zu erwarten. Eine kritische Diskussion der Ergebnisse zeigt jedoch auch die Grenzen der RBA in Bezug auf die Qualität der verwendeten Daten und den hohen Aufwand für die Anpassung der Parameter. Hier untersucht das Projekt die Integration von Algorithmen des maschinellen Lernens zur automatischen Fehlererkennung und zur verbesserten Schätzung der in der RBA verwendeten Parameter. Diese Methoden haben sich als vielversprechend erwiesen, wenn sie in Kombination mit der RBA eingesetzt werden, um ein automatisiertes und genaueres System zu erhalten.
Experimental demonstration of grid-supportive scheduling of a polygeneration system using economic-MPC
Drawing off the technical flexibility of building polygeneration systems to support a rapidly expanding renewable electricity grid requires the application of advanced controllers like model predictive control (MPC) that can handle multiple inputs and outputs, uncertainties in forecast data, and plant constraints amongst other features. In this original work, an economic-MPC-based optimal scheduling of a real-world building energy system is demonstrated and its performance is evaluated against a conventional controller. The demonstration includes the steps to integrate an optimisation-based supervisory controller into a standard building automation and control system with off-the-shelf HVAC components and usage of state-of-art algorithms for solving complex nonlinear mixed integer optimal control problems. With the MPC, quantitative benefits in terms of 6–12% demand-cost savings and qualitative benefits in terms of better controller adaptability and hardware-friendly operation are identified. Further research potential for improving the MPC framework in terms of field-level stability, minimising constraint violations, and inter-system communication for its deployment in a prosumer-network is also identified.
Development and experimental evaluation of grey-box models of a microscale polygeneration system for application in optimal control
Multiperspective analysis of microscale trigeneration systems and their role in the crowd energy concept
The energy system of the future will transform from the current centralised fossil based to a decentralised, clean, highly efficient, and intelligent network. This transformation will require innovative technologies and ideas like trigeneration and the crowd energy concept to pave the way ahead. Even though trigeneration systems are extremely energy efficient and can play a vital role in the energy system, turning around their deployment is hindered by various barriers. These barriers are theoretically analysed in a multiperspective approach and the role decentralised trigeneration systems can play in the crowd energy concept is highlighted. It is derived from an initial literature research that a multiperspective (technological, energy-economic, and user) analysis is necessary for realising the potential of trigeneration systems in a decentralised grid. And to experimentally quantify these issues we are setting up a microscale trigeneration lab at our institute and the motivation for this lab is also briefly introduced.
Comparison of component-oriented and system-oriented modeling in the context of operational energy system analysis
Simulation based studies for operational energy system analysis play a significant role in evaluation of various new age technologies and concepts in the energy grid. Various modelling approaches already exist and in this original paper, four models representing these approaches are compared in two real-world hybrid energy system scenarios. The models, namely TransiEnt, µGRiDS, and OpSim (including pandaprosumer and mosaic) are classified into component-oriented or system-oriented approaches as deduced from the literature research. The methodology section describes their differences under standard conditions and the necessary parameterization for the purpose of creating a framework facilitating a closest possible comparison. A novel methodology for scenario generation is also explained. The results help to quantify primary differences in these approaches that are also identified in literature and qualify the influence of the accuracy of the models for application in a system-wide analysis. It is shown that a simplified model may be sufficient for the system-oriented approach especially when the objective is an optimization-based control or planning. However, from a field level operational point of view, the differences in the time series signify the importance of the component-oriented approaches.
An energy-economic analysis of real-world hybrid building energy systems
A coordinated operation of decentralised micro-scale hybrid energy systems within a locally managed network such as a district or neighbourhood will play a significant role in the sector-coupled energy grid of the future. A quantitative analysis of the effects of the primary energy factors, energy conversion efficiencies, load profiles, and control strategies on their energy-economic balance can aid in identifying important trends concerning their deployment within such a network. In this contribution, an analysis of the operational data from five energy laboratories in the trinational Upper-Rhine region is evaluated and a comparison to a conventional reference system is presented. Ten exemplary data-sets representing typical operation conditions for the laboratories in different seasons and the latest information on their national energy strategies are used to evaluate the primary energy consumption, CO2 emissions, and demand-related costs. Various conclusions on the ecologic and economic feasibility of hybrid building energy systems are drawn to provide a toe-hold to the engineering community in their planning and development.
Application and analysis of a model based controller for cooling towers in compression chiller plants
Cooling towers or recoolers are one of the major consumers of electricity in a HVAC plant. The implementation and analysis of advanced control methods in a practical application and its comparison with conventional controllers is necessary to establish a framework for their feasibility especially in the field of decentralised energy systems. A standard industrial controller, a PID and a model based controller were developed and tested in an experimental set-up using market-ready components. The characteristics of these controllers such as settling time, control difference, and frequency of control actions are compared based on the monitoring data. Modern controllers demonstrated clear advantages in terms of energy savings and higher accuracy and a model based controller was easier to set-up than a PID.
Simulationsgestützte Systemauslegung für sorptionsgestützte, solare Kälteanlagen
Die fluktuierende Verfügbarkeit regenerativer Energiequellen stellt eine Herausforderung bei der Planung und Auslegung regenerativer Gebäudeenergiesysteme dar. Die in einem System benötigten Speicherkapazitäten hängen dabei sowohl von der eingesetzten Regelungsstrategie als auch von den temperaturabhängigen Wirkungsgraden der Anlagenkomponenten ab. Genauere Einblicke in das Betriebsverhalten eines Gesamtsystems können dynamische Simulationen liefern, die eine Analyse der Systemtemperaturen und von Teilenergiekennwerten ermöglichen.
Demonstration of optimal scheduling for a building heat pump system using eEconomic-MPC
It is considered necessary to implement advanced controllers such as model predictive control (MPC) to utilize the technical flexibility of a building polygeneration system to support the rapidly expanding renewable electricity grid. These can handle multiple inputs and outputs, uncertainties in forecast data, and plant constraints, amongst other features. One of the main issues identified in the literature regarding deploying these controllers is the lack of experimental demonstrations using standard components and communication protocols. In this original work, the economic-MPC-based optimal scheduling of a real-world heat pump-based building energy plant is demonstrated, and its performance is evaluated against two conventional controllers. The demonstration includes the steps to integrate an optimization-based supervisory controller into a typical building automation and control system with off-the-shelf HVAC components and usage of state-of-art algorithms to solve a mixed integer quadratic problem. Technological benefits in terms of fewer constraint violations and a hardware-friendly operation with MPC were identified. Additionally, a strong dependency of the economic benefits on the type of load profile, system design and controller parameters was also identified. Future work for the quantification of these benefits, the application of machine learning algorithms, and the study of forecast deviations is also proposed.
Quasi-first-principle based grey-box modelling of microscale trigeneration systems for application in automatic control
With the need for automatic control based supervisory controllers for complex energy systems, comes the need for reduced order system models representing not only the non-linear behaviour of the components but also certain unknown process dynamics like their internal control logic. At the Institute of Energy Systems Technology in Offenburg we have built a real-life microscale trigeneration plant and present in this paper a rational modelling procedure that satisfies the necessary characteristics for models to be applied in model predictive control for grid-reactive optimal scheduling of this complex energy system. These models are validated against experimental data and the efficacy of the methodology is discussed. Their application in the future for the optimal scheduling problem is also briefly motivated.