Dynamic Time Warping und Synthesedaten zur Validierung von Seq2Seq für die Simulation. - In: ASIM Workshop 2023, (2023), S. 133-142
Seq2Seq is a machine learning method that allows to translate sequences into other sequences. This method has been tried in hybrid simulation of machine tools. The method has been used to generate time series of energy consumption of jobs from the corresponding numerical control code that runs on a machine tool. Seq2Seq suffers from various problems. Firstly, the creation of training data is costly. Secondly, standard Seq2Seq metrics only allow for the evaluation of a prediction of one timestamp at a time, not an entire time series. Thirdly, training metrics are failing when vanilla data is used, as two identical numerical control codes can result in deviating time series. This causes confusion for the model in the training loop, as it is not clear which time series should be considered correct. Here we propose a holistic framework to all three problems, that contains synthetic data, additional metrics for time series and dynamic time warping.
https://doi.org/10.11128/arep.21.a2113
Hochaufgelöste Energieprofile durch hybride Simulation. - In: ASIM SST 2022 Proceedings Langbeiträge, (2022), S. 243-251
The price of a commodity, as electricity, is determined on a commodity market. A market is efficient when the supply and demand in the market are at an equilibrium. Efficient markets run on information. Information can cause a spontaneous and instantaneous change within the supply and demand in a market. The market communicates this new equilibrium through the change of the price of a commodity. In the electricity market the supplier and consumer communicate through electrical load profiles. A load profile signals when and how much energy should be consumed within a certain time frame without causing a change in the price of electricity. Creating such load profiles is commonly done by the supplier of energy by means of standard load profiles. Here we propose a data-driven simulation-based method that allows for the consumer to create its own specific load profile, which potentially will bring down the cost of energy consumed.
https://dx.doi.org/10.11128/arep.20.a2004
A method for predicting high-resolution time series using sequence-to-sequence models. - In: 2020 Winter Simulation Conference (WSC), (2020), S. 1075-1086
With the increasing availability of data, the desire to interpret that data and use it for behavioral predictions arises. Traditionally, simulation has used data about the real system for input data analysis or within data-driven model generation. Automatically extracting behavioral descriptions from the data and representing it in a simulation model is a challenge of these approaches. Machine learning on the other hand has proven successful to extract knowledge from large data sets and transform it into more useful representations. Combining simulation approaches with methods from machine learning seems therefore promising to combine the strengths of both approaches. Representing some aspects of a real system by a traditional simulation model and others by a model incorporating machine learning, a hybrid system model (HSM) is generated. This paper suggests a specific HSM incorporating a deep learning method for predicting the anticipated power usage of machining jobs.
https://doi.org/10.1109/WSC48552.2020.9383969
On the usage of deep learning for modelling energy consumption in simulation models. - In: Simulation Notes Europe, ISSN 2164-5353, Bd. 30 (2020), 4, S. 165-174
With the increasing availability of data, the desire to interpret that data and use it for behavioral predictions arises. Traditionally, simulation has used data about the real system for input data analysis or within data-driven model generation. Automatically extracting behavioral descriptions from the data and representing it in a simulation model is a challenge for these approaches. Machine learning on the other hand has proven successful in extracting knowledge from large data sets and transforming it into more useful representations. Combining simulation approaches with methods from machine learning seems, therefore, promising. Representing some aspects of a real system by a traditional simulation model and others by a model generated from machine learning, a hybrid system model (HSM) is generated. This paper discusses such HSMs and suggests a specific HSM incorporating a deep learning method for predicting the power consumption of machining jobs.
https://doi.org/10.11128/sne.30.tn.10536
Sequence to Sequence Modelle zur hochaufgelösten Prädiktion von Stromverbrauch. - In: Proceedings ASIM SST 2020, (2020), S. 149-157
Modelling power consumption for jobs on a ma-chine tool is commonly performed by measuring the real power consumption of comparable jobs and machines. The so gathered data is then processed to represent the time-av-eraged sums of power consumptions of previous jobs. These values of power consumption are then used for upcoming comparable jobs. This approach allows for no high-resolution prediction of power consumption and further presumes static processing times of jobs. Here we propose a new approach to model power consumption that incorporates a Sequence-to-Sequence model, which generates time series according to dynamic data, that describes a numerical control code and environment settings such as state of tools, etc.
https://doi.org/10.11128/arep.59.a59021
Deep learning based prediction of energy consumption for hybrid simulation :
Deep-Learning-basierte Prognose von Stromverbrauch für die hybride Simulation. - In: Simulation in Produktion und Logistik 2019, (2019), S. 121-131
Modern production facilities need to prepare for changing market conditions within the energy market due to ongoing implementation of governmental policies. This results in higher volatility of the availability of energy and therefore energy costs. If a simulation model of a machinery model can estimate its own future consumption, and according time frames for said consumption, this information could be used for optimized scheduling of energy consuming jobs. This would result in lower procurement costs. To make said estimation about the dynamic behaviour of jobs, methods of time series prediction tend to be applied. Here a proposal is made to apply a Hybrid System Model incorporating a recurrent neural network (RNN)-Encoder-Decoder-Architecture, which returns a discrete times series when a behavioural sequence (such as an NC-Code) has been put into a neural net model of the respective machinery. Those discrete time series reflect the machines energy consumption for each job that it has been operated on. This neural net, if weighted and called, emits the length value of a job and an according time series which displays the quasi-continuous time consumption of said job. Such generative models combined with classic simulation paradigm qualify as potent applications of hybrid simulation approaches.
Hybride Simulation zur realitätsnahen Abbildung von Energieströmen : Kombination von Materialflusssimulation und maschinellem Lernen. - Ilmenau. - 117 Seiten
Technische Universität Ilmenau, Masterarbeit 2018
Moderne Produktionsunternehmen müssen im Kontext der Energiewende sich auf veränderte Marktverhältnisse innerhalb des Energiemarktes einstellen. Diese sind durch eine höhere Volatilität der Verfügbarkeit und daher auch der Kosten von Energien geprägt. Könnte ein Unternehmen vor Abruf seines Energiebedarfs diesen realitätsnah ermitteln und an sein Versorgungsunternehmen weitergeben, so würde dies in fallende Beschaffungskosten resultieren. Um Voraussagen über das dynamische Verhalten einer Organisation zu treffen, werden Methoden der Simulation angewandt. Die Qualität der aus Simulation erzeugten Erkenntnisse nimmt gemeinhin mit einer erhöhten Modellkomplexität und Datenlage zu. Desto komplexer sich eine Simulationsumgebung darstellt, desto höher ist der Aufwand diese mit klassischen Methoden des gewählten Simulationsparadigmas darzustellen. Ziel dieser Arbeit war es die Grenzen der Modellierung einer solchen realitätsnahen Systems anhand des Bedienraums einer Werkzeugmaschine aufzuzeigen, und einen Vorschlag einer alternativen Modellierung über hybride Simulationsansätze zu liefern. Hierfür wurde einmal der Vorschlag einer ereignisdiskreten Modellierung einer Werkzeugmaschine geschaffen, welche eine Simulation des zeitlichen Verlaufs des Energieverbrauchs darstellt und den Aufwand für ebenso eine Erstellung charakterisiert. Die Anforderungen an das diskret ereignisorientierte Modell wurden mit möglichen alternativen Modellen des maschinellen Lernens abgeglichen. Hierbei wurde eine RNN-Encoder-Decoder-Architektur identifiziert, welche bei Eingabe einer Verhaltensbeschreibung des Bedienraums, als Ausgabe eine diskrete Zeitreihe hatte. Diese diskrete Zeitreihe stellt den Energieverbrauch der Werkzeugmaschine für jeden Fertigungsauftrag dar. Das Modell der diskret ereignisorientierten Simulation wurde mit dem Modell der RNN-Encoder-Decoder-Architektur kombiniert, um eine ganzheitliche Beschreibung der Energieverbrauchszeiten simulieren zu können. Bei dem entstandenen Simulationsmodell handelt es sich um ein Modell der diskret ereignisorientierten Weltsicht, welches beim Durchlauf eines Fertigungsauftrags durch den Spanraum (Bedienraum) einer Werkzeugmaschine, ein zuvor gewichtetes rekurrentes, neuronales Netz und den vektorisierten NC-Code des spezifischen Fertigungsauftrags einlädt. Dieses RNN wird bei Simulationslauf aktiviert und hat als Ausgabe die zeitliche Länge des Fertigungsauftrages und eine Zeitreihe des kontinuierlichen Energieverbrauchs des Fertigungsauftrags.