6 research outputs found
Iterative Learning Control for Fuel Robust HCCI
No full textA dual-fuel system is used to control HCCI combustion timing and load in a CFR engine. The systems steady-state response to iso-octane and n-heptane injected energy is investigated then a dynamic ARMAX model is found using system identification. This model is used to design a norm-optimal Iterative Learning Controller (ILC). A new design for an ILC is developed that requires minimal system information and is called the model-less ILC. The stability and noise transference of this new design is investigated. Both ILC designs are then optimized for experimental implementation. To attenuate noise in the system two non-causal filters are developed to use with the ILC: a Gaussian and a zero-phase Butterworth. The filters are optimized on the ARMAX model and then implemented on the CFR engine. The three best ILC designs are are found to be the model-less with zero-phase Butterworth, norm-optimal with zero-phase Butterworth and a norm optimal without a filter. These ILC designs are compared to a Proportional-Integral (PI) controller and all three ILCs are found to out perform the PI controller. The three best ILC designs are then implemented on the engine with different operating conditions to explore the controller robustness. The intake temperature and compression ratio are varied and all ILC designs performed well with the ILC convergence time being most affected by these disturbances. The disturbance rejection of the controllers is then tested with the addition of biofuels: ethanol and biodiesel. The controllers are able to converge with the new fuel and out-performed the PI controller subject to the same biofuel disturbance. For the CFR engine with HCCI combustion performing repetitive steps in combustion timing and load, ILC outperforms PI control and is robust to changes in intake temperature, compression ratio and the addition of biofuels with little changes in the final iteration reference tracking error
Mapping hierarchical sources into RDF using the RML mapping language
No full textIncorporating structured data in the Linked Data cloud is still complicated, despite the numerous existing tools. In particular, hierarchical structured data (e. g., JSON) are underrepresented, due to their processing complexity. A uniform mapping formalisation for data in different formats, which would enable reuse and exchange between tools and applied data, is missing. This paper describes a novel approach of mapping heterogeneous and hierarchical data sources into RDF using the RML mapping language, an extension over R2RML (the W3C standard for mapping relational databases into RDF). To facilitate those mappings, we present a toolset for producing RML mapping files using the Karma data modelling tool, and for consuming them using a prototype RML processor. A use case shows how RML facilitates the mapping rules' definition and execution to map several heterogeneous sources
usc-isi-i2/Web-Karma: Karma Version v2.051
No full textRelease Highlights:
<ol>
<li>Ability to Set Java Home from Karma-App <a href="https://github.com/usc-isi-i2/Web-Karma/wiki/Installation%3A-One-Click-Install">One click Karma Installer</a></li>
<li>Show rdfs:label in English by default</li>
</ol>
usc-isi-i2/Web-Karma: Karma Version 2.048
No full textRelease Highlights:
<ul>
<li>Show rdfs:label for Classes and Properties</li>
<li>New <a href="https://github.com/usc-isi-i2/Web-Karma/wiki/Installation%3A-One-Click-Install">one-click installation for Karma</a></li>
</ul>
usc-isi-i2/Web-Karma: Karma Release 2.2
No full text<ul>
<li>Updated electron version to fix security vulnerability</li>
</ul>
<B>Please view Installation Instructions for the 1-Click Installer <a href='https://github.com/usc-isi-i2/Web-Karma/wiki/Installation:-One-Click-Install'>here</a></B>
