Multi-Criteria optimisation using past,real time and predictive performance benchmarks

Performance based design, construction, commissioning and operation of buildings requires virtual testing and validation of project alternatives. In the case of environmental and energy management of buildings, whole Building Energy Simulation (BES) models can be used to determine indoor environmental conditions, building energy consumption, system performance, and associated CO₂ emissions, etc. BES is currently used at the design and commissioning phases of the Building Life Cycle (BLC) but not during the operational phase. This paper defines a methodology that incorporates predictive BES into building operation while acknowledging present technological limitations with respect to model accuracy and required resources. This predictive model also requires detailed definition and characterisation of inputs including: Historical data from buildings; Real time data such as measurements from meters and wired and wireless sensors underpinned by a Building Management System (BMS) and Future data such as short term weather forecast values and expected occupancy schedules. The paper concludes with a demonstration of the predictive BES model methodology using an existing building at University College Cork, Ireland.     

DuD Recht

Dud Recht

DuD Recht     

An iterated sample construction with path relinking method: Application to switch allocation in electrical distribution networks

We present a metaheuristic approach which combines constructive heuristics and local searches based on sampling with path relinking. Its effectiveness is demonstrated by an application to the problem of allocating switches in electrical distribution networks to improve their reliability. Our approach also treats the service restoration problem, which has to be solved as a subproblem, to evaluate the reliability benefit of a given switch allocation proposal. Comparisons with other metaheuristics and with a branch-and-bound procedure evaluate its performance.     

Hyperspectral reflectance features of water hyacinth growing under feeding stresses of Neochetina spp. and different heavy metal pollutants

Spectral signatures of water hyacinth grown with biocontrol agents (Neochetina eichornia and N. bruchi) and various heavy metal pollutants were collected at the plant canopy level using a hand-held spectrometer to detect the biocontrol agent and heavy metal-induced plant stresses and the interaction between the two stressors. Water hyacinth was grown in 65l tubs, each with a single element from one of: As (1 mg l^?1), Au (1 mg l^?1), Cu (2 mg l^?1), Fe (0.5, 2 and 4 mg l^?1), Hg (1 mg l^?1), Mn (0.5, 2, and 4 mg l^?1), U (1 mg l^?1), and Zn (4 mg l^?1), with the exception of the control treatment. Three weeks after the metal treatments, the weevils were added to each tub, including those of the control treatment. Spectral measurements were taken before and after the addition of the weevils. Several spectral indicators of plant stress including red edge normalized difference vegetation index (RE-NDVI), modified red edge NDVI (mNDVI₇₀₅), modified simple ratio (mSR), photochemical reflectance index (PRI), and red edge position (REP) calculated using first derivative and linear extrapolation and water band index (WBI) were used to identify the plant stresses of water hyacinth. The spectral indicators of both metal and weevil plant stressors were correlated with the leaf chlorophyll content from the SPAD-502 readings at the end of the experiment. Correlations of mNDVI₇₀₅ with SPAD-502 readings were the highest followed by the indicators of REP. Cu-, Hg-, and Zn-treated plants showed significantly lower chlorophyll contents compared with the control treatment. A similar trend with four additional treatments (As, Fe-M, Mn-L, and Mn-H) was seen after the release of the weevils, indicating plant stress due to feeding by the biocontrol agent. However, adult and larval feeding was significantly reduced by Cu, Hg, and Zn elements, of which Cu was the most stressful. These results indicate that hyperspectral remote sensing has potential as a tool to determine the health status of water hyacinth from a remote location, to inform management interventions in control of the weed. However, its usage at a larger scale requires further studies.     

Implicit upper bound error estimates for combined expansive model and discretization adaptivity

A computational methodology for goal-oriented combined discretization and expansive (refined) model adaptivity by overall implicit error control of quantities of interest is presented, requiring estimators of primal and dual discretization and model errors. In the case of dimensional within model adaptivity, prolongations of coarse model solutions into the solution space of a fine model for defining a consistent model error are necessary, which can be achieved at the element level by two strategies. The first one is an orthogonalized kinematic prolongation of nodal displacements, whereas the second one uses prolongations of the external loads which are then used to solve additional local variational problems thus yielding prolongated solutions which a priori fulfill the required orthogonality relations at the element level. Finally, a numerical example of an elastic continuous T-beam is presented with comparative results where goal-oriented error estimation is applied to linear elasticity with a $2\frac{1}{2}\text{D}$ discrete Reissner–Mindlin plate model as the coarse model and the 3D theory as the fine model.     

Ranking software components for reuse based on non-functional properties

One of the biggest obstacles to software reuse is the cost involved in evaluating the suitability of possible reusable components. In recent years, code search engines have made significant progress in establishing the semantic suitability of components for new usage scenarios, but the problem of ranking components according to their non-functional suitability has largely been neglected. The main difficulty is that a component’s non-functional suitability for a specific reuse scenario is usually influenced by multiple, “soft” criteria, but the relative weighting of metrics for these criteria is rarely known quantitatively. What is required, therefore, is an effective and reliable strategy for ranking software components based on their non-functional properties without requiring users to provide quantitative weighting information. In this paper we present a novel approach for achieving this based on the non-dominated sorting of components driven by a specification of the relative importance of non-functional properties as a partial ordering. After describing the ranking algorithm and its implementation in a component search engine, we provide an explorative study of its properties on a sample set of components harvested from Maven Central.     

Towards improved benchmarking of black-box optimization algorithms using clustering problems

The field of Metaheuristics has produced a large number of algorithms for continuous, black-box optimization. In contrast, there are few standard benchmark problem sets, limiting our ability to gain insight into the empirical performance of these algorithms. Clustering problems have been used many times in the literature to evaluate optimization algorithms. However, much of this work has occurred independently on different problem instances and the various experimental methodologies used have produced results which are frequently incomparable and provide little knowledge regarding the difficulty of the problems used, or any platform for comparing and evaluating the performance of algorithms. This paper discusses sum of squares clustering problems from the optimization viewpoint. Properties of the fitness landscape are analysed and it is proposed that these problems are highly suitable for algorithm benchmarking. A set of 27 problem instances (from 4-D to 40-D), based on three well-known datasets, is specified. Baseline experimental results are presented for the Covariance Matrix Adaptation-Evolution Strategy and several other standard algorithms. A web-repository has also been created for this problem set to facilitate future use for algorithm evaluation and comparison.     

TCP is Competitive with Resource Augmentation

The well-known Transport Control Protocol (TCP) is a crucial component of the TCP/IP architecture on which the Internet is built, and is a de facto standard for reliable communication on the Internet. At the heart of the TCP protocol is its congestion control algorithm. While most practitioners believe that the TCP congestion control algorithm performs very well, a complete analysis of the congestion control algorithm is yet to be done. A lot of effort has, therefore, gone into the evaluation of different performance metrics like throughput and average latency under TCP. In this paper, we approach the problem from a different perspective and use the competitive analysis framework to provide some answers to the question “how good is the TCP/IP congestion control algorithm?” We describe how the TCP congestion control algorithm can be viewed as an online, distributed scheduling algorithm. We observe that existing lower bounds for non-clairvoyant scheduling algorithms imply that no online, distributed, non-clairvoyant algorithm can be competitive with an optimal offline algorithm if both algorithms were given the same resources. Therefore, in order to evaluate TCP using competitive analysis, we must limit the power of the adversary, or equivalently, allow TCP to have extra resources compared to an optimal, offline algorithm for the same problem. In this paper, we show that TCP is competitive to an optimal, offline algorithm provided the former is given more resources. Specifically, we prove first that for networks with a single bottleneck (or point of congestion), TCP is ${\mathcal{O}}(1)The well-known Transport Control Protocol (TCP) is a crucial component of the TCP/IP architecture on which the Internet is built, and is a de facto standard for reliable communication on the Internet. At the heart of the TCP protocol is its congestion control algorithm. While most practitioners believe that the TCP congestion control algorithm performs very well, a complete analysis of the congestion control algorithm is yet to be done. A lot of effort has, therefore, gone into the evaluation of different performance metrics like throughput and average latency under TCP. In this paper, we approach the problem from a different perspective and use the competitive analysis framework to provide some answers to the question “how good is the TCP/IP congestion control algorithm?” We describe how the TCP congestion control algorithm can be viewed as an online, distributed scheduling algorithm. We observe that existing lower bounds for non-clairvoyant scheduling algorithms imply that no online, distributed, non-clairvoyant algorithm can be competitive with an optimal offline algorithm if both algorithms were given the same resources. Therefore, in order to evaluate TCP using competitive analysis, we must limit the power of the adversary, or equivalently, allow TCP to have extra resources compared to an optimal, offline algorithm for the same problem. In this paper, we show that TCP is competitive to an optimal, offline algorithm provided the former is given more resources. Specifically, we prove first that for networks with a single bottleneck (or point of congestion), TCP is O(1){\mathcal{O}}(1)-competitive to an optimal centralized (global) algorithm in minimizing the user-perceived latency or flow time of the sessions, provided we allow TCP O(1){\mathcal{O}}(1) times as much bandwidth and O(1){\mathcal{O}}(1) extra time per session. Second, we show that TCP is fair by proving that the bandwidths allocated to sessions quickly converge to fair sharing of network bandwidth.     

Transport properties of nonelectrolyte liquid mixtures—V. Viscosity coefficients for binary mixtures of benzene plus alkanes at saturation pressure from 283 to 393 K

Viscosity coefficient measurements at saturation pressure are reported for benzene + n-hexane, benzene + n-octane, benzene + n-decane, benzene + n-dodecane, benzene + n-hexadecane, and benzene + cyclohexane at temperatures from 283 to 393 K. The characteristic parameter G in the Grunberg and Nissan equation $$\ell n\eta = x_1 \ell n\eta _1 + x_2 \ell n\eta _2 + x_1 x_2 G$$ is found to be both composition and temperature dependent for benzene + n-alkane mixtures, but it is independent of composition for the system benzene + cyclohexane.     

Expressibility of output equals input

We examine the common and seemingly simple specification that the output stream equals the input stream. We show that this is not in full generality expressible in first-order or temporal logic by an infinite set of sentences or a recursive specification, but requires certain extra assumptions, such as the existence of a clock or discrete input values.The main negative results are stated for first-order expressibility and have direct corollaries for inexpressibility in first-order temporal logic: output equals input with arbitrary delay is not expressible by a (perhaps infinite) theory (Theorems 2 and 3), even with a timestamp (Theorem 8), and is not expressible for an timeline by a sentence, even with a timestamp (Theorem 10). Output equals input with constant delay cannot be expressed for timeline by a sentence with extra unary predicates over the timeline.As an example of the positive results, we show output equals input can be expressed by a sentence in the language with a (weak) clock if the base model contains either an extra function (Theorem 14), or arithmetic (Theorem 15).Supported by Aerospace Sponsored Research