The task of our work is segmentation of a speech signal when having a speech waveform and parameters of the segments. We used dynamic programming approach.

No description available.

Over the last 20 years, electricity markets around the world have gradually been deregulated, creating wholesale markets in which generating companies compete for the right to supply electricity, through an offering system. This thesis considers the optimisation of the offering process from the perspective of an individual generator, subject to intertemporal constraints including fuel limitations, correlated rest-of-market behaviour patterns and unit operational decisions. Contributions from the thesis include a Pre-Processing scheme that results in considerable computational benefits for a two-level Dynamic Programming method, in addition to the development of a new process that combines the techniques of Decision Analysis and Dynamic Programming.

Generation and transmission investment planning in deregulated markets faces new challenges particularly as deregulation has introduced more uncertainty to the planning problem. Tradi- tional planning techniques and processes cannot be applied to the deregulated planning problem as generation investments are profit driven and competitive. Transmission investments must facilitate generation access rather than servicing generation choices. The new investment plan- ning environment requires the development of new planning techniques and processes that can remain flexible as uncertainty within the system is revealed. The optimisation technique of Stochastic Dual Dynamic Programming (SDDP) has been success- fully used to optimise continuous stochastic dynamic planning problems such as hydrothermal scheduling. SDDP is extended in this thesis to optimise the stochastic, dynamic, mixed integer power system investment planning problem. The extensions to SDDP allow for optimisation of large integer variables that represent generation and transmission investment options while still utilising the computational benefits of SDDP. The thesis also details the development of a math- ematical representation of a general power system investment planning problem and applies it to a case study involving investment in New Zealand’s HVDC link. The HVDC link optimisation problem is successfully solved using the extended SDDP algorithm and the output data of the optimisation can be used to better understand risk associated with capital investment in power systems. The extended SDDP algorithm offers a new planning and optimisation technique for deregulated power systems that provides a flexible optimal solution and informs the planner about investment risk associated with uncertainty in the power system.

Markov decision processes (MDPs) are a general framework used by Artificial Intelligence (AI) researchers to model decision theoretic planning problems. Solving real world MDPs has been a major and challenging research topic in the AI literature. This paper discusses two main groups of approaches in solving MDPs. The first group of approaches combines the strategies of heuristic search and dynamic programming to expedite the convergence process. The second makes use of graphical structures in MDPs to decrease the effort of classic dynamic programming algorithms. Two new algorithms proposed by the author, MBLAO* and TVI, are described here.

This thesis first develops a range of wholesale water market design options, based on an optimisation approach to market-clearing, as in electricity markets, focusing on the extent to which uncertainty is accounted for in bidding, market-clearing and contract formation. We conclude that the most promising option is bidding for, and trading, a combination of fixed and proportionally scaled contract volumes, which are based on optimised outputs. Other options include those which are based on a post-clearing fit (e.g. regression) to the natural optimised outputs, or constraining the optimisation such that cleared allocations are in the contractual form required by participants. Alternatively, participants could rely on financial markets to trade instruments, but informed by a centralised market-clearing simulation. We then describe a computational modelling system, using Stochastic Constructive Dynamic Programming (CDDP), and use it to assess the importance of modelling uncertainty, and correlations, in reservoir optimisation and/or market-clearing, under a wide range of physical and economic assumptions, with or without a market. We discuss a number of bases of comparison, but focus on the benefit gain achieved as a proportion of the perfectly competitive market value (price times quantity), calculated using the market clearing price from Markov Chain optimisation. With inflow and demand completely out of phase, high inflow seasonality and volatility, and a constant elasticity of -0.5, the greatest contribution of stochastic (Markov) optimisation, as a proportion of market value was 29%, when storage capacity was only 25% of mean monthly inflow, and with effectively unlimited release capacity. This proportional gain fell only slowly for higher storage capacities, but nearly halved for lower release capacities, around the mean monthly inflow, mainly because highly constrained systems produce high prices, and hence raise market value. The highest absolute gain was actually when release capacity was only 75% of mean monthly inflow. On average, over a storage capacity range from 2% to 1200%, and release capacity range from 100% to 400%, times the mean monthly inflow, the gains from using Markov Chain and Stochastic Independent optimisation, rather than deterministic optimisation, were 18% and 13% of market value, respectively. As expected, the gains from stochastic optimisation rose rapidly for lower elasticities, and when vertical steps were added to the demand curve. But they became nearly negligible when (the absolute value of) elasticity rose to 0.75 and beyond, inflow was in-phase with demand, or the range of either seasonal variation or intra-month variability reduced to ±50% of the mean monthly inflow. Still, our results indicate that there are a wide range of reservoir and economic systems where accounting for uncertainty directly in the water allocation process could result in significant gains, whether in a centrally controlled or market context. Price and price risk, which affect individual participants, were significantly more sensitive. Our hope is that this work helps inform parties who are considering enhancing their water allocation practices with improved stochastic optimisation, and potentially market based mechanisms.

Advanced hybrid vehicle architectures can exploit multiple power sources and optimal control to achieve high efficiency operation. In this work, a method for generating the best-possible energy efficiency benchmark for a hybrid architecture is introduced. The benchmark program uses Dynamic Programming to analyse a reduced-fidelity MATLAB model over standard driving cycles, and bypasses vehicle controls to identify the optimal control actions and resulting fuel consumption of the Series-Parallel Multiple-Regime retrofitted PHEV of the UVic EcoCAR2 program. The simulation results indicate an optimal fuel consumption value of 4.74L/100km, in the parallel regime, compared to the stock Malibu's 8.83L/100km. The results are found to be sensitive to the allowed level of regenerative braking, with an optimal consumption value of 6.56L/100km obtained with restricted regen power limits. The parallel regime provided more efficient operation overall, especially during more aggressive driving conditions. However, the series regime provided more desirable operation during gentle driving conditions, where opportunities for regenerative braking are limited. The generated powertrain control profiles were then used to drive a higher-fidelity Simulink model. Due to the significant difference between the model structures of the MATLAB and Simulink models, comparison of results were not conclusive. A different simulation approach is required to make this proof-of-concept more useful for controls development. This research forms the foundation for further studies. / Graduate / 0540 / snkaban@gmail.com

This thesis (1) reports a new Dynamic Programming (DP) approach, and (2) reports a Real Time Control strategy to optimize the energy management of a Hybrid Electric Vehicle(HEV). Increasing environmental concerns and rise in fuel prices in recent years has escalated interest in fuel efficient vehicles from government, consumers and car manufacturers. Due to this, Hybrid electric vehicles (HEV) have gained popularity in recent years. HEV’s have two degrees of freedom for energy flow controls, and hence the performance of a HEV is strongly dependent on the control of the power split between thermal and electrical power sources. In this thesis backward-looking and forward-looking control strategies for two HEV architectures namely series and parallel HEV are developed. The new DP approach, in which the state variable is not discretized, is first introduced and a theoretical base is established. We then prove that the proposed DP produces globally optimal solution for a class of discrete systems. Then it is applied to optimize the fuel economy of HEV's. Simulations for the parallel and series HEV are then performed for multiple drive cycles and the improved fuel economy obtained by the new DP is compared to existing DP approaches. The results are then studied in detail and further improvements are suggested. A new Real Time Control Strategy (RTCS) based on the concept of preview control for online implementation is also developed in this thesis. It is then compared to an existing Equivalent Cost Minimization Strategy (ECMS) which does not require data to be known apriori. The improved fuel economy results of the RTCS for the series and parallel HEV are obtained for standard drive cycles and compared with the ECMS results

Data Grids provide services and infrastructure for distributed data-intensive applications accessing massive geographically distributed datasets. An important technique to speed access in Data Grids is replication, which provides nearby data access. Although data replication is one of the major techniques for promoting high data access, the problem of replica placement has not been widely studied for large-scale Grid environments. In this thesis, I propose improved data placement techniques useful when replicating potentially large data files in wide area data grids. These techniques are aimed at achieving faster data access as well as efficient utilization of bandwidth and storage resources. At the core of my approach is a new highly distributed replica placement algorithm that places data in strategic locations to improve overall data access performance while satisfying varying user/application and system demands. This improved efficiency of access to large data will improve the practicality of large-scale data and compute intensive collaborative scientific endeavors. My thesis makes several contributions towards improving the state-of-the-art for replica placement in large-scale data grid environments. The major contributions are: (i) development of a new popularity-driven dynamic replica placement algorithm for hierarchically structured data grids that balance storage space utilisation and access latency; (ii) creation of an adaptive version of the base algorithm to dynamically adapt the frequency and degree of replication based on such factors as data request arrival rates, available storage capacities, etc.; (iii) development of a new highly distributed algorithm to determine a near-optimal replica placement while minimizing replication cost (access and update) for a given traffic pattern; (iv) creation of a distributed QoS-aware replica placement algorithm that supports multiple quality requirements both from user and system perspectives to support efficient transfers of large replicas. Simulation results using widely observed data access patterns demonstrate how the effectiveness of my replica placement techniques is affected by various factors such as grid network characteristics (i.e. topology, number of nodes, storage and workload capacities of replica servers, link capacities, traffic pattern), QoS requirements, and so on. Finally, I compare the performance of my algorithms to a number of relevant algorithms from the literature and demonstrate their usefulness and superiority for conditions of interest.

A continuous service innovation such as Cloud Computing is highly attractive in the business-to-business world because it brings the service provider both billions of dollars in profits and superior competitive advantage. The success of such an innovation is strongly tied to a consumer’s adoption decision. When dealing with a continuous service innovation, the consumer’s decision process becomes complicated. Not only do consumers need to consider two different decisions of both whether to adopt and how long to adopt (contract length), but also the increasing trend of the service-related technological improvements invokes a forward-looking behavior in consumer’s decision process. Moreover, consumers have to balance the benefits and costs of adoption when evaluating decision alternatives. Consumer adoption decisions come with the desire to have the latest technology and the fear of the adopted technology becoming obsolete. Non-adoption prevents consumers from being locked-in by the service provider, but buying that technology may be costly. Being bound to a longer contract forfeits the opportunity to capitalize on the technological revolution. Frequently signing shorter contracts increases the non-physical efforts such as learning, training and negotiating costs. Targeting the right consumers at the right time with the right service offer in the business-to-business context requires an efficient strategy of sales resource allocation. This is a “mission impossible” for service providers if they do not know how consumers make decisions regarding service innovation. In order to guide the resource allocation decisions, we propose a complex model that integrates the structural, dynamic, and learning approaches to understand the consumer’s decision process on both whether or not to adopt, and how long to adopt a continuously updating service innovation in a B2B context.