Simulated Annealing in Optimization of Energy Production in a Water Supply Network

In water supply systems, the potential exists for micro-hydropower that uses the pressure excess in the networks to produce electricity. However, because urban drinking water networks are complex systems in which flows and pressure vary constantly, identification of the ideal locations for turbines is not straightforward, and assessment implies the need for simulation. In this paper, an optimization algorithm is proposed to provide a selection of optimal locations for the installation of a given number of turbines in a distribution network. A simulated annealing process was developed to optimize the location of the turbines by taking into account the hourly variation of flows throughout an average year and the consequent impact of this variation on the turbine efficiency. The optimization is achieved by considering the characteristic and efficiency curves of a turbine model for different impeller diameters as well as simulations of the annual energy production in a coupled hydraulic model. The developed algorithm was applied to the water supply system of the city Lausanne (Switzerland). This work focuses on the definition of the neighborhood of the simulated annealing process and the analysis of convergence towards the optimal solution for different restrictions and numbers of installed turbines.

Optimization of low-head hydropower recovery in water supply networks

Irene Almeida Samora

Small scale hydropower is emerging as a decentralized source to satisfy local demand for electricity. The interest in micro-hydropower, which refers to installed power below 100 kW, is increasing since this is a solution with low environmental impact. Water supply systems are one of the main manmade water systems presenting potential for micro-hydropower. Although some applications already exist in adduction lines, the urban distribution networks remain unexplored. Because these are complex systems in which flows and pressure vary constantly, specific technology and installation schemes for energy recovery are lacking. This work accesses the potential for hydropower within water supply networks (WSN) and presents an arrangement of micro-turbines specially conceived for this type of installation. The arrangement is based on a novel inline turbine suitable for pressurized systems and its best location within the networks is studied. The five blade tubular propeller (5BTP), preliminarily designed in the framework of the European Project HYLOW, was further developed and tested in this work. An experimental campaign was conducted with a large range of heads and torque measurements to access its characteristic curves and to obtain hill charts. The relative positioning of the turbine-generator shaft regarding the pipe bend was modified from a downstream position to an upstream position. Efficiencies of around 60% were found. The adequate locations in WSN for micro-hydropower plants are identified using an optimization algorithm which considers both the assessment of the energy production and sizing of the main equipment and works. A concept for the implementation of the 5BTP in the field was developed, consisting of an arrangement with up to four turbines inline within a buried chamber created around an existing pipe. Two objective functions, energy production and economic value respectively, are used. A simulated annealing process is developed to optimize the location of a given number of turbines. This procedure takes into account the hourly variation of flows throughout an average year and its effect on the turbine efficiency. The optimization is achieved by considering the characteristic and efficiency curves of a turbine with different impeller diameters and simulating the annual energy production in a coupled hydraulic model. After a convergence analysis for different restrictions and numbers of installed turbines, the algorithm was applied to analyze the feasibility of the proposed arrangement in two case studies in Switzerland: a sub-grid of the city of Lausanne and the complete WSN of the city of Fribourg. In both cases, the implementation of the proposed energy recovery solution seems to be feasible. A detailed analysis of the cost breakdown revealed that the cost of additional pipe work, which are required in each layout to guarantee a by-pass supply in case of maintenance, may have an important role on the investments. Also, the pressure reduction valves locations, if they exist, are likely to be the optimal solutions. Finally, a methodology to quantify the potential for hydropower based on the excess energy in a WSN is proposed and applied to case studies. It allowed to conclude how much the proposed arrangement can extract from the networksâ energy potential. In addition, an attempt was made to produce an expedite method to estimate the energy produced with one 5BTP based on network parameters and dimensional analysis.

EPFL2016

Optimal location of micro-turbines in water supply network

Irene Almeida Samora, Mário Jorge Rodrigues Pereira Da Franca, Anton Schleiss

Micro-hydropower is currently expanding as a solution to improve the efficiency of water systems by using energy excesses which are typically lost. In the particular case of water supply systems, often excessive pressure exists in zones of the network connected to other areas situated at higher altitudes. Pressure reducing valves are commonly used as a mean for dissipation of this excess energy. In this work, the installation of micro-turbines in a closed water supply network is analyzed as a way to recover the existing surplus of energy by converting it into electricity. The flow in water supply systems is highly variable, with a direct impact on the efficiency of the turbine and, as most water-supply networks are meshed, the optimal location of the energy converters is not straightforward and needs assessment by simulation processes. For this purpose, an optimization tool based on the application of an evolutionary algorithm was developed to select the best location, model and runner size of a selected number of turbines to install in a network. Using the characteristic and efficiency curves of a micro-turbine and a database of flow demand given every hour, a simulated annealing process is applied to maximize the energy production while pressure restrictions imposed by regulation must be respected. The methodology was applied to a case study in a sub-grid of the water supply system of the city of Lausanne, Switzerland, and the considered micro-hydro converter was a five blade tubular propeller. This study is focused on the simulation problem and the convergence to optimal solution is analyzed under different restrictions and number of turbines to install.

2015

Optimization of low-head hydropower recovery in water supply networks

Irene Almeida Samora

EPFL2016