Per Printz Madsen

The HomePort System

Residential gateways for home automation alre prerequisites to obtain optimal exploitation of energy resources, and they also have the potential to provide a unified operation of various home devices and appliances. Although a number of protocol standards have been proposed, the number of commercially available systems is still very limited. One reason for this is the diversity of device manufacturing standards; another is the lack of efficient and expressive middleware for defining control algorithms and usage scenarios. In this paper we present the architectural design of a distributed middleware system for residential gateways including a simple composition language. Also, we present the initial experiences obtained from a prototype implementation of the system.

An intuitive definition of demand flexibility in direct load control

Two control approaches: direct and indirect control of demand side energy management in a smart grid are studied. Indirect control of energy demands is based on economic incentives. In this approach, consumers will shift their energy consumption with the benefit of a cut down in the electricity bill and in the cost of being flexible in terms of accepting a wider comfort zone. The other approach is to directly command a power signal to end-user loads provided that the electricity provider company is already aware of the consumers flexibility degree. We have framed a measure of the flexibility in terms of the amount of energy that can be shifted without sacrificing comfort. A specific case study i.e. a residential building equipped with an electric floor heating system is used for demonstration of simulation results. Finally, we proposed a control method to serve both direct and indirect load shifting schemes. Simulation results show a sub-optimal result with the combination approach compared to the optimal solution with the indirect approach.

Neural network for optimization of existing control systems

The purpose of this paper is to develop methods to use neural network based controllers (NNC) as an optimization tool for existing control systems. Two different methods are suggested. One uses the NNC as a feedforward controller and the other uses the NNC as a feedback controller. The main advantage of these methods is that they make it possible to retain an existing traditional control system. In these methods the NNC is doing the optimization and not the stabilization of the process. A thermal mixing process is used as a test system, which is a multivariable and nonlinear process. The method based on feedforward has been tested and shows very good performance.

Energy Storage in Ordinary Houses. A Smart Grid Approach

Abstract The last couple of years more and more non-controllable energy sources, e.g. wind turbines, have been connected to the power grid. This has caused an inefficient energy production and a huge variation in the energy prices. In the near future (10 to 15 years) the amount of non-controllable energy sources will double or more. This again will result in a number of problems: Inefficient energy production, problems with controlling the power grid, the wind turbines have to stop when the wind is optima and so on. A way to reduce these problems is to use the possibility to store energy in ordinary houses. Ordinary houses are the main electricity consumer. More than 60% of the total electricity production is consumed in family homes. A way to overcome these problems with non-controllable electricity production is to use residential homes as energy storage. For instance floor heating systems, hot water tanks, refrigerators and freezers are all examples of systems and appliance that can be used for storing energy in an efficient way. Beside of that it is possible to use the washing machine, the dryer and the ventilation in a more flexible manner which will have the same effect than storing energy. This paper analyzes the possibility for using floor heating systems and hot water tanks as energy storage in ordinary family houses. Beside of that the size and the efficiency of this energy storage are analyzed. The main idea are to make the power consumer more flexible, or in other word the consumer have to use the energy when it is available, and by that mean secure a more efficient and green energy production. If houses and private households have to play that role, then it is very important to find a solution that minimizes the impact on the living comfort. The way to do that is to make an automatic and intelligent house control system that maximizes the consumption flexibility based on the energy users behavior without affection the living comfort. This behavior is of course different from household to household, because of that it is necessary to include an adaptive behavior prediction system. This paper describes a method for making houses more energy flexible based on pre knowledge of the user behavior. The energy flexibility is calculated based on simulation for 200000 typical houses in Denmark. The simulations shows that it is possible based on these 200000 houses, to store enough energy to absorb the variation in the production in Denmark over a time horizon on several hours.

HomePort: An intelligent home control system

In the last couple of years computer based home control systems are getting more and more common in modern homes. For instance these systems take care of light control, heat control and security systems. The latest trend is to use wireless communication like Z-Wave and ZigBee to interconnect different components in these systems. One of the characteristics is that each system, like for instance heat and light, has their own specific way of using the communication system. This paper describes a way to connect different home control systems through an intelligent gateway, called a HomePort. The HomePort consists of a number of Subsystem communication drivers, a virtual communication layer, an interpreter and a PC-based compiler for a high level control language, called GIL (Gateway intelligence language). The focus in this paper will be on the upper two layers in the HomePort, the interpreter and the application layer.

Economic energy distribution and consumption in a microgrid Part1: Cell level controller

We have investigated energy management of a small scale electrical microgrid comprised of local renewable generation, consumption and storage units. The microgrid has the possibility of connection to the electricity grid as well to compensate for energy deficit. The objective is to fulfill microgrids energy demands from the local electricity producers as much as possible. The other objective is to manage the consumption such that consumption costs are minimum for all households. To fulfill the objectives, as the first step of designing a hierarchical controller, we focused on designing an energy and cost minimizing controller for one building. To this aim, a model predictive controller is formulated to schedule the buildings energy consumption using potential load flexibility. Simulation results show the economically optimal energy consumption of one building based on the defined load types, a user-specified comfort criterion and the grid electricity price.

Dedicated Programming Language for Small Distributed Control Devices

Small control computers are more and more used in modern households. These computers are for instance in washing machines, heating systems, secure systems, televisions and stereos. In the future all these computers will communicate with each other to implement the intelligent house. This can only become a reality if each of these controlling computers can be configured to perform a cooperative task. This again requires the necessary communicating facilities. In other words this requires that all these simple and distributed computers can be programmed in a simple and hardware independent way. This paper describes a new, flexible and simple language for programming distributed control tasks. The compiler for this language generates a target code that is very easy to interpret. An interpreter, that can be easy ported to different hardwares, is described. The new language is simple and easy to learn, but, on the other hand, it is general enough to be used in a large group of house control systems.

Neural network for combining linear and non-linear modelling of dynamic systems

The purpose of this paper is to develop a method to combine linear models with MLP networks, i.e. to find a method to make a nonlinear and multivariable model that performs at least as good as a linear model, when the training data lacks information. First, the MLP network for predicting the output from a dynamic system is described. Then two methods are proposed to combine linear and nonlinear modelling. The first method is the MLP network with linear path through, and the second method is a linear model with nonlinear error correction. Finally the two methods are tested. A thermal mixing process is used as a test system. This system is a multivariable and nonlinear process. The test is partly based on a simulation of the process and partly on data from a physical process. The results are given and discussed.<<ETX>>

Changes in roles and behavior of learners and teachers when going for POPBL

It is accepted world wide; that project-organized problem-based learning (POPBL) is the right method to improve learning and satisfy the customers being more innovative and creative. Shifting to POPBL in universities seems uncomplicated when it comes to changes in materials and procedures, but for the people involved, mentally, it is a very difficult change in behavior and attitudes. Organizations and staff in many universities are skeptical about POPBL concerning student competence measurement and assessments. On the personal level the teachers major apprehension is the loss of prestige, templates, and traditions. Students need to be more active in the learning process so the concern is more work. It is always difficult to make major changes in an (mature) educational institution, but shifting to POPBL collaborative learning has been proven to be a successful learning method at Aalborg University, Denmark (AAU) for more than 30 years and it has provided great satisfaction for all parties including students and customers. Although, it was difficult in the beginning and after much iteration, AAU is now the leading university in on-campus POPBL, as well as on-line POPBL. However, the ongoing development of information technology (IT), pedagogy and didactic methods requires new demands to all parties in openness, willingness, and readiness in adapting new roles and behavior. This article describes challenges and changes in roles and behavior when going for and maintaining a modern and successful POPBL environment. Based on years of POPBL development and experience - methods and recommendations will be presented to staff and students when working with the courses and projects in an on-campus, as well as on-line POPBL educational environment.

ENCOURAGEing results on ICT for energy efficient buildings

This paper presents how the ICT infrastructure developed in the European ENCOURAGE project, centered around a message oriented middleware, enabled energy savings in buildings and households. The components of the middleware, as well as the supervisory control strategy, are overviewed, to support the presentation of the results and how they could be achieved. The main results are presented on three of the pilots of the project, a first one consisting of a single household, a second one of a residential neighborhood, and a third one in a university campus.