A. C. de Keizer, K. Vajen and U. Jordan

Kassel University, Institute of Thermal Engineering, 34125 Kassel, Germany, www. solar. uni-kassel. de

Corresponding Author, solar@uni-kassel. de

Abstract

Continuous monitoring and failure detection during the life time of a solar thermal system is important to detect occurring failures as quick as possible. Therefore, several methods have been developed during the last decades. However, so far application is mainly limited to research and demonstration projects. In this paper several failure detection methods are described and compared with a partial multi-criteria analysis.

Up to now monitoring approaches have primarily been applied with data analysis by an expert, but without an automatic analysis of the data through the method. There are some methods that include automatic failure detection; which is based on a static function control or on a simulation based comparison. Up to now none of the systems include the auxiliary heating. Keywords: monitoring, failure detection, solar thermal systems

1. Introduction

The solar thermal energy market is growing. Solar thermal systems are designed to function for at least 25 years, but failures and malfunctions of parts of the system are likely to occur at a certain time. This can cause energy and economic losses. These can be minimized or largely avoided with the right monitoring approaches. System failures are not easily noticed without performance checks, since the auxiliary heating system always backups the hot water supply. Furthermore, changing weather circumstances and hot water demand make a prediction of the energy yield difficult. During the last decades several methods for monitoring have been developed [2-14], however, so far they have mainly been applied in research and demonstration projects.

Several terms will be distinguished here. Monitoring is defined as data logging of a variable amount of measurement data, however this data is not automatically analysed, so it does not automatically lead to a failure declaration. In a failure detection method, measurement data is automatically analysed and in case of a malfunction a failure indication follows. Failure identification or localisation goes further in that it requires the identification of the type of failure. This will make reparation much easier.

In this paper a partial Multi-Criteria Analysis (MCA) will be used to describe the performance of several monitoring methods for solar thermal systems. The MCA procedure will be described in chapter 2, consecutively the monitoring methods will be described in section 3. The results of the MCA will be given in section 4, conclusion and discussion are described in section 5.

2. Method