Implementation of the QD test method
3.1. Necessary steps
In order to implement the QD test methodology at LECS, three main aspects had to be dealt with:
a) Install an operational test circuit
The available test circuit was initially used for SS tests. It allows for the control of the collector inlet temperature at a constant value. Calibrated sensors were installed and a fixed stand for installation of the collectors was used. For measurement of irradiance on the collector aperture, two pyranometers were used, one with diffuse band for measurement of diffuse irradiance. Beam irradiance was calculated based on measurement of global and diffuse irradiance.
b) Develop a data acquisition programme
The data acquisition system was composed by a DMM equipment allowing reading of several sensors with voltage and resistance signals. A data acquisition program was developed using Visual Basic 6 language. The program’s purpose is to gather data from the measurement sensors, according to the methodology set out in the standard [1]. The measured data is recorded on a file with the collector name, test date and ORI extension. Files with ORI extension contain sensors readings in volts and ohms. The information on sensor readings is supplemented by information such as the date and time (hour, minute, second) and reading number.
The acquisition program also calculates physical values using the calibration parameters of each sensor and records them on a file with the collector name, test date and TPO extension.
The time step for data acquisition is 5 s, which is in agreement with the suggestions given in the standard.
c) Develop a tool for parameter identification
The software chosen for the development of the tool for parameter identification is Mathlab. With this software a pre-processing of the data collected with the data acquisition programme is also performed. This pre-processing generates files with mean values of relevant data for five minutes intervals. Generation of graphic for analyses of the data collected according to the recommendations of the standard is also performed. Example of these graphs can be seen in section
4., Fig. 1 of this work. This tool also includes a multilinear analyses of the data collected, in order to determine the characteristic parameters according to equation (4).
3.2. Multilinear regression
The following (conventional) multiple linear regression model describes a relationship between the k independent variables, xj, and the dependent variable Y
Y = Д) + Ax1 + Ax24 + Pa + є-> (6)
This model was applied to equation (4), heat balance equation for QD test method, for determination of the collector characteristic parameters (5). Since the parameter IAM for beam radiation, Keb(0), is dependent on the incidence angle, 0, an assumption for its functional form is needed. In a first approach this dependence was considered to be given by;
Keb(0)= 1 - b0(— -1 I cos 0
 |
 |
and equation (4) could be re-written as:
Characteristic parameters: Pi= F(xa)en ;p2= F(xa)enb0; p3= F(xa)enK0d ;p4=d; p5=c2; p6=c5
A programme was developed for determination of the characteristic parameters based on the measured data (5 minute average values).
The same programme also produces graphs for analyses of data variability, according to the EN 12975-2; section 6.3. These graphs are:
• Difference between mean fluid temperature and ambient temperature versus global irradiance;
• Beam versus global irradiance;
• Beam irradiance versus incidence angle;
Based on the calculated parameters and measured values, it calculates the power delivered by the collector using the collector heat balance model of equation (8) and represents it graphically for comparison with the measured power delivered by the collector during the test sequences.
