Now that Performance Map Tutorial: Creating a Performance Map for Heat Pump Water Heaters has begun, those interested in following along will benefit from having access to a companion data set. Reading the blog posts would be valuable because it gives exposure to the information, but having a companion set allows you to follow along as you read. This way you can write your own code based on what’s read in the blog posts, do the analysis yourself, see how it works, check your results, and leave the tutorial with the confidence that you can write a program to automatically generate a performance map of heat pump water heaters.
Announcing a Sample Data Set for the Tutorial
To help with that goal, I have published my data set in the Store. This is a manufactured data set containing data representing what happens in performance map testing of heat pump water heaters. It emulates the data sets you would receive from the laboratory and includes important measurements including ambient temperature, inlet and outlet water temperature, water temperature measurements at several depths in the storage tank, water flow rate, and electricity consumption. This data will form the basis of the project as we analyze the data to find the change in energy stored in the tank, identify the coefficient of performance (COP) of the heat pump, and create a performance map predicting the COP as a function of the ambient and storage temperatures.
The data set contains manufactured data representing three different tests. In each test, the water in the tank is pushed out and replaced with 72 deg F water. This is done until all of the water stored in the tank is 72 deg F. After the tank is set at the starting temperature, the ambient temperature is set to the intended air temperature. Once the ambient temperature is at the set temperature, the heat pump is engaged and allowed to bring the water temperature to the set temperature of 140 deg F. Measurements of the electricity tell us how much energy is consumed by the heat pump, and measurements of the water temperature in the tank tells us how much the energy stored in the tank has changed. Therefore, during each test we can calculate the COP of the device with changes in water temperature for a specific ambient temperature. There are three tests to provide this information at three different ambient temperatures.
Overview of the Tutorial Data Set
Figure 1 shows the water temperatures in the tank during the three tests. The red lines and text describe what is occurring during the testing. The vertical lines and text at the bottom break the entire test period into the three separate tests, showing the repeated nature of the tests. These three tests are each performed at different ambient temperatures. The same pattern is followed within each of the three tests, and is highlighted with the text in test 1. First the water in the tank is replaced with 72 deg F water. Then, when the water in the tank is all at that temperature, the heat pump is used to heat the water up to 140 deg F. Remember that this is a manufactured data set, not actual test data. Real test data will never be as perfect as this data set, but this set does capture all of the main concepts and allow practicing the techniques.
Figure 2 presents the ambient temperatures during the same test period. In this case the data is red, and the descriptive information is black. As in Figure 1, the data is divided into three sections representing the data from the three tests. The descriptive information within each test states the ambient temperature used for that test period. The three tests were done with 55, 70, and 95 deg F ambient air temperatures. Combined with the data during each individual test, tracking the water temperature as energy is added, this provides the information needed to determine the curve predicting the COP of the heat pump as a function of water temperature at three different ambient temperatures. These curves provide the basis of a performance map for the device.
The next post will begin the process of analyzing this data set, and providing detailed tutorials so you can follow along using this data set yourself. We will begin with splitting this single data set into three different files, one for each test. During the process, our program will study the data contained within each test and write a file name detailing the test, and providing all information needed for future analysis. This will be done using the techniques described in How to Identify the Conditions of Laboratory Tests and Split Large Data Files. Once this process is complete, future posts will detail how to write a program which analyzes those test files and creates the final performance map.