Thermal Systems: Next Generation Control and Monitoring
By Ron Gehl
The International Energy Agency estimates that in 2012, worldwide production of solar thermal energy was 225 terawatt-hours (TWh), second only to wind power in total renewable energy produced. Approximately 16 TWh of that total was generated in the United States and Canada. How much of that total did your solar thermal system produce? Unfortunately for most system owners and operators, the answer is, “I really don’t know.” And without good monitoring data, how could you know?
Today, the web can tell you the instantaneous watt output of an individual PV panel on Aunt Edna’s roof,so it shouldn’t be too hard to do something similar for solar thermal energy, right? What follows is an appeal to advance toward more modern, efficient solar thermal systems by implementing better monitoring and control technology. This will be key to the continued development of solar thermal as one of the most effective means of displacing fossil fuel use, driven in part by the prospect of monetizing thermal output, and in part by the expectations brought about by the smartphone in the consumer’s pocket.
A Need for Better Monitoring
As I wrote in the July/August 2013 issue of Solar Today (“Commercial Applications Ramping Up”), larger solar thermal applications are becoming more prevalent, as power purchase agreement (PPA) models are being applied to thermal output, and states are beginning to incorporate renewable thermal energy in their renewable portfolio standards (RPS). Accurate and reliable monitoring of thermal output is a basic requirement for these situations, and industry has recognized that greater credibility will come with a common standard that addresses heat metering — in fact, financial institutions will demand it. Under the direction of James Critchfield of the U.S. EPA, ASTM International has organized subcommittee E44.25 to address this need, and many control/monitoring system providers and instrument manufacturers are actively involved in the standard’s development, to be released in 2014. State initiatives such as California’s performance-based initiative for solar thermal also require monitoring for all larger systems.
But monitoring is much more than just tallying heat output. Relevant, actionable data from a system enables a new model for solar thermal technology, one in which thermal energy is recognized as a service, not just some nebulous offset against fossil fuel consumption. Astute professionals and eager consumers will take advantage of this model, which has been so successful for PV. A good monitoring system allows the energy professional to easily determine that a system is running properly, and gives access to information that will help troubleshoot a malfunctioning system. Consumers (including commercial and industrial consumers) generally want to know overall system production, along with cost savings or environmental benefit derived from that production. And these days, everyone wants that information NOW. Professionals need that data pushed to them in the form of alerts to system malfunctions (via email or text message), along with remote access to real-time data that allow them to drill down to the performance of individual system components and sensors. An internet-accessible “dashboard” should permit the consumer to observe near real-time production and review the operating history.
Monitoring Best Practices
The best monitoring dashboards available from providers these days can be quite attractive and contain most of what the average consumer might be interested in. However, all the eye candy in the world won’t save a monitoring dashboard if the data being presented are just plain wrong. While a “sanity check” on the operation may be sufficient for small domestic water heating systems, larger commercial systems, particularly those where there is a financial incentive for the thermal output produced, must adhere to higher standards. Promising new approaches to heat metering are being developed (e.g., enthalpy-based measurement of storage volumes), but the most common form of heat metering employs data from a fluid flow meter and two matched temperature sensors (positioned before and after the heat exchanger) to calculate a heat transfer rate. To help minimize some of the common errors we have found in heat metering, following are some principles that should be adhered to for the proper measurement of heat produced, and the accurate presentation of data. A more complete treatment of some of these details and achievable degrees of accuracy will be available in the aforementioned ASTM standard.
Temperature Sensor Selection – First and foremost, make sure that temperature sensors involved in BTU calculations are matched, are suitably accurate for the application, and are appropriate for the monitoring device (calculator) being used. Thermistors and resistance temperature detectors (RTDs) are commonly used for this purpose, but it is a common misconception that one 10K Ohm thermistor will have the same characteristic curve as another (they’ll all have the same resistance at 25°C, but not at 80°C). Look at the sensor specification and make sure the measuring device will properly interpret it.
Temperature Sensor Placement – Sensors must be placed near the point of use (immediately upstream and downstream of the heat exchanger) and must accurately represent the temperature of the water or other heat transfer fluid being used. Tom Henkel, Ph.D., of Henkel Solar, Inc., who has designed and deployed some of the most advanced solar heating and cooling systems in theUnited States, emphasizes that accurate temperature measurement requires excellent heat transfer between the fluid and the sensor. He recommends that temperature sensors be placed in thermal wells immersed in the fluid at a location where there is turbulent flow, using thermal paste in the well to insure good heat transfer. He has found that the most accurate temperature measurement will occur at a right angle elbow in the pipe, with the thermal well installed parallel to and pointing “upstream” in relation to flow direction. Sorry, strapping a sensor to the outside of the pipe just doesn’t cut it.
Flow Meter Selection – While there is quite a variety of flow metering technology available over a wide range of accuracy specifications (and cost), a couple of basic facts to remember are that a) you’re not going to measure low flow rates accurately in large pipes, b) turbulence can throw things off and c) viscosity and density of the heat transfer fluid matter (a lot). In other words, don’t expect to measure 1 gallon per minute of 50 percent glycol solution in a 3” pipe using a flow meter located next to a pipe elbow. Make sure your flow meter will measure the range expected for your application, reduce your pipe sizing in the vicinity of the flow meter, and then (if applicable) make sure your BTU calculator can correct for glycol characteristics.
Electrical Noise/Signal Loss – Signals from RTD temperature sensors will quickly lose accuracy due to higher wire resistance as cable runs increase in length. Most thermistors are impacted less by wire resistance; however, thermistor signals are subject to AC power interference due to capacitive coupling on the cables. For longer cable runs or electrically “noisy” environments, temperature transmitters that output a 4-20 milliamp signal should be used, as these signals can be transmitted thousands of feet without signal degradation. Purely digital and/or wireless signal transmission is also being adopted more frequently.
Data Presentation – Dashboards must present information that is accurate for the system’s location and purpose. For consumer dashboards, correct assumptions about the local cost of electricity or type of fuel being offset are important for comparison purposes. But while consumer dashboards are fine for the consumer, the energy equivalent of miles driven in a car is probably not relevant to a professional installer or manager. Here is where the ability to alert to malfunctions or substandard performance becomes important. The monitoring system must flag clues to problems with the system: Why is there no flow? Unusually high collector temperatures? Can that delta-T be real?
A Need for Better Control
Solar thermal technology is inherently efficient, but we can do a lot better in putting the energy we capture to good use. A run-of-the-mill flat plate collector can harvest an impressive 60 percent of the sun’s energy that strikes it; in comparison, a cutting-edge, high-end PV panel will struggle to convert even 20 percent of that energy to electricity. Solar thermal is a fairly “mature” technology, and efficiency gains for the collectors at the core of these systems have only improved marginally in recent years. So where can the production gains come from? The answers lie in better managing the fluid flow through the collectors and optimizing the heat distribution through the balance of the system. It is not unusual to derive a 15 to 25 percent increase in overall system output when these aspects are improved; these strategies simply require more intelligent controllers to execute.
Practitioners are building increasingly sophisticated solar thermal systems, yet many larger systems are still saddled with fairly primitive, consumer-grade controls, or worse, a cobbled-together amalgam of single-purpose controllers that often operate at cross purposes. As a result, much of that great energy harvested at the collector is lost through poor management downstream. It’s time for the industry to recognize that efficiency gains derived from better control are the “low-hanging fruit” of solar thermal system improvement, and that the small investment required will pay back quickly. We provide an example below of how heat production in the collector loop can be maximized, but there’s plenty more to be gained in the management of downstream storage and heat distribution.
Enhancing Thermal Production
For the collector loop, the greatest gains in production efficiency are achieved by implementing some form of variable-speed control of the primary circulator pump(s). Solar irradiance varies significantly over the course of a day, and can swing wildly over short time intervals for the majority of us living in partly cloudy climates. The traditional fixed-rate, on/off control method for solar loops can throw away a lot of energy under these circumstances. Collectors operate most efficiently when energy is harvested continuously and fluid doesn’t stagnate in pipes leading to and from the collectors (which results in greater heat loss).
We find that continuously-variable flow rate adjustment to maintain a constant differential temperature (delta-T) across the collectors works best; this can be achieved through the standard industrial control algorithm known as proportional-integral-derivative (PID) control. It sounds like heavy math, but it’s widely available in industrial controllers and proven over decades of use. Variable-rate control requires paying a bit more for a circulator pump that can accept a speed signal; highly efficient electronically-commutated (ECM) pumps are ideal for this purpose and continue to come down in price.
Figure 1 provides a good example of the benefit that can be derived from variable-rate control under real-world conditions. We equipped two identical flat plate, two-panel racks with matched 4-20 mA temperature transmitters and vortex flow meters, operating one with conventional on/off differential temperature control, and the other with PID variable-rate, constant delta-T control. The graph focuses in on BTU output to a storage tank over a portion of a partly cloudy day. The variable-rate rack provides useful heat almost continuously, while the on/off rack produces it in shorter bursts. The result is 26 percent higher total heat output using the PID approach over the test day; and while benefits will not be as great on full-sun days, longer-term testing over several months showed a net gain of 13 percent over the full time period. That’s real value that can’t be ignored.
Better Monitoring and Control Come Together
It becomes clear that to maximize effectiveness, advanced monitoring and more intelligent control should be joined at the hip, and brought together in one smart, connected device. An integrated control/monitoring system is the most powerful tool available to efficiently manage solar thermal operations. Once the domain of expensive industrial supervisory control and data acquisition (SCADA) systems, these capabilities are now available in lower-cost, easy-to-use controllers. In the work we do at EOS Research, we stress the importance of real-time, live interaction with your systems, with a complete, remotely accessible view of device status and history. Imagine the value of checking in on a solar thermal system a couple of weeks after installation; not by traveling to that site that’s 200 miles away, but simply by peeking in from your office or home. Review recent operating data, tweak a few setpoints, and immediately see the results of those changes. Don’t like how the loads have been prioritized? Change it on the fly, and maybe you’ve earned a bit more on your PPA.
What stands to be gained is a lot more than bragging rights — it’s real dollars, real energy and real progress toward reducing fossil fuel use. With smarter controls and monitoring, we can demonstrate the full potential of the systems we build so that solar thermal energy becomes an even larger part of the energy mix.
Ron Gehl, P.E. is president of EOS Research, Ltd., providing control and remote monitoring solutions for renewable energy and environmental protection applications. He has nearly 30 years of experience in environmental engineering, and in the automation of processes that function at the boundary between industry and the natural world. Ron holds B.S. and M.S. degrees in engineering from Rensselaer Polytechnic Institute, and is Chair of the ASES Solar Thermal Division.