Impact of Improved Measurements on Performance of a Smart Thermal Energy System Huibert Verra, Jacob Henderson, John Dyer, and John N. Jiang D ecentralized electrical energy systems will play an essential role in the future smart grid. In this paper, we introduce a new industrial scale electric ther- mal energy system using a high-voltage electrode boiler, of which the level of electricity consumption can be adjusted as desired. It is also shown that, with appropriate instrumenta- tion and measurement algorithms, the energy consumption can be adjusted quickly and accurately. This feature makes such a thermal energy system an ideal candidate to deal with the variability of renewable energy, to respond to the prices in the electric power markets and support the frequency stability of the power grid. Recent specific applications for the boiler demand new methods of measurement for critical variables, which are needed to offer a fail-safe or fault-tolerant class in addition to more reliable measurements in general. Implementation of specific instrumentation, many of which require a redundant setup, and control software is necessary to provide this criti- cal level of operation. Using the instrumentation to measure variables such as temperature, pressure, conductivity, etc., a control algorithm was implemented to improve the overall performance of the boiler system. The new control method sig- nificantly improves the performance of this electric thermal energy system, reducing the mean power errors to 0 ± 0.5%. The smart electric thermal energy system presented in this paper provides a fundamental building block toward expanded use of high-voltage electrode boilers for load-fre- quency control, with the added benefit of thermal energy as a by-product for use by various consumers. The enhanced control algorithm is particularly relevant in electrical power applications with increased volatility from energy production by wind turbines. Decentralized Electrical Power Generation Historically, the generation of electrical energy has been ac- complished in large centralized facilities, often by means of natural gas, coal and nuclear or hydropower plants. Large facilities offer great advantages economically, but long transmission lines are needed, which can have a negative effect on the environment. As centralized power produc- tion increased, the distribution grid became so large that the power companies could not guarantee cheap and reliable electricity to more remote consumers. Increasing the central generation was no longer enough to achieve better efficiency; however, better efficiency might be achieved by implement- ing smaller generating units collocated with the consumers. A much-used decentralized generation method is cogene- ration, which is the use of an engine to generate both electricity and heat used for heating applications. Fig. 1 shows a typical layout of a co-generation grid. Another method of decentralized generation is to generate power with wind turbines. It was a method used for thousands of years but became obsolete after the Industrial Revolu- tion. In the early 1970s, during the first oil price shock, wind power generation made a comeback as a reliable and consis- tent power source by using other sources as back up [1]. The only pollution associated with wind power generation is dur- ing manufacturing and maintenance. Wind energy is experiencing the most substantial growth of all sources of power generation. A status report from the global Renewable Energy Council (REN) [2] shows that wind power has had an annual growth rate of more than 20%. In many areas, wind power is supplying as much as 20% of the total energy demand [3]. By the end of the year 2012, the to- tal installed wind power capacity worldwide was more than 282 GW, and 21% of this was generated in the United States (60 GW) [4]. Not only is the number of installations rapidly in- creasing, but also the capacity per installation has increased immensely [5]. In the 1980s, 300 kW turbines were state of the art; by the time of this writing in 2014, the Danish company Ves- tas has deployed an 8-MW turbine [6]. However, wind electric power generation systems, in particular those with variable- speed wind turbines, are very different from conventional thermal power generation, because they are not synchro- nous to the electrical frequency of the power distribution grid (mains frequency or grid frequency are synonyms for the fre- quency of electrical generation). Due to the design of modern February 2015 IEEE Instrumentation & Measurement Magazine 25 1094-6969/15/$25.00©2015IEEE