A Generalized Fuzzy Linguistic Model for Predicting Component Concentrations in an Optical Gas Sensing System

Motivated by environmental protection concerns, monitoring the flue gas of thermal power plant is now often mandatory due to the need to ensure that emission levels stay within safe limits. Optical based gas sensing systems are increasingly employed for this purpose, with regression techniques used to relate gas optical absorption spectra to the concentrations of specific gas components of interest (NOx, SO2 etc.). Accurately predicting gas concentrations from absorption spectra remains a challenging problem due to the presence of nonlinearities in the relationships and the high-dimensional and correlated nature of the spectral data. This article proposes a generalized fuzzy linguistic model (GFLM) to address this challenge. The GFLM is made up of a series of “If-Then” fuzzy rules. The absorption spectra are input variables in the rule antecedent. The rule consequent is a general nonlinear polynomial function of the absorption spectra. Model parameters are estimated using least squares and gradient descent optimization algorithms. The performance of GFLM is compared with other traditional prediction models, such as partial least squares, support vector machines, multilayer perceptron neural networks and radial basis function networks, for two real flue gas spectral datasets: one from a coal-fired power plant and one from a gas-fired power plant. The experimental results show that the generalized fuzzy linguistic model has good predictive ability, and is competitive with alternative approaches, while having the added advantage of providing an interpretable model.

A generalized fuzzy linguistic model for predicting component concentrations in an optical gas sensing system

Motivated by environmental protection concerns, monitoring the flue gas of thermal power plant is now often mandatory due to the need to ensure that emission levels stay within safe limits. Optical based gas sensing systems are increasingly employed for this purpose, with regression techniques used to relate gas optical absorption spectra to the concentrations of specific gas components of interest (NO_x, SO₂ etc.). Accurately predicting gas concentrations from absorption spectra remains a challenging problem due to the presence of nonlinearities in the relationships and the high-dimensional and correlated nature of the spectral data. This article proposes a generalized fuzzy linguistic model (GFLM) to address this challenge. The GFLM is made up of a series of “If-Then” fuzzy rules. The absorption spectra are input variables in the rule antecedent. The rule consequent is a general nonlinear polynomial function of the absorption spectra. Model parameters are estimated using least squares and gradient descent optimization algorithms. The performance of GFLM is compared with other traditional prediction models, such as partial least squares, support vector machines, multilayer perceptron neural networks and radial basis function networks, for two real flue gas spectral datasets: one from a coal-fired power plant and one from a gas-fired power plant. The experimental results show that the generalized fuzzy linguistic model has good predictive ability, and is competitive with alternative approaches, while having the added advantage of providing an interpretable model.

Designing an interactive laboratory experiment to investigate buoyancy

Laboratory classes provide a visual and practical way of supplementing traditional teaching through lectures and tutorial classes. A criticism of laboratories in our School is that they are largely based on demonstration with insufficient participation by students. This provided the motivation to create a new laboratory experiment which would be interactive, encourage student enthusiasm with the subject and improve the quality of student learning.

The topic of the laboratory is buoyancy. While this is a key topic in the first-year fluids module, the laboratory has been designed in such a way that prior knowledge of the topic is unnecessary and therefore it would be accessible by secondary school pupils. The laboratory climaxes in a design challenge. However, it begins with a simple task involving students identifying some theoretical background information using given websites. They then have to apply their knowledge by developing some equations. Next, given some materials (a sheet of tinfoil, card and blu-tack), they have to design a vessel to carry the greatest mass without sinking. Thus, they are given an open-ended problem and have to provide a mathematical justification for their design. Students are expected to declare the maximum mass for their boat in advance of it being tested to create a sense of competition and fun. Overall, the laboratory involves tasks which begin at a low level and progressively get harder, incorporating understanding, applying, evaluating and designing (with reference to Bloom’s taxonomy).

The experiment has been tested in a modern laboratory with wall-mounted screens and access to the internet. Students enjoyed the hands-on aspect and thought the format helped their learning.

The use of cheap materials which are readily available means that many students can be involved at one time. Support documentation has been produced, both for the student participants and the facilitator. The latter is given advice on how to guide the students (without simply giving them the answer) and given some warning about potential problems the students might have.

The authors believe that the laboratory can be adapted for use by secondary school pupils and hope that it will be used to promote engineering in an engaging and enthusing way to a wider audience. To this end, contact has already been made with the Widening Participation Unit at the University to gain advice on possible next steps.