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.

Evaluation of an alternative spectroscopic approach for aflatoxin analysis: Comparative analysis of food and feed samples with UPLC-MS/MS

Increasing research has highlighted the effects of changing climates on the occurrence and prevalence of toxigenic Aspergillus species producing aflatoxins. There is concern of the toxicological effects to human health and animal productivity following acute and chronic exposure that may affect the future ability to provide safe and sufficient food globally. Considerable research has focused on the detection of these toxins, based on the physicochemical and biochemical properties of the aflatoxin compounds, in agricultural products for human and animal consumption. As improvements in food security continue more regulations for acceptable levels of aflatoxins have arisen globally; the most stringent in Europe. These regulations are important for developing countries as aflatoxin occurrence is high significantly effecting international trade and the economy. In developed countries analytical approaches have become highly sophisticated, capable of attaining results with high precision and accuracy, suitable for regulatory laboratories. Regrettably, many countries that are affected by aflatoxin contamination do not have resources for high tech HPLC and MS instrumentation and require more affordable, yet robust equally accurate alternatives that may be used by producers, processors and traders in emerging economies. It is especially important that those companies wishing to exploit the opportunities offered by lucrative but highly regulated markets in the developed world, have access to analytical methods that will ensure that their exports meet their customers quality and safety requirements.

This work evaluates the ToxiMet system as an alternative approach to UPLC–MS/MS for the detection and determination of aflatoxins relative to current European regulatory standards. Four commodities: rice grain, maize cracked and flour, peanut paste and dried distillers grains were analysed for natural aflatoxin contamination. For B1 and total aflatoxins determination the qualitative correlation, above or below the regulatory limit, was good for all commodities with the exception of the dried distillers grain samples for B1 for which no calibration existed. For B1 the quantitative R² correlations were 0.92, 0.92, 0.88 (<250 μg/kg) and 0.7 for rice, maize, peanuts and dried distillers grain samples respectively whereas for total aflatoxins the quantitative correlation was 0.92, 0.94, 0.88 and 0.91. The ToxiMet system could be used as an alternative for aflatoxin analysis for current legislation but some consideration should be given to aflatoxin M1 regulatory levels for these commodities considering the high levels detected in this study especially for maize and peanuts

Implementation of a drive-by monitoring system for transport infrastructure utilising GNSS

Ageing and deterioration of infrastructure is a challenge facing transport authorities. In particular, there is a need for increased bridge monitoring in order to provide adequate maintenance, prioritise allocation of funds and guarantee acceptable levels of transport safety. Existing bridge structural health monitoring (SHM) techniques typically involve direct instrumentation of the bridge with sensors and equipment for the measurement of properties such as frequencies of vibration. These techniques are important as they can indicate the deterioration of the bridge condition. However, they can be labour intensive and expensive due to the requirement for on-site installations. In recent years, alternative low-cost indirect vibrationbased SHM approaches have been proposed which utilise the dynamic response of a vehicle to carry out “drive-by” pavement and/or bridge monitoring. The vehicle is fitted with sensors on its axles thus reducing the need for on-site installations. This paper investigates the use of low-cost sensors incorporating global navigation satellite systems (GNSS) for implementation of the drive-by system in practice, via field trials with an instrumented vehicle. The potential of smartphone technology to be harnessed for drive by monitoring is established, while smartphone GNSS tracking applications are found to compare favourably in terms of accuracy, cost and ease of use to professional GNSS devices.

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.