This paper introduces a new structure in neural networks called TD-CMAC, an extension to the conventional Cerebellar Model Arithmetic Computer (CMAC), having reasonable ability in time series PREDICTION. TD-CMAC, the conventional CMAC and a classical neural network model called Multi-Layer Perceptron (MLP) are simulated and evaluated for 1-hour-ahead PREDICTION and 24-hour-ahead PREDICTION of CARBON MONOXIDE as one of primary air pollutants. CARBON MONOXIDE data used in this evaluation were recorded and averaged at Villa station in Tehran, Iran from October 3thrd. 2001 to March 14th. 2002 at one-hour intervals. The results show that the errors made by TD-CMAC is fewer than those made by other models.

This paper predicts and compares the CARBON MONOXIDE (CO) concentration levels along Sembulan Road for years 2004 and 2014 using CAL3QHC air dispersion model at two major locations, i.e., at Sembulan Roundabout and Sutera Harbour Intersection, Kota Kinabalu, Sabah, Malaysia. The CO concentration “hot-spots” were also identified at Sutera Harbour Intersection, and the highest maximum 1-hr average ground level concentrations of CO modeled for Kpg. Air Sembulan located in the northeast of idling road was 9.33 ppm for year 2004. This study showed that there would be no extreme changes in CO concentration trends for year 2014 although a substantial increase in the number of vehicles is assumed to affect the level of CO concentrations. It was also found that the CO levels would be well below the Malaysian Ambient Air Quality Guidelines of 30 ppm for 1-hour Time-Weighted Average (TWA). Comparisons between the modeled and observed outputs using quantitative data analysis technique and statistical methods indicated that the CAL3QHC predicted results correlated well with measured data. It was predicted that receptors located near to the major intersection, in the long-term would be potentially exposed to relatively higher CO levels.

Nowadays, air pollution is the main environmental challenge in metropolises. Therefore, it is essential to monitor and forecast air quality parameters in urban areas. It depends upon various factors, including topography, climate, population and transportation network. The relationship between these special factors has been considered as a dynamic, the nonlinear and ambiguous phenomenon. In this study, an adaptive Neuro –fuzzy system and GIS have been used to extract knowledge of environment from data, in terms of fuzzy rules. These rules were used to predict and model CARBON MONOXIDE (CO) pollutant concentration. Tehran has been selected as the case study. The data gathered from six meteorological stations, for four consecutive years in summer, in this city were used separately to train the neural network. Fuzzy rules (Sugeno and Mamdani) were extracted for each station and then, using these rules; pollutant concentration was estimated. Having concentration PREDICTIONs at station points, log- Kriging was used to model the spatial concentration in the area selected as the case study. The results showed that average RMSE of all stations using Sugeno rules is 1.445 and using Mamdani rules is 1.374.

Separation of CARBON MONOXIDE, hydrogen and methane from gas mixtures is an important industrial process. In this work, the adsorption capacity of CARBON molecular sieve (CMS) for these gases was studied by volumetric method at temperature of 20oC and pressure in the range of 200- 3500 kPa. Maximum adsorption of CARBON MONOXIDE, hydrogen and methane was 6.98, 3.39 and 2.77 mmol.g-1, respectively. Adsorption isotherms of these gases by using Excel showed to be a Langmuir type adsorption. Gas adsorption rate was calculated and showed that CARBON MONOXIDE, hydrogen and methane adsorptions by CMS were based on the difference between their adsorption rates and followed a kinetic rate of pseudo second-order. The maximum volume of adsorption gas of micro pores and potential energy as the parameter for predicting the type of adsorption were calculated using Dubinin-Radushkevich (DR) method. Adsorption potential energy for CARBON MONOXIDE, hydrogen and methane were in the order of 7.95, 7.78 and 7.79 kJ.mol-1.

Background and Aim: The CARBON MONOXIDE is an important compound which can convert to different chemical materials which is harmful to environment and humans. Selecting the mechanism and absorbent to absorption pollutant gases is very important due to the efficiency. This study aimed to optimize the CARBON molecular sieves prepared which cause the adsorption of more CARBON MONOXIDE. Materials and methods: In this study, a walnut shell was used to prepare a CARBON molecular sieve and activate the CARBON was used physical method. The adsorbents were optimized during two modifying and coating processes after granulation. The adsorbents were modified at pH = 4. 5, 6. 5, and9 which layered with oil-kerosene and oil-thinner mixtures (1: 1 ratio). The adsorption capacity of CARBON MONOXIDE by the adsorbents was measured by the volumetric method under the same conditions (2g of adsorbent, 25 °, C temperature, and 2 bar pressure). Results: The highest efficiency among those with modified adsorbents for CMS with 0. 281 mmol CO/g adsorbents and among those coated with (Ac) oil-kerosene adsorbents for CMS with 0/591 mmol CO/g adsorbent and (Al-K) among the adsorbents coated with oil-thinner and among all adsorbents for CMS with 0. 858 mmol CO/g adsorbents. (Al-T) Conclusion: A CARBON molecular sieve can absorb and reduce the concentration of CO, which is an environmental pollutant. Optimization causes to increase the adsorption efficiency.