پديد آورندگان :
معظمي، صابر دانشگاه آزاد اسلامي - مركز تحقيقات علوم زيست محيطي - گروه مهندسي عمران، واحد اسلامشهر , نوري، روح اله دانشگاه تهران - دانشكده تحصيلات تكميلي محيط زيست , سليميان، محمد دانشگاه تهران - دانشكده تحصيلات تكميلي محيط زيست , مؤمني، محمودرضا دانشگاه تهران - دانشكده تحصيلات تكميلي محيط زيست , وصالي ناصح، محمدرضا دانشگاه اراك - گروه مهندسي عمران
كليدواژه :
ماشين بردار پشتيبان , آلودگي هوا , تحليل عدم قطعيت , تهران , مونوكسيد كربن
چكيده فارسي :
عدم قطعيت موجود در مطالعات مدل سازي آلودگي هوا، نتايج شبيه سازي را به شدت تحت تاثير قرار مي دهد. در اين راستا تعيين عدم قطعيت مدل هاي پيش بيني كيفي هوا، به دليل تاثيرگذاري بر سلامت افراد در معرض آلودگي، از اهميت ويژه اي برخوردار است. بنابراين در اين تحقيق اقدام به تعيين عدم قطعيت ماشين بردار پشتيبان (SVM) براي تخمين مونوكسيد كربن (CO) در ايستگاه پايش قلهك، در شمال تهران، شده است. براي تعيين عدم قطعيت مدل SVM، با تغيير الگوهاي انتخابي براي كاليبراسيون، به دفعات زياد واسنجي مدل انجام شد. اين امر با خروجي هاي مختلف در هر مرحله، به دليل حساسيت مدل به الگوهاي واسنجي انتخابي، همراه بود. سپس با استفاده از فاكتور عرض باند (d-factor) و درصد داده هاي اندازه گيري قرار گرفته در باند 95 درصد عدم قطعيت مدل (95PPU)، عملكرد SVM در فرآيند پيش بيني CO مورد ارزيابي قرار گرفت. نتايج اين تحقيق بيانگر عملكرد مناسب مدل SVM در پيش بيني غلظت روزانه CO در شهر تهران بود به گونه اي كه از ضريب تعيين (R2) معادل 0/89 و 0/88 به ترتيب در مراحل كاليبراسيون و صحت سنجي برخوردار بود. همچنين يافته هاي تحليل عدم قطعيت مدل SVM مشخص نمود كه مقدار d-factor و درصد داده هاي اندازه گيري محصور شده توسط 95PPU براي اين مدل در مرحله صحت سنجي به ترتيب معادل 0/74 و 76 درصد مي باشد. بر اساس شاخص هاي آماري d-factor و 95PPU، عدم قطعيت مدل SVM براي پيش بيني غلظت روزانه CO در تهران قابل قبول بوده به نحوي كه اين مدل از عدم قطعيت اندكي در پيش بيني اين شاخص كيفيت هوا در شهر تهران برخوردار بود.
چكيده لاتين :
Carbon monoxide (CO) is one of the main air pollutant parameters in the atmosphere of Tehran, Iran. Generally, it is difficult to predict and control the intensity of CO concentration because it essentially has nonlinear time-varying nature. Environmental control such as controlling the CO concentration level is recently regarded as one of the most important factors in environmental protections. This paper describes forecasting and - more specifically - determination of uncertainty in evaluating the CO concentration during the modeling process using a support vector machine (SVM) technique. Uncertainty of air pollution modeling highly affects the simulation results. In this regards, it is very important to determine the uncertainty of air pollution models due to consequences on the health of people exposed to the pollution. Therefore, this research aims to calibrate, verify, and also determine the uncertainty of SVM in the process of air pollution modeling in the atmosphere of Tehran. To achieve this goal, the SVM model was applied to predict arithmetic average of daily measured CO concentration in the atmosphere of Tehran. In this regard, the SVM model was calibrated and verified using six daily air pollutants including particulate matter (PM), total hydrocarbons (THC), nitrogen oxides (NOx), methane (CH4), sulfur dioxide (SO2) and ozone (O3) and also six daily meteorological variables including pressure (Press), temperature (Temp), wind direction (WD), wind speed (WS) and relative humidity (Hum). The data was collected from Gholhak station located in the north of Tehran, Iran, during 2004-2005. Thereafter, the best developed SVM model for predicting CO concentration was chosen based on determination of coefficient (R2). Finally, to determine the SVM uncertainty, the model was run many times with different calibration data. It led to many different results because of the model sensitivity to the selected calibration data. Then, the model uncertainty in the CO prediction process was evaluated using the width of uncertainty band (d-factor) and the percentage of measured data bracketed by the 95 percent prediction uncertainties (95PPU). Generally, the results confirmed the strong performance of the SVM model in predicting CO concentration in the atmosphere of Tehran. The predicted average daily CO concentrations by SVM model had a good agreement with the measured ones in the Gholahak air quality monitoring station. It was found that determination of coefficient for calibration and validation of SVM model were equal to 0.89 and 0.88, respectively. Furthermore, the results indicated that the SVM model has an acceptable level of uncertainty in predicting CO concentration. In this case, the level of d-factor and the percentage of measured data - bracketed by the 95PPU in the validation step - were 0.74 and 76, respectively. Therefore, obtained results indicated that the SVM model had an acceptable level of uncertainty in prediction of CO concentration. The proposed methodology is general, therefore; it is suggested to be applied for analyzing the uncertainty of SVM model in other fields of science and engineering.
