The aim of the short term load forecasting is to forecast the electric power load for unit commitment, evaluating the reliability of the system, economic dispatch, and so on. Short term load forecasting obviously plays an important role in traditional non-cooperative power systems. Moreover, in a restructured power system a generator company (GENCO) should predict the system demand and its corresponding price for efficient decision making. The task of a forecasting engine is to find the relation of the inputs and outputs of the system and also predicts the outputs for a given inputs. Therefore, the accuracy of forecasting is highly affected by the inputs of the forecasting engine. This effect can be studied from two points of view; First, extracting the more informative inputs and second, reducing the dimension of input space, both make it possible to learn the forecasting network via more simple models with more generalization. As a result, a reduced informative input space leads to lower prediction error. In many previous load forecasting methods, the inputs have been selected empirically. In this manner, the more correlative factors with the load in the forecasting day have been chosen as the inputs. They are generally a combination of load history and weather conditions. Several researches are focused on mathematical approaches of the input selection which are mainly based on principal component analysis (PCA) method as well as some intelligent algorithms. In this paper, a manifold learning method namely Locally Linear Embedding (LLE) is proposed, aiming to extract more informative inputs and to reduce the dimension of input space for short term load forecasting. Among all methods based on manifold learning, it can be seen that LLE performs very well in extracting the electric load curve features. The aim of this paper is to analyze the features of the load curve for estimating this curve in future. The extensive computational experiments show that the extracted features by LLE results in less prediction error than two other methods. Furthermore, LLE acts faster and makes input dimension lower than the two other methods. In the following section we will discuss the LLE method. The LLE method finds the nonlinear relationships among features by mapping a locally linear manifold in the feature space. Extracting the more informative inputs by extracting the combinational features by finding the nonlinear dependences of the features, results in reducing the dimension of input space. The resulted inputs from feature extraction and dimension reduction are utilized for load forecasting. To examine the effect of the proposed feature extraction method on load prediction error, a hybrid prediction system is proposed which is a combination of a radial basis function (RBF) network and a fuzzy system. The RBF network is the core of the prediction engine and works with historical load data as its inputs. The fuzzy inference system is combined with the RBF network to incorporate the impact of temperature on load. The case studies are carried out on the real data of electric power load of Mazandaran area in Iran. The efficiency of the proposed forecasting engine is compared with three benchmarks, the artificial neural network, time series and neuro-fuzzy methods. Furthermore, the proposed input selection method (LLE) is compared with principal component analysis (PCA) and empirical selection of inputs. Simulation results with statistical significance analysis show that the LLE method with the proposed forecasting engine is superior to other input selection methods and forecasting engines in sense of lower input dimension and lower prediction error.

Study of vapor-liquid and liquid – liquid Equilibrium plays an important role in the design, optimization and control of separation processes. In this research phase equilibrium of binary systems (1-propanol, water and 1-propanol, ethyl acetate) also ternary systems (water, ethylene glycol, 2-ethyl 1-hexanol and water, ethylene glycol, 1-heptanol) using thermodynamic models of NRTL and UNIQUAC were studied. Also Adaptive Nero-Fuzzy Inference System (ANFIS) and group method of Data handling (GMDH-type neural network) were used for modeling of systems. The VLE graphs (temperature based on mole fraction of vapor-liquid phases) for binary systems and LLE graphs for ternary systems at various temperatures by thermodynamics models were drawn. Accuracy of thermodynamic models, ANFIS model and GMDH type-Neural Network for the binary and ternary systems studied and compared with experimental data. Comparison of results shows that NRTL models have good fitness with the experimental data and the minimum error is related to the ANFIS Statistical model.

Experimental solubility curve and tie-line data were obtained for a ternary system of water, phosphoric acid, and 1-undecanol at T = 303. 2 K and ambient pressure. The cloud point method was employed to determine the binodal curve data. The mass fraction compositions of each layer were explored by acid-base and the Karl Fisher titration methods accompanied by mass balance calculations. For the mentioned ternary system, a type-1 LLE behavior was observed. The reliability of the experimental compositions was verified using the Othmer-Tobias and Hand plots. The UNIQUAC and NRTL thermodynamic models were applied to correlate the equilibrium tie-line points. The experimental points were adequately regressed using the thermodynamic models. Distribution coefficients and separation factors were computed over the non-miscible region. For the separation process, separation factor should be greater than one. For the present system, this separation factor was suitable to apply separation treatment in all concentrations of phosphoric acid at the experiment temperature. The Katritzky parameters were utilized for LSER modeling of distribution coefficients of the target mixtures. The ternary system showed good fittings with the LSER model.

Persistent organic pollutants, (POPs), are vast distributed compounds in environment which are recognized as one of the global pollution problems. These groups of materials being dangerous due to their high stability are accumulated in animal tissues and occurring in the food chain. One of the major paths through which persistent organic combinations access to human body is consuming polluted foods, particularly, fishes. Among aquatic animals, trout as one of the mostly consumed fishes in Tehran’ s food basket was studied. In this study, two categories of persistent organic pollutants: Organochlorine pesticides (OCPs) including HCB, Dieldrin, Methoxychlor, α-, ϒ-Chlordane, α-, β-Endosulfan and o, p’-DDE, p, p’-DDE, o, p’-DDT, p, p’-DDT and the second group Polychlonitated biphenyls (PCBs) including seven PCB congeners which are caLLEd indicator PCBs (IUPAC nos.: 28, 52, 101, 118, 138, 153 and 180) were determined in trout by GC-MS/MS in MRM monitoring mode and LLE extraction. The average recoveries of OCPs and PCBs at five concentration levels (1, 2, 5, 10 and 20 ng/g for PCBs and 5 times for OCPs) were in the range of 73-112%. The relative standard deviations of POPs in fish were in the range of 1. 4-17. 9% for all of the concentration levels. The limit of detections (LODs) and the limit of quantitations (LOQs) were between 0. 6-8. 3 and 2-25 μ g/kg, respectively. The results indicated the presence of organochlorine pesticides in trout and the levels of p, p’-DDE and p, p’-DDT were within the range of < LOQ-12. 83 and < LOQ-10. 2 ng/g ww (wet weight), respectively. According to the results, OCPs residues were lower than maximum residue levels set by European Council Directives.

The effect of NaCl on the LLE data for the ternary system, water + 1-propanol + hexane, were determined at 298Kunder atmospheric pressure. The curve of solubility and tie-line data of this system were determined by experimental analysis. The experimental results showed that the addition of NaCl significantly affected the two-phase region of the ternary system, the enlargement of the two-phase region occurred with an increase of salt concentration in the initial aqueous phase. Distribution coefficients and separation factors were also calculated and compared at different NaCl concentrations. In addition, the experimental tie-line data were correlated with the NRTL model. Also, the parameters of this model were estimated with five nature-inspired metaheuristic solvers. These algorithms are Genetic Algorithms (GA), Monkey-Krill Herd Hybrid (MAKHA), Intelligent Firefly Algorithm (IFA), Cuckoo Search (CS), and Harmony Search Algorithm(HAS). The capabilities and limitations of these solvers have been analyzed. Results showed that all algorithms outperformed, in particular, MAKHA give the best efficiency-reliability for the prediction of LLE.