Zhihong Zou

Entropy method for determination of weight of evaluating indicators in fuzzy synthetic evaluation for water quality assessment

Considering the difficulty of fuzzy synthetic evaluation method in calculation of the multiple factors and ignorance of the relationship among evaluating objects, a new weight evaluation process using entropy method was introduced. This improved method for determination of weight of the evaluating indicators was applied in water quality assessment of the Three Gorges reservoir area. The results showed that this method was favorable for fuzzy synthetic evaluation when there were more than one evaluating objects. One calculation was enough for calculating every monitoring point. Compared with the original evaluation method, the method predigested the fuzzy synthetic evaluation process greatly and the evaluation results are more reasonable.

Adaptive neuro fuzzy inference system for classification of water quality status

An adaptive neuro fuzzy inference system was used for classifying water quality status of river. It applied several physical and inorganic chemical indicators including dissolved oxygen, chemical oxygen demand, and ammonia-nitrogen. A data set (nine weeks, total 845 observations) was collected from 100 monitoring stations in all major river basins in China and used for training and validating the model. Up to 89.59% of the data could be correctly classified using this model. Such performance was more competitive when compared with artificial neural networks. It is applicable in evaluation and classification of water quality status.

Water quality evaluation based on improved fuzzy matter-element method

For natural water, method of water quality evaluation based on improved fuzzy matter-element evaluation method is presented. Two important parts are improved, the weights determining and fuzzy membership functions. The coefficient of variation of each indicator is used to determine the weight instead of traditional calculating superscales method. On the other hand, fuzzy matter-elements are constructed, and normal membership degrees are used instead of traditional trapezoidal ones. The composite fuzzy matter-elements with associated coefficient are constructed through associated transformation. The levels of natural water quality are determined according to the principle of maximum correlation. The improved fuzzy matter-element evaluation method is applied to evaluate water quality of the Luokou mainstream estuary at the first ten weeks in 2011 with the coefficient of variation method determining the weights. Water quality of Luokou mainstream estuary is dropping from level I to level II. The results of the improved evaluation method are basically the same as the official water quality. The variation coefficient method can reduce the workload, and overcome the adverse effects from abnormal values, compared with the traditional calculating superscales method. The results of improved fuzzy matter-element evaluation method are more credible than the ones of the traditional evaluation method. The improved evaluation method can use information of monitoring data more scientifically and comprehensively, and broaden a new evaluation method for water quality assessment.

Water quality evaluation of Haihe River with fuzzy similarity measure methods.

Fuzzy similarity measures, which are used to judge the closeness of two fuzzy sets, are presented to evaluate the water quality of the Haihe River. Based on the membership functions and coefficient of variation as the weights, four fuzzy similarity measures (including Lattice similarity measure, Hamming similarity measure, Euclidean similarity measure and the max-min similarity measure) are used to classify the 299 samples into the proper water quality standard ranks. The results are compared with the traditional distance discriminant methods. The calculation of two traditional distance discriminant methods (both Euclidean distance and absolute value distance) is also based on the use of coefficients of variation as the weights. Without the Lattice similarity measure, for this method loses some information, the correct assignment of samples classified into the same water quality ranks is 75.92% with the other three similarity measures and two distance discriminant methods. This result shows the reliability of the five methods. Only considering the three similarity measures, there were only 1.01% of the samples that did not classify to the same ranks, while the corresponding ratio of the two distance discriminant methods was 5.69%. The results of leave-one-out cross validation show that more than 88% of the samples are classified to the proper ranks, which demonstrates that the similarity measures are suitable to evaluate the water quality of the Haihe River.

Forecasting of dissolved oxygen in the Guanting reservoir using an optimized NGBM （1,1） model

An optimized nonlinear grey Bernoulli model was proposed by using a particle swarm optimization algorithm to solve the parameter optimization problem. In addition, each item in the first-order accumulated generating sequence was set in turn as an initial condition to determine which alternative would yield the highest forecasting accuracy. To test the forecasting performance, the optimized models with different initial conditions were then used to simulate dissolved oxygen concentrations in the Guanting reservoir inlet and outlet (China). The empirical results show that the optimized model can remarkably improve forecasting accuracy, and the particle swarm optimization technique is a good tool to solve parameter optimization problems. Whats more, the optimized model with an initial condition that performs well in in-sample simulation may not do as well as in out-of-sample forecasting.

Water Quality Assessment in the Harbin Reach of the Songhuajiang River (China) Based on a Fuzzy Rough Set and an Attribute Recognition Theoretical Model

A large number of parameters are acquired during practical water quality monitoring. If all the parameters are used in water quality assessment, the computational complexity will definitely increase. In order to reduce the input space dimensions, a fuzzy rough set was introduced to perform attribute reduction. Then, an attribute recognition theoretical model and entropy method were combined to assess water quality in the Harbin reach of the Songhuajiang River in China. A dataset consisting of ten parameters was collected from January to October in 2012. Fuzzy rough set was applied to reduce the ten parameters to four parameters: BOD5, NH3-N, TP, and F. coli (Reduct A). Considering that DO is a usual parameter in water quality assessment, another reduct, including DO, BOD5, NH3-N, TP, TN, F, and F. coli (Reduct B), was obtained. The assessment results of Reduct B show a good consistency with those of Reduct A, and this means that DO is not always necessary to assess water quality. The results with attribute reduction are not exactly the same as those without attribute reduction, which can be attributed to the α value decided by subjective experience. The assessment results gained by the fuzzy rough set obviously reduce computational complexity, and are acceptable and reliable. The model proposed in this paper enhances the water quality assessment system.

Quantificational analysis on progress of river water quality in China.

In order to understand the dynamic change of water quality in a specific period of time, a type of possibility transition matrix based on the theory of Markov process was established. The transition possibility with a weight to calculate the degree of absolute advancement was given based on the result of water quality evaluation. The concept of relative advancement was presented. It was used to evaluate the extent of water quality changed in a period of time. The method was used to calculate the degrees of relative advancement for 4 rivers in China, and the results were analyzed.

Descriptive Characteristics of Surface Water Quality in Hong Kong by a Self-Organising Map.

In this study, principal component analysis (PCA) and a self-organising map (SOM) were used to analyse a complex dataset obtained from the river water monitoring stations in the Tolo Harbor and Channel Water Control Zone (Hong Kong), covering the period of 2009–2011. PCA was initially applied to identify the principal components (PCs) among the nonlinear and complex surface water quality parameters. SOM followed PCA, and was implemented to analyze the complex relationships and behaviors of the parameters. The results reveal that PCA reduced the multidimensional parameters to four significant PCs which are combinations of the original ones. The positive and inverse relationships of the parameters were shown explicitly by pattern analysis in the component planes. It was found that PCA and SOM are efficient tools to capture and analyze the behavior of multivariable, complex, and nonlinear related surface water quality data.

An Enhanced K-Means Algorithm for Water Quality Analysis of The Haihe River in China

The increase and the complexity of data caused by the uncertain environment is today’s reality. In order to identify water quality effectively and reliably, this paper presents a modified fast clustering algorithm for water quality analysis. The algorithm has adopted a varying weights K-means cluster algorithm to analyze water monitoring data. The varying weights scheme was the best weighting indicator selected by a modified indicator weight self-adjustment algorithm based on K-means, which is named MIWAS-K-means. The new clustering algorithm avoids the margin of the iteration not being calculated in some cases. With the fast clustering analysis, we can identify the quality of water samples. The algorithm is applied in water quality analysis of the Haihe River (China) data obtained by the monitoring network over a period of eight years (2006–2013) with four indicators at seven different sites (2078 samples). Both the theoretical and simulated results demonstrate that the algorithm is efficient and reliable for water quality analysis of the Haihe River. In addition, the algorithm can be applied to more complex data matrices with high dimensionality.

Using discriminant analysis to assess polycyclic aromatic hydrocarbons contamination in Yongding New River

Yongding New River has been polluted by polycyclic aromatic hydrocarbons (PAHs) which are carcinogenic and mutagenic. In three periods (the abundant water period, mean water period, dry water period), ten sites (totally 30 samples) in Yongding New River were clustered into four categories by hierarchical cluster analysis (hierarchical CA). In the same cluster, the samples had the same approximate contamination situation. In order to eliminate the dimensional differences, the data in each sample, containing 16 kinds of PAHs, were standardized with normal standardization and maximum difference standardization. According to the results of the cubic clustering criterion, pseudo F, and pseudo t2 (PST2), the proper number of clustering for the 30 samples is 4. Before conducting hierarchical CA and K-means cluster analysis on the samples, we used principal component analysis to obtain another group data set. This data set was composed of the principal component scores which are uncorrelated variables. Hierarchical CA and K-means cluster analysis were used to classify the two data sets into four categories. With the classification results of hierarchical CA and K-means cluster analysis, discriminant analysis is applied to determine which method was better for normalization of the original data and which one was proper to cluster the samples and establish discriminant functions so that a new sample can be grouped into the right categories.