Chongli Di

A Four-Stage Hybrid Model for Hydrological Time Series Forecasting

Hydrological time series forecasting remains a difficult task due to its complicated nonlinear, non-stationary and multi-scale characteristics. To solve this difficulty and improve the prediction accuracy, a novel four-stage hybrid model is proposed for hydrological time series forecasting based on the principle of ‘denoising, decomposition and ensemble’. The proposed model has four stages, i.e., denoising, decomposition, components prediction and ensemble. In the denoising stage, the empirical mode decomposition (EMD) method is utilized to reduce the noises in the hydrological time series. Then, an improved method of EMD, the ensemble empirical mode decomposition (EEMD), is applied to decompose the denoised series into a number of intrinsic mode function (IMF) components and one residual component. Next, the radial basis function neural network (RBFNN) is adopted to predict the trend of all of the components obtained in the decomposition stage. In the final ensemble prediction stage, the forecasting results of all of the IMF and residual components obtained in the third stage are combined to generate the final prediction results, using a linear neural network (LNN) model. For illustration and verification, six hydrological cases with different characteristics are used to test the effectiveness of the proposed model. The proposed hybrid model performs better than conventional single models, the hybrid models without denoising or decomposition and the hybrid models based on other methods, such as the wavelet analysis (WA)-based hybrid models. In addition, the denoising and decomposition strategies decrease the complexity of the series and reduce the difficulties of the forecasting. With its effective denoising and accurate decomposition ability, high prediction precision and wide applicability, the new model is very promising for complex time series forecasting. This new forecast model is an extension of nonlinear prediction models.

Modeling and Dynamical Analysis of the Water Resources Supply-Demand System: A Case Study in Haihe River Basin

The relationship between water resources supply and demand is very complex and exhibits nonlinear characteristics, which leads to fewer models that can adequately manage the dynamic evolution process of the water resources supply-demand system. In this paper, we propose a new four-dimensional dynamical model to simulate the internal dynamic evolution process and predict future trends of water supply and demand. At the beginning, a new four-dimensional dynamical model with uncertain parameters is established. Then, the gray code hybrid accelerating genetic algorithm (GHAGA) is adopted to identify the unknown parameters of the system based on the statistic data (1998–2009). Finally, the dynamical analysis of the system is further studied by Lyapunov-exponent, phase portraits, and Lyapunov exponent theory. Numerical simulation results demonstrate that the proposed water resources supply-demand system is in a steady state and is suitable for simulating the dynamical characteristics of a complex water supply and demand system. According to the trends of the water supply and demand of several nonlinear simulation cases, the corresponding measures can be proposed to improve the steady development of the water resources supply-demand system.

Multi-scale analysis of streamflow using the Hilbert-Huang Transform

Purpose – The purpose of this paper is to simulate and analyze accurately the multi-scale characteristics, variation periods and trends of the annual streamflow series in the Haihe River Basin (HRB) using the Hilbert-Huang Transform (HHT). Design/methodology/approach – The Empirical Mode Decomposition (EMD) approach is adopted to decompose the original signal into intrinsic mode functions (IMFs) in multi-scales. The Hilbert spectrum is applied to each IMF component and the localized time-frequency-energy distribution. The monotonic residues obtained by EMD can be treated as the trend of the original sequence. Findings – The authors apply HHT to 14 hydrological stations in the HRB. The annual streamflow series are decomposed into four IMFs and a residual component, which exhibits the multi-scale characteristics. After the Hilbert transform, the instantaneous frequency, center frequency and mean period of the IMFs are obtained. Common multi-scale periods of the 14 series exist, e.g. 3.3a, 4∼7a, 8∼10a, 11-14...

Integrated assessment of water resources vulnerability under climate change in Haihe River Basin

Purpose – The purpose of this paper is to assess the water resources vulnerability (WRV) rationally in Haihe River Basin (HRB) using set pair analysis (SPA) theory. Design/methodology/approach – An improved intelligent set pair analysis (IISPA) model is established, in which intelligent SPA theory is introduced and the weights are determined by use of the maximum entropy principle and the improved analytic hierarchy process method. The index systems and criteria of WRV assessment in terms of water cycle, socio-economy, and ecological environment are established based on the analysis of sensibility and adaptability. Findings – The authors apply IISPA to the WRV assessment of seven administrative divisions in HRB. Results show IISPA can fully take advantage of certain and uncertain information compared with fuzzy assessment and topsis assessment models. For present situation, Shanxi, Shandong, Tianjing, Inner Mongolia, Hebei are higher, Henan and Beijing are the middle vulnerability. But Henan will become w...

Refined gray-encoded evolution algorithm for parameter optimization in convection-diffusion equations

Purpose – The purpose of this paper is to reduce the computational burden and improve the precision of the parameter optimization in the convection-diffusion equation, a new algorithm, the refined gray-encoded evolution algorithm (RGEA), is proposed. Design/methodology/approach – In the new algorithm, the differential evolution algorithm (DEA) is introduced to refine the solutions and to improve the search efficiency in the evolution process; the rapid cycle operation is also introduced to accelerate the convergence rate. The authors apply this algorithm to parameter optimization in convection-diffusion equations. Findings – Two cases for parameter optimization in convection-diffusion equations are studied by using the new algorithm. The results indicate that the sum of absolute errors by the RGEA decreases from 74.14 to 99.29 percent and from 99.32 to 99.98 percent, respectively, compared to those by the gray-encoded genetic algorithm (GGA) and the DEA. And the RGEA has a faster convergent speed than doe...

Assessment of residential district ecological value based on the FAHP

Setting up a reasonable ecological value evaluation system for residential quarters is important to improve peoples living level. An evaluating index system with contains 2 factors, 12 sub-factors, and 50 elements is established. Then we use FAHP in determining the weights of the factors and sub-factors, which can reflect the human thinking style. At last, we establish a multi-grade fuzzy synthetical evaluation model and apply it to the assessment of a residential district in Tianjin. The result of assessment is constructive and reasonable.