A Non-Deterministic Integrated Optimization Model with Risk Measure for Identifying Water Resources Management Strategy

Abstract

Water resources system planning often exhibits high modeling error and uncertainty. Uncertainty in system parameters as well as their interrelationships can strengthen the conflict-laden issue of water allocation among competing interests. In this study, a nondeterministic integrated optimization model with risk measure is developed for planning water resources management. It can (i) deal with complex uncertainties described as probability distributions, fuzzy sets, and their combinations, (ii) provide an effective linkage between the predefined policies and the associated economic implications, and (iii) reflect policymakers' preferences to the tradeoff between system benefit and economic loss. The developed model is then applied to planning water resources allocation of the Heshui River Basin (China), where 960 scenarios are analyzed under various uncertainty and risk measures. Results disclose that (i) not only uncertainties of fuzziness and randomness but also risk attitudes of decision makers have significant impacts on water-allocation scheme and system benefit; (ii) the selection of a suitable alternative among solutions under different α, μ and λ values is complicated; (iii) water shortage would occur when water availability is less than the promised target; (iv) agriculture would encounter most serious scarcity compared to municipal and industry; (v) the conflict between economic development and agricultural sustainability would be a challenged issue that would enforce the local authority to adjust water-allocation policy. The findings can provide superior fundamental understanding of the study basin to improve water-allocation decisions under complex uncertain condition.

Environmental and Resource Impacts from an Aggressive Regionalized Carbon Peak Policy

“…These interactions further complexities and uncertainties from various carbon policies. Policy interactions influence different industrial, political, and regional systems, and within these policies there are also multiple factorial interactions all contributing to various uncertainties. , These uncertainties and various relationships will require special attention for carbon policy analysis so that more accurate and globally optimized results can be used for broad-based regional and national carbon management. The policies are relegated not only to carbon policies but also to water policies.…”

A Fuzzy-Interval Dynamic Optimization Model for Regional Water Resources Allocation under Uncertainty

“…On account of parameter uncertainties and objective inconsistency in multi-objective programming [38][39][40], the FILP model can well handle the uncertainty parameters denoted by interval numbers, and also coordinate the conflicts among different objective functions by introducing membership function λ, which makes the resulting solutions more scientific and reliable. The model is summarized as follows [26]:…”

Assessing Distributed Solar Power Generation Potential under Multi-GCMs: A Factorial-Analysis-Based Random Forest Method

“…Climate change can influence the availability of solar energy resources, and climate variables such as sunshine duration, solar radiation intensity (SRI), temperature, and humidity can further affect the feasibility of the solar energy system. , Changes in climate variables are spatially heterogeneous, which can cause the spatial and temporal variations of solar power generation across regions . Global climate models (GCMs) are useful for future climate projections and relevant impact evaluations, such as atmospheric condition prediction, energy system planning, and hydrological modeling. , Due to the course spatial resolution, GCMs’ outputs cannot be used directly as inputs to local scale. , Random forest (RF) is desired for bridging the empirical relationship between large-scale atmospheric variables and local climate observations. , Previously, the RF method was widely used for cleaner energy generation, solar radiation and wind speed forecasting, water resources management, and vegetation classification, especially showing advantages in climate analysis and renewable energy sources prediction. , Srivastava et al used the RF model to forecast the solar radiation, and results indicated that the RF model has a good performance. Karasu and Altan proposed a high-performance recognition model for solar radiation based on RF with a feature selection approach to cope with nonlinear dynamics in time series.…”

“…17,18 Random forest (RF) is desired for bridging the empirical relationship between largescale atmospheric variables and local climate observations. 19,20 Previously, the RF method was widely used for cleaner energy generation, 21 solar radiation and wind speed forecasting, 22 water resources management, 23 and vegetation classification, 24 especially showing advantages in climate analysis and renewable energy sources prediction. 25,26 Srivastava et al 27 used the RF model to forecast the solar radiation, and results indicated that the RF model has a good performance.…”