In order to study the influence of water and carbon constraint policies on coal power industry under different scenarios, the government can adjust policies timely according to the simulation results. In this paper, a system dynamics model of coal power industry development under water-carbon dual constraints is constructed. Eight provinces in the Yellow River Basin are selected as the research objects, and the year 2020 is taken as the base year and 2021–2030 is taken as the research time zone to carry out an empirical study. The results show that: (1) under the existing water and carbon quota allocation policy, the profit of coal and power industry in the Yellow River basin will decrease obviously, and the development pressure of coal and power enterprises in the Yellow River basin will increase. (2) Water-carbon constraint has obvious extrusion effect on coal power industry. According to the calculation in this paper, the extrusion capacity will reach 395.17 TWh. (3) The water and carbon quota policy does not constrain the coal power industry at the same time. The existing water quota allocation method matches the ‘2030’ water consumption target, but the carbon quota allocation scheme has weak constraint on the coal and power industry in the Yellow River Basin, so the carbon quota should be tightened. (4) After the tightening of carbon quota, the coal emission reduction technology should be upgraded and improved before 2025. After 2025, the coal emission reduction technology can be mainly restricted by quota. (5) New energy planning should be combined with its own development situation and quota allocation policy. When the amount of coal and electricity extruded by water-carbon constraint is not enough to meet the target of new energy installation, it should be planned according to its own development situation and extrusion amount. When the amount of coal power extruded by water-carbon constraint is enough to meet the target of new energy installation, the amount of coal extruded should be taken as the main planning basis.

Environmental Kuznets Curve (EKC) is one of the main tools to analyze the response relationship between water environmental quality and economic development. In classical EKC curve modeling process, the environmental index time series and economic index time series must be the same frequency. Because basin water-quality in China has been gradually monitored and released with high frequency, and most macroeconomic indicators are on annual basis, therefore, by the construction of three classes mix data polynomial equation of EKC curve, the mixing data modeling method of the response relationship between water quality and economic development is proposed to adapt to the situation. The mix data model not only avoids destroying original information due to simple low-frequency measurement of high frequency water quality data using classical model, but also through direct mining of high frequency water quality data, the lag time of water environment quality response to economic development can be obtained. To verify the validity of the model, according to the monitoring and statistical data of 2009-2017, a prefecture level city in Henan province of Haihe River Basin is taken as the research object, and EKC three classes mix data polynomial equation of chemical oxygen demand (COD) concentration and ammonia nitrogen concentration is established.

Under China’s dual carbon goals, the phase-out of coal-fired power units is imperative for the energy system transition. This study simulates the phase-out pathways of China's coal-fired power units from 2025 to 2060, analyzing the associated risks and spatiotemporal dynamics.Using survival analysis, we constructed a simulation framework that integrates historical empirical patterns with future demand scenarios to quantify phase-out drivers and assess relative risks.Temporally, the phase-out follows a three-stage pathway: gentle initiation, accelerated reduction, and final peaking.Spatially, the phase-out rhythm varies significantly across regions, falling into three types: “Early-Advancing and Later-Lagging” “Slow Start and Late-Surge” and “Mid-Stage Sprinting” Under three scenarios, the average phase-out age ranges from 37 to 39 years, with 13%–16% of units retired prematurely, risking asset stranding. Mechanisms like capacity pricing can alter these pathways by influencing the economic viability and thus the retirement schedule of units.It provides a quantitative basis for policymakers to anticipate phase-out trends across stages and regions and to formulate differentiated policies that balance energy security with the prevention of asset stranding.