Fully tapping adjustable load demand response capabilities is necessary to guide massive and dispersed demand-side resources to participate in power dispatching operations. This study considers intelligent buildings (IBS) as the research object and proposes an adjustable load demand response capability assessment method. First, response capability assessment indicators, including adjustment power, adjustment response time, and adjustment sustainable time, are established based on the technical characteristics of adjustable loads. Second, the heating, ventilation, and air conditioning system model is established for IBS. Based on this, an intelligent building is used as an adjustable load unit for parameter identification using the recursive least squares estimation method. Finally, the response capability of the intelligent building is assessed based on the identified data and established response capability assessment indicators that can provide support for power dispatch.
The continuous increase of green energy supply on the power generation side gives rise to green power consumption challenges, and the rapid growth of emerging loads on the consumer side makes a large amount of adjustable load resources to be tapped. A green consumption pattern that coordinates production and consumption contributes to sustainable development. Considering the two-way interaction between sources and loads, the article designs a credit points response mechanism for green and friendly electricity consumption by users. The proposed credit points mechanism aims to increase the green power usage on the power generation side, and at the same time, explore the adjustable load on the user side to guide the users to use electricity in a friendly way. Based on the credit points response mechanism, a user green-friendly power consumption regulation model is constructed with the objective of maximizing the points. A park microgrid is selected as the research object, and the simulated annealing particle swarm optimization algorithm is used to solve the changes in the user load after the implementation of the credit points mechanism. The research results show that after implementing the green-friendly electricity consumption credit mechanism, the photovoltaic consumption in the park increased from 80,125.21 kW to 87,345.69 kW, and the photovoltaic consumption rate increased by 6.92%. Through optimized scheduling, users have obtained 50,082.57 green electricity points and 5923.68 demand response points. After considering the role of energy storage, the photovoltaic consumption capacity was further increased to 95,854.69 kW, the photovoltaic consumption rate reached 91.69%, and the total points increased to 61,650.62. This mechanism effectively smooths out peak loads and promotes efficient consumption of green electricity. The findings validate the rationality and effectiveness of the proposed credit points mechanism, which has certain reference value for green power consumption certification and user friendly power usage.
The full penetration of electric vehicle (EV) is the support for China's dual carbon goal of decarbonizing urban transportation. However, the inadequate layout of EV service facilities, especially the battery swapping station network (BSSN), has hindered the development of EV in cities and the promotion of decarbonization of urban transportation. Therefore, in this paper, a novel BSSN two-stage layout model for EV is proposed by combining the characteristics of urban road net. At the first stage, a comprehensive utility model is constructed to select the candidate site nodes of BSSN by combining the node characteristics of urban road net based on Social Network Analysis (SNA). At the second stage, a layout model of BSSN is proposed to quantify the battery swapping demand and provide the optimal solution of BSSN, based on the comprehensive utility and the cost objective of the battery swapping station (BSS). Finally, the BSSN optimal solution is provided based on the layout model to cover the swapping demand within the Fourth Ring Road of Beijing as a case. In the case study, the consideration of multi-objective is proved to be effective. The optimal service radius for BSS is 5-7 km and 399,127.2t carbon emission reductions that can be generated based on this plan. In addition, the sensitivity analysis of BSS service radius, minimum station number and EV scale is carried out. The layout model of BSSN introduces comprehensive utility into the road net and optimizes the location of BSSN from the perspective of urban planning, and it is of significance to the planning of urban transportation.
Thermal effects play a crucial role in cavitating flows involving thermo-sensitive fluids, particularly in high-pressure fuel injection systems, hydrofoils, and aerospace propulsion, where phase transition thresholds are approached. Unlike isothermal cavitation in water at room temperature, the significant variations in vapor pressure and liquid-vapor density under thermal conditions introduce complex dynamics that profoundly influence bubble growth, phase transition mechanisms, and overall system performance. While cavitation models based on bubble growth theory are commonly employed to model inertia-driven cavitation phase change processes, the modeling of thermal-diffusion-driven or coupled-phase transition processes in thermo-sensitive fluids remains a challenge. This study presents a comparative analysis of multistage bubble growth dynamics under varying thermodynamic conditions. A novel quantitative approach is introduced, incorporating key parameters such as the relative error rate between different phase change models and the critical time, thereby facilitating the differentiation of bubble growth stages and enabling a detailed assessment of thermal effects. The results show that the second critical time is approximately 250 times the first critical time. Based on an analysis of the bubble growth process, the Mikic-Rohsenow-Griffith-Theofanous-Patel (MRG-TP) bubble growth model, which accounts for both inertial forces and thermal diffusion and is applicable across all stages of bubble growth, is integrated into the cavitation phase change model. Consequently, a new generalized thermodynamic phase transition model (TPTM) is developed for simulating thermo-sensitive fluid cavitating flows. The accuracy of this model is validated through comparisons with experimental data on cryogenic liquid nitrogen cavitation flow around hydrofoils.
The spatiotemporal energy-shifting and moving flexibility of mobile energy storage (MES) can be explored to effectively support the operation security and resilience of distribution network. With this regard, a discrete multilayer spatiotemporal coupled network is first established to model the power charging/discharging, energy-shifting and moving-trajectory coupling of MESs; then a two-stage optimal operation strategy of distribution network simutaneously considering the operation flexibility of MESs and distributed generators (DGs) is proposed. In the first stage, the optimized connecting points of MES and optimal power output of DGs are optimized to achieve the minimum distribution network load shedding cost during fault; in the second stage, MES moving trajectory and charging/discharging power, as well as the power output of DGs are collaboratively optimized to minimize the comprehensive operation cost of the distribution network in post-fault. Finally, the effectiveness of the proposed two-stage optimal operation strategy considering MES flexibility was verified by the IEEE 33-bus network. Compared to scheme 2 with location-fixed energy storage and scheme 3 with DGs, the proposed approach reduced the load shedding costs by 4.5% and 2.2%, with the distribution network comprehensive operation costs decreased by 3.7% and 10%, respectively.
Abstract High penetration of uncertain wind power generation brings challenges to power system operational security and economy. Here, an adjustable box uncertainty set is first presented to characterize the spatiotemporal correlation of multiple wind power generations. Afterward, a two‐stage risk‐adjusted robust energy dispatching model is established as a min‐max‐min optimization problem for active distribution networks operating with high penetration wind power generation. Then a three‐layer Nest Column and Constraint Generation (NCCG) decomposition approach is designed to efficiently solve the proposed model. Finally, the proposed two‐stage adaptive robust optimization model and NCCG approach are validated by a distribution network with uncertain wind power generations, and the simulation results indicate the distribution network operation economy could be compromised with robustness by adjusting the conservative regulation parameter.
5G base stations have experienced rapid growth, making their demand response capability non-negligible. However, the collaborative optimization of the distribution network and 5G base stations is challenging due to the complex coupling, competing interests, and information asymmetry among different stakeholders. In this paper, a distributed collaborative optimization approach is proposed for power distribution and communication networks with 5G base stations. Firstly, the model of 5G base stations considering communication load demand migration and energy storage dynamic backup is established. Afterward, a collaborative optimal operation model of power distribution and communication networks is designed to fully explore the operation flexibility of 5G base stations, and then an improved distributed algorithm based on the ADMM is developed to achieve the collaborative optimization equilibrium. Finally, the effectiveness of the proposed distributed collaborative optimization model is validated by a modified IEEE 33-bus power distribution and communication networks with 5G base stations.
B-site Co -Fe-based perovskite materials have shown potential as ceramic cathodes in solid oxide electrolysis cells (SOECs) for direct CO 2 electrolysis. Nevertheless, their operational utilization is constrained by insufficient catalytic activity and durability in pure CO 2 . In this work, F - doping and A-site deficiency strategies are proposed and La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3 (LSCF), F -doped La 0.6 Sr 0.4 Co 0.2 Fe 0.8 F 0.1 O 2.9 (LSCFF) and A-site deficiency combined with F-doped (La 0.6 Sr 0.4 ) 0.95 Co 0.2 Fe 0.8 F 0.1 O 2.9 (LSCFF95) are prepared as cathode materials for CO 2 electrolysis. After doping with F - , the valence states of both Co and Fe elements decrease, while generating more oxygen vacancies. With the further introduction of A-site deficiency, the valence states further decrease, resulting in more oxygen vacancies. The enhancement of oxygen vacancies leads to an improvement in the oxygen surface exchange coefficient ( k chem ) and the bulk diffusion coefficient ( D chem ) of LSCFF and LSCFF95 materials. At 800 degrees C, k chem of LSCFF and LSCFF95 reach 24.21 x 10 -4 and 33.54 x 10 -4 cm s - 1 , respectively, which are 51% and 110% higher than the value of 16.03 x 10 -4 cm s - 1 for LSCF. Meanwhile, the D chem values of LSCFF and LSCFF95 are measured to be 37.52 x 10 -5 and 47.81 x 10 -5 cm 2 s -1 , respectively. These values are 49% and 90% higher than the value of 25.22 x 10 -5 cm 2 s -1 for LSCF. Consequently, F - doping and A-site deficiency significantly increase the electrochemical performance. For example, F - doping decreases the polarization resistance ( R p ) value of the SOEC with LSCF cathode from 0.19 Omega cm 2 to 0.13 Omega cm 2 at 800 degrees C. Furthermore, when the A-site deficiency is introduced, the R p value is further reduced to 0.11 Omega cm 2 . Therefore, the SOEC with the LSCFF95 cathode exhibits excellent performance. Specifically, it has demonstrated an impressively high current density of 2.85 A cm -2 at 800 degrees C and 1.5 V, and has exhibited exceptional durability over prolonged operation.
Within the transportation sector, hydrogen fuel cell vehicles represent the most efficient method of harnessing hydrogen energy. The establishment of a hydrogen refueling facility network that encompasses the entire cycle of hydrogen energy production, transportation, and application contributes to the effective promotion of hydrogen fuel cell vehicles. However, existing research predominantly focuses on hydrogen production and the conversion of refueling stations, neglecting the economic and stability considerations of the full-cycle use of hydrogen energy. This study proposes a hydrogen refueling facility network planning model that utilizes hydrogen energy throughout its full cycle. Firstly, a hydrogen demand estimation model based on the application scenarios of hydrogen fuel cell vehicles which clarifies the hydrogen requirements in each scenario is proposed. Regional hydrogen energy needs and the costs of building hydrogen refueling stations are then considered, using a set cover model to optimize the overall layout of existing and new refueling stations. Finally, taking into account the supply costs of renewable energy hydrogen production points and regional supply stability, a hydrogen refueling facility network planning model is constructed. The Beijing-Tianjin-Hebei region is selected as a case study, establishing a large-scale hydrogen supply network based on renewable energy hydrogen production, transportation, and usage.
Protonic ceramic fuel/electrolysis cells (PCFCs/PCECs) show great potential for the efficient and reversible conversion of chemical and electrical energy. The electrolyte and oxygen electrode interface in PCFCs/PCECs is particularly crucial for achieving optimal performance. In this work, Ba evaporation and Y enrichment phenomena are investigated on the surface of a BaCe0.7Zr0.1Y0.2O3-delta electrolyte during high-temperature sintering. Furthermore, an interfacial engineering strategy is proposed by removing the non-stoichiometric perovskite surface layer via a facile polishing process to enable a comparative analysis of the detrimental effects of the undesired layer on the cell performance. Polishing the electrolyte leads to a remarkable -50 % enhancement in PCFC performance and -120 % enhancement in PCEC performance at 650 degrees C. Electrochemical impedance spectroscopy coupled with distribution of relaxation time analysis reveals that the improved performance of the polished cell can be attributed to reductions in both the ohmic and polarization losses, with the latter primarily influenced by the facilitated oxygen electrode reactions. This interfacial engineering approach restores the high proton conductivity at the electrolyte surface and strengthens the physical and chemical adhesion between the electrolyte and the oxygen electrode. Our findings shed light on the existing challenges at the electrolyteelectrode interface and underscore the significance of interfacial engineering in enhancing PCFC/PCEC performance.
Large-scale access to distributed energy resources leads to new energy consumption problems and safe operation risks in the power system. Virtual power plants and shared energy storage are effective ways to promote the flexible consumption of distributed energy resources and improve the reliability and economy of power system operation. Based on the concept of sharing economy and considering the complementary characteristics of source and load resources between different virtual power plants, this paper focuses on the optimisation of shared energy storage and multi-virtual power plant operation. Firstly, distributed wind power, distributed photovoltaic and flexible load resources are aggregated into virtual power plants to analyze the cooperative operation mode of shared energy storage and multi-virtual power plant systems. Secondly, a two-layer decision model for shared energy storage configuration and multi-VPP system operation optimisation is constructed, with the upper model solving the optimal energy storage configuration scheme by maximising the revenue of the shared energy storage operator, and the lower model optimising the multi-VPP system operation strategy by minimising the total operation cost, the maximum amount of new energy consumed, and the minimum amount of carbon emission as the multi-objective optimisation strategy. Finally, a simulation analysis is carried out, and the results show that compared with the independent operation mode of each virtual power plant, the model proposed in this paper increases the annual profit of the shared energy storage operator by 7180 & YEN;, reduces the operating cost of the VPP system by 7.08%, improves the rate of renewable energy consumption by 0.82%, and reduces the annual carbon emission by 54.91 t, which realizes the mutual benefits of the virtual power plant and the shared energy storage.
The large-scale development of household photovoltaic in rural areas increases grid operation challenges and leads to higher costs for its access to the grid. To promote self-generation and self-consumption of photovoltaic, this paper investigates methods to enhance the photovoltaic consumption potential in village microgrid. An electric vehicles orderly charging scheduling model and an energy storage configuration model are proposed respectively, solved by CPLEX tool and genetic algorithm, and the photovoltaic consumption under different charging scenarios is compared with disordered or orderly charging and with or without energy storage. The results show that electric vehicles orderly charging scheduling not only reduces the load peak-valley difference, but also increases the photovoltaic consumption, and the configuration of energy storage enhances the photovoltaic consumption potential higher than electric vehicles charging scheduling, but its investment cost is larger, and it needs to be reasonably planned for its capacity. Comprehensive analysis suggests that load management and energy storage configuration co-optimization is the optimal choice to increase photovoltaic consumption. This study verifies the potential of load management and energy storage configuration to enhance household photovoltaic consumption, which can provide an application reference for the sustainable development of household photovoltaic and village microgrid.
Perovskite oxides show great promise as an alternative catalyst to the conventional nickel cermets for CO2 reduction reactions (CO2RR) in solid oxide electrolysis cells (SOECs) owing to their advantages of redox stability and coking resistance. Nevertheless, practical applications of these oxides are prevented largely by their poor CO2RR activities. Herein, a novel donor and acceptor co-doped nonstoichiometric double perovskite, La0.3Sr1.55Fe1.5Ni0.1Mo0.4O6-delta (LSFNM), is developed with in situ exsolved FeNi3 nanoparticles to efficiently catalyze CO2RR in SOECs. Pure CO2 electrolysis over the impregnated FeNi3@LSFNM catalysts is evaluated on two types of SOECs-one with thin (ZrO2)(0.89)(Sc2O3)(0.1)(CeO2)(0.01) (SSZ) electrolytes supported on 430L alloys and the other with thin La0.9Sr0.1Ga0.8Mg0.2O3-delta (LSGM) electrolytes supported on impregnated SmBa0.5Sr0.5CO2O5+delta (SBSCO)@LSGM anodes, producing unprecedently high current densities of 2.84 A cm(-2) for the former and 3.07 A cm(-2) for the latter at 1.5 V and 800 degrees C. Experimental analysis and density-functional theory (DFT) calculations reveal collective synergistic catalysis of oxygen vacancies (V-(O)double over dot), the doping Ni2+ ions and FeNi3 nanoparticles via the cooperative V-(O)double over dot-O(CO2), and Ni(II)-C(sp) and Ni(0)-O(CO2) interactions in LSFNM, not only facilitating CO2 chemisorption on oxygen vacancies but also destabilizing and dissociating surface carbonates in the vicinity of FeNi3 spontaneously into CO.
Praseodymium-based perovskites, like Pr0.4Sr0.6Ni0.2Fe0.7Mo0.1O3-delta (PSNFM), has excellent bifunctional electrocatalytic activity, making it suitable as a semiconductor material for reversible single-layer cell (RSLC) device, which can realize the electrochemical energy conversion and storage. To improve the electrocatalytic activity of PSNFM, Pr0.4Sr0.6Ni0.2Fe0.7Mo0.1F0.1O2.9-delta (F0.1-PSNFM), and (Pr0.4Sr0.6)0.95Ni0.2Fe0.7Mo0.1F0.1O2.9-delta (F0.1(PS)0.95NFM) perovskite oxides are synthesized. In H2 atmosphere, it is found that NiFe10.8 alloy is deposited on the surface of reduced PSNFM and reduced F0.1-PSNFM, while NiFe alloy is deposited on the surface of reduced F0.1-(PS)0.95NFM, suggesting that the presence of A-site defects promoting the precipitation of metallic Ni. Moreover, the concentration of oxygen vacancies can be increased by both A-site defects and F- doping, which results from the reduction of average valence states of Ni, Fe, and Mo. This indicates that F0.1-(PS)0.95NFM has the highest number of oxygen vacancies serving as oxygen reduction reaction (ORR) sites and the reduced F0.1(PS)0.95NFM exhibits the best hydrogen oxidation reaction (HOR) activity. For HOR, the rate-determining steps (RDS) on reduced PSNFM series semiconductor materials are charge transfer reaction and ORR process is controlled by a mixture of two elementary steps: the conversion of adsorbed oxygen to lattice oxygen and the reduction of O to O-. The combination of A-site defects and F- doping has been found to have a synergistic effect, leading to a significant enhancement of the discharge power density in single-layer fuel cell (SLFC) as well as the water electrolysis current density in single-layer electrolysis cell (SLEC).
The number of retired batteries from electric vehicles continues to grow, which not only puts pressure on battery recycling but also leads to environmental pollution. By replacing the conventional batteries with the second-use batteries in configuring the energy storage system, the investment cost can be effectively reduced, and the issue of retired batteries disposal can be alleviated. In view of this, the paper investigates the quantification of the environmental benefits of second-use batteries, and comprehensively evaluates the second-use batteries energy storage system (SUBESS) in terms of environmental and economic benefits. The energy storage configuration model aimed at maximizing annual net income is developed, and the improved particle swarm optimization algorithm is applied to determine the optimal capacity and power in a photovoltaic (PV) pilot area. Under the same capacity condition, several evaluation indexes are used to compare the economics of the SUBESS with the conventional batteries energy storage system (CBESS). The results show that: (1) Compared to end-of-life disposal of batteries, secondary utilization will yield greater environmental benefits. (2) When the cost per unit capacity of the SUBESS is lower than 63.99% of the CBESS, the economy of the SUBESS is better than that of the CBESS. Finally, targeted suggestions are made to further promote the scale development of the retired batteries secondary utilization industry. The research results not only validate the economic feasibility of the SUBESS but also have significant environmental benefits, which can provide a reference for potential investors.
In this paper, a day-ahead active and reactive power coordinated optimization strategy for active distribution networks (ADNs) with dynamic reconfiguration and soft open point (SOP) is proposed. Firstly, the model of dynamic reconfiguration of distribution networks and SOP are established. Afterward, the day-ahead active and reactive power coordinated optimization model of ADNs is established to minimize ADNs daily operation cost with consideration of the operation constraints of distributed generation, reactive power compensation equipment, dynamic reconfiguration of distribution networks and SOP. Then the model is converted into a mixed-integer second-order cone programming problem by using the Big-M method and the second-order cone relaxation and afterward is solved by YALMIP and CPLEX. Finally, the effectiveness of the proposed second-order cone optimization model is validated by a modified IEEE 33-node power distribution network.
Increased urbanization has caused problems such as increasing water consumption and the continuous deterioration of the groundwater environment. It is necessary to consider the groundwater quality in the water resource optimization system and increase the rate of reclaimed water development to reduce the amount of groundwater exploitation and achieve sustainable development of water resources. This study used the Daxing District, a region of Beijing’s southern plain, as an example to evaluate water quality by analyzing water quality data of surface and groundwater from 2012 to 2016 and actual water-use schemes from 2006 to 2016. Three groundwater extraction modes were set up based on NO3–N concentrations, water resources were optimized under three extraction modes, and water resource optimization schemes were determined based on the improved connection entropy. The results show that (1) the surface water quality was poor, and the proportion of V4 type water in the indexes of NH3–N and chemical oxygen demand (COD) was the largest. The surface water can only be used for agricultural irrigation. The pollution sources contributing most to NH3–N and COD were domestic and agricultural pollution sources. (2) The groundwater quality was good. The NO3–N index was primarily type I–III water, accounting for 95.20% of the total samples. Severe NH3–N pollution areas were mainly in the northern region, and most regional groundwater can be used for various purposes. (3) Taking 2016 as an example, three groundwater exploitation modes were set to optimize water resource allocation, and the results showed that the rate of groundwater development and NO3–N pollution decreased significantly after optimization. (4) Connection entropy is an evaluation method that combines connection numbers and entropy, including identify, difference, and opposition entropy. As connection entropy being a kind of complete entropy, which can reflect the difference of the system in different states, based on the improved connection entropy, the connection entropies of optimal water resource allocation and actual water-use schemes were calculated. The connection entropy of groundwater exploitation mode 3 was less than that of groundwater exploitation modes 1 and 2 and actual water-use schemes from 2006 to 2016. Therefore, exploitation mode 3’s water resource optimization scheme was recommended. In the paper, satisfactory results have been obtained. As a kind of complete entropy, connection entropy has great research value in dealing with complex hydrological problems. This study’s research methods and outcomes can provide methodological and theoretical lessons for water management in freshwater-deficient areas.
To address the issue of low load recovery rate after distribution network failure, this paper proposes a load recovery strategy that considers the operation flexibility of mobile energy storage (MES) and distribution network reconfiguration. Firstly, the spatiotemporal position mobility and energy shifting properties of MES are analyzed. Constrained by the operating limits of DG and the static security of distribution network, a load recovery model based on MES flexible mobility and distribution network reconfiguration is established to minimize the total cost of load reduction, MES operation and DG generation. Afterward, the proposed load recovery model is transformed into a mixed integer second-order cone programming problem through linearization and solved by YALMIP toolkit. Finally, the simulations of IEEE 33-bus distribution system validated that the proposed load recovery strategy could fully exploit the spatiotemporal operation flexibility of MES to improve the load recovery rate and reduce the comprehensive operation cost of distribution network.
Geophysical survey of basal structure is important for providing information that can help us to understand the genesis of heat flow anomaly. Using data from the recently completed regional gravity survey, the field source edge detection and density interface inversion are used to obtain the Cenozoic basement morphology and fault distribution characteristics in Yanqing basin. Meanwhile, we calculate the heat flow of 17 boreholes from its temperature logging data and the regional thermal conductivity measurement data, then analyze the tectonic setting of heat flow anomaly. The result reveals that (1) The basement relief of Yanqing basin can be divided into "one uplift, four sags and one monoclinic", including Dongwuliying uplift, Yaojiaying sag, Zhanglaoying sag, Tiansongying sag, Zhuojiaying sag and Kangzhuang-Shenjiaying monoclinic. The Yanqing basin is mainly developed by two NE-trending faults, and six nearly SN-trending faults which connect with the recharge area of south and north mountains and cut NE-trending faults, forming a crisscross shallow fault network. (2) The average value of heat flow in Yanqing basin is 65. 8 +/- 113. 0 mW . m(2), which is slightly lower than that of Bohai Bay basin. There are two abnormal areas of high heat flow, one is the convergence area of Kangzhuang-Shenjiaying fault and Jinjiapu fault, the other is the transition zone between the Xizhuojiaying sag and the Dongwuliying uplift. (3) The heat flow anomaly in Yanqing basin is mainly related to the distribution and thickness variation of the Jixian carbonates with high thermal conductivity and the heat flow redistribution promoted by groundwater thermal activity which is controlled by multi-system shallow faults, while it has little relationship with the development of FoyukouHuangbaisi fault and the intrusion of Yanshanian granite.
The purpose of this paper is to provide new ideas and methods for the sustainable use of groundwater in areas with serious groundwater overexploitation and serious groundwater pollution. Geographic information systems (GIS) were combined with machine learning algorithms, water resources optimization technology, and groundwater numerical simulation to optimize the regulation of the groundwater table and quality beneath the Daxing District in the southern plain of Beijing. By collecting local consumption and supply data and observations of the groundwater table and quality in the connected aquifer beneath Daxing for the years 2006–2020, the corresponding water demands and groundwater impact were extrapolated for the years 2021–2025 based on the basis of the existing development model. Through the combination of GIS and machine learning algorithms, the NO3-N concentration of local groundwater monitoring points in wet years, normal years, and dry years were predicted. With respect to NO3-N pollution, three new groundwater exploitation regimes were devised, which we numbered 1 to 3. The optimal allocation of water resources was then calculated for wet year, typical year, and dry year scenarios for the year 2025. By comparing the water shortage, groundwater utilization rate, and NO3-N pollution under the new groundwater exploitation regimes, the optimal groundwater exploitation mode for the three different types of hydrological year was determined. The results indicate that NO3-N pollution was greatly reduced after the adoption of the optimal regimes and that the groundwater table demonstrated rapid recovery. These results can be of great help in realizing the management, supervision, and regulation of groundwater by combining GIS with machine learning algorithms.