Threshold value calculation method for water resource population carrying capacity based on hierarchical water consumption evaluation

Abstract

This invention relates to the field of water resource management technology, specifically disclosing a method for calculating the population carrying capacity threshold of water resources based on hierarchical water use assessment. The method includes: Step 1, deploying a sensor network in the target area to monitor water resource status, climate change, socio-economic indicators, and residents' water use behavior in real time via edge computing units, and synchronizing the collected multidimensional data to the cloud; Step 2, using big data analytics and probability density methods, subdividing residential water use into eight aspects based on residents' water use behavior, and dividing residents' water demand into three levels and setting water use intervals; This invention enables water resource management to flexibly respond to changes in supply and demand by dynamically adjusting thresholds, introducing intelligent algorithms and reinforcement learning, and optimizing water resource allocation and water conservation policies; comprehensively considering regional characteristics and socio-economic factors, it provides support for formulating personalized water conservation policies and long-term implementation strategies for different regions, while simultaneously optimizing policy implementation in real time through an intelligent management platform.

Description

Technical Field

[0001] This invention belongs to the field of water resources management technology, specifically relating to a method for calculating the population carrying capacity threshold of water resources based on hierarchical water use assessment. Background Technology

[0002] Water resource carrying capacity is a key indicator for measuring the sustainable use of water resources in a region. However, traditional methods for analyzing water resource carrying capacity ignore the hierarchical nature of water demand, making it impossible to accurately predict and dynamically adjust the water resource carrying capacity.

[0003] In existing technologies, most calculations of water resource carrying capacity are based on traditional water resource supply and demand balance models. However, due to the lack of real-time data monitoring and dynamic adjustment mechanisms, these methods often suffer from inaccurate predictions and an inability to cope with complex climate changes and diverse water use behaviors. Furthermore, traditional water conservation policies are typically static and unilateral, failing to optimize and adjust in real time according to changes in water resource carrying capacity. The lack of comprehensive consideration of regional characteristics, water use behaviors, and socio-economic indicators also leads to insufficient accuracy in calculation results, making it difficult to meet practical application needs. This poses significant challenges to water resource prediction and management, especially in a rapidly changing social environment, making it difficult to accurately assess future water resource demand and population carrying capacity, thus affecting the scientific allocation and management of water resources.

[0004] Therefore, it is necessary to propose a water resource population carrying capacity threshold calculation method based on hierarchical water use assessment to solve the problem that the traditional water resource population carrying capacity threshold analysis and calculation in the existing technology is not accurate enough. Summary of the Invention

[0005] The purpose of this invention is to provide a method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The method for calculating the population carrying capacity threshold of water resources based on hierarchical water use assessment includes the following steps:

[0008] Step 1: Deploy a sensor network in the target area to monitor water resources, climate change, socio-economic indicators, and residents' water use behavior in real time through edge computing units, and synchronize the collected multi-dimensional data to the cloud.

[0009] Step 2: Using big data analysis technology and probability density method, residential water use is subdivided into eight aspects according to residents' water use behavior, and residents' water demand is divided into three levels and water use intervals are set.

[0010] Step 3: Combining historical water use data, real-time sensor information, and economic and social indicators, use deep learning algorithms to predict water demand at various levels, and calculate the threshold range of water resource carrying capacity based on water resource availability.

[0011] Step 4: Real-time monitoring and prediction of the changing trend of water resource population carrying capacity through deep learning algorithms, and calculation of population carrying limits under different scenarios, dynamically adjusting the threshold range of water resource population carrying capacity;

[0012] Step 5: Based on the dynamically adjusted water resource carrying capacity threshold range, formulate personalized water conservation policies, and monitor water resource use efficiency in real time through an intelligent management platform to optimize the implementation of water conservation policies;

[0013] Step 6: Introduce reinforcement learning algorithms to automatically adjust water resource allocation and usage strategies, and update the water resource carrying capacity threshold range in real time to optimize water resource management strategies.

[0014] Preferably, step one includes:

[0015] Deploy multiple types of high-precision intelligent sensors within the target area to form a fully covered Internet of Things system, and monitor water resources, climate change, economic and social indicators and residents' water use behavior in real time.

[0016] The smart sensor transmits the collected real-time data to the edge computing unit within the target area via wireless communication technology;

[0017] The edge computing unit performs preliminary analysis and processing on real-time data, filters out noisy data, and performs time-window weighted averaging to obtain multidimensional data.

[0018] A differentiated transmission algorithm is used to synchronously upload multidimensional data to the cloud platform through a high-speed network channel, and the cloud platform is used to store, analyze and mine the data.

[0019] Preferably, step two includes:

[0020] Combining time-series data-based clustering algorithms, and through big data analysis and clustering of residents' water use behavior, it is subdivided into eight aspects, including drinking water, toilet flushing water, personal hygiene water, cooking water, laundry water, household cleaning water and aquaculture water. Personal hygiene water is further subdivided into washing and bathing water.

[0021] Based on the characteristics of the target area, the water use situation of each aspect is calculated using the formula for calculating the additional water consumption, and the proportion of each aspect in the total water consumption of residents is analyzed.

[0022] Formulas for calculating additional water consumption for different water use behaviors:

[0023] (1) Calculation of drinking water consumption:

[0024]

[0025] In the formula, v drinking.r For each household resident's single water consumption, in liters (L / time), in grams (E) r1 The number of times each household resident drinks water per day, times / day, where m is the number of permanent residents in the household;

[0026] (2) Calculation of cooking water consumption:

[0027] W cook =v cook ·E2

[0028] In the formula, v cook E2 represents the amount of water used for a single household cooking session, while E2 represents the number of times a household cooks per day.

[0029] (3) Calculation of water consumption for toilet flushing:

[0030] W toilet =v toilet.j E3

[0031] In the formula, v toilet.j E3 represents the water consumption corresponding to the water efficiency level of the toilet, in L / (person·d), where j is 1 for level 1 water efficiency, j is 2 for level 2 water efficiency, and j is 3 for level 3 water efficiency. E3 represents the flushing frequency, in times / day.

[0032] (4) Calculation of water consumption for laundry:

[0033]

[0034] In the formula, v laundry The water consumption per cycle, expressed in L / cycle, is the water efficiency rating of the washing machine. 41 The number of times a household uses the washing machine per week, times / week, v hw Water consumption per hand wash, L / wash, E 42 The number of times a household washes its hands per week, expressed as times / week, where m represents the number of permanent residents in the household.

[0035] (5) Calculation of water consumption for washing and grooming:

[0036]

[0037] In the formula, v wash E5 represents the water consumption per wash, in L / wash; E5 represents the number of washes per day, in times / day.

[0038] (6) Calculation of water consumption for bathing:

[0039]

[0040] In the formula, W bath E6 represents the water usage per shower, in liters (L / shower), while E6 represents the number of showers per week, in times / week.

[0041] (7) Calculation of household water consumption for cleaning:

[0042]

[0043] In the formula, v e Water volume per unit area for mopping, in L / m² 2 S represents the residential housing area, in meters. 2 / person, E7 represents the number of times the floor is mopped per week, times / week;

[0044] (8) Calculation of water consumption for aquaculture:

[0045]

[0046] In the formula, v breed.r For each household's single-use aquaculture water consumption, in L / use, E r8 The number of times each household uses water for aquaculture per week, per week;

[0047] By using the probability density method to conduct frequency analysis on water consumption for each water use category, residential water demand is divided into three levels: rigid water demand, elastic water demand, and luxury water demand.

[0048] Probability density function:

[0049]

[0050] In the formula, f(x) is the probability density of a certain water use behavior, K is the kernel function, h is the bandwidth parameter, and x i For sample points;

[0051] The water usage distribution at each level was calculated using the probability density method, and a probability density distribution map was generated.

[0052] Water usage zones are defined for each level using probability density distribution maps, and adjustments are made as needed based on the characteristics of the target area.

[0053] Preferably, step three includes:

[0054] Based on historical water use data, an LSTM network is used to conduct a preliminary analysis of changes in water demand. A spatiotemporal relationship model is constructed by combining real-time sensor information for real-time monitoring of fluctuations in water demand.

[0055] By integrating economic and social indicators and fusing spatiotemporal information with nonlinear variation characteristics through deep learning algorithms, water demand at various levels can be predicted within a preset time period.

[0056] Nonlinear variation characteristics arise from complex climate patterns, water supply and demand relationships, population changes, economic activities, social behaviors, and policy factors.

[0057] Based on the established water use intervals for each level, historical water use data is analyzed and calculated to obtain the per capita water use for each water use interval.

[0058] A water resource supply and demand balance model is introduced to compare the available water resources with the predicted water demand at various levels and calculate the water resource population carrying capacity range.

[0059] Water resource supply and demand balance model:

[0060]

[0061] In the formula, R j W represents the available water volume of the j-th water source. loss It is the amount of water resources lost, D l This is the l-th type of water demand, where m and L are the quantities of water sources and water demand types, respectively.

[0062] Monte Carlo simulation was used to sample multiple times under different water supply and demand scenarios to predict the water carrying capacity under various scenarios and optimize the water supply and demand balance model.

[0063] Based on the water resource carrying capacity range and per capita water consumption, the carrying capacity limit, warning line, moderate line and surplus line are derived, and the rational allocation, scheduling and optimization schemes of water resources are provided for decision-makers.

[0064] Preferably, step four includes:

[0065] A water resource carrying capacity prediction model based on LSTM was constructed using multidimensional data, and cross-validation and hyperparameter optimization were employed to improve the model's prediction accuracy.

[0066] Based on the characteristics of the target area, a prediction model is used to conduct simulation analysis under different scenarios, and the corresponding water resource supply and demand balance is calculated to further estimate the population carrying capacity limit.

[0067] By calculating and predicting the water resources carrying capacity for the population, and combining the changing trends under various scenarios, the threshold range of the water resources carrying capacity for the population is dynamically adjusted.

[0068] By utilizing multi-source data fusion technology, data from different types of high-precision sensors are integrated, and water use efficiency is analyzed in real time and combined with regional characteristics to automatically generate customized water-saving suggestions.

[0069] Multiple optimization algorithms are introduced to optimally allocate water resources and population carrying capacity based on the current situation, and relevant optimization suggestions are automatically adjusted.

[0070] When it is predicted that the water resource carrying capacity is about to reach a critical value, or when the predicted changes in the scenario may lead to water shortage, an early warning mechanism will be automatically triggered and a policy adjustment will be requested.

[0071] Preferably, step five includes:

[0072] Based on the water resource carrying capacity of different regions and their dynamic adjustment thresholds, regional personalized water conservation policies are formulated using nonlinear programming or genetic algorithms.

[0073] The intelligent management platform comprehensively monitors the efficiency of water resource use. The platform integrates technologies such as IoT sensors, data acquisition modules, and cloud computing analysis to collect water usage data of residents in various regions in real time.

[0074] The LCA method is used to assess the long-term impact of water conservation policies on water resources, the economy and society, and to balance the short-term benefits and long-term sustainability of water conservation policies.

[0075] Through a real-time feedback and decision support mechanism, water-saving policies are automatically adjusted based on feedback data, and the effectiveness of policy implementation is monitored and adjusted.

[0076] By introducing a social behavior model, we can assess the social acceptance and implementation effectiveness of different water conservation policies and optimize the implementation methods of these policies.

[0077] Preferably, step six includes:

[0078] Based on reinforcement learning algorithms, a reward and punishment mechanism is constructed according to the current water conservation policies of the region to continuously optimize water resource allocation and use strategies;

[0079] Incentive Mechanism: A reward function is designed based on the effectiveness of water conservation policies to encourage the selection of reasonable and efficient water resource allocation strategies. For example:

[0080] Water conservation rewards: Rewards are given when a region or its residents reduce water waste.

[0081] Balanced Incentive: If water resources can be allocated fairly and reasonably to various regions and the water needs of all regions are guaranteed, an incentive will be given.

[0082] Sustainability incentives: Long-term incentives are given when water use strategies can be sustained for a considerable period of time without water shortages or extreme waste.

[0083] Penalty mechanisms: These penalize strategies that inefficiently or excessively consume water resources. For example:

[0084] Water shortage penalty: When a region experiences a water shortage (i.e., the water volume falls below a certain threshold), it will be penalized.

[0085] Waste penalties: Penalties will also be imposed if water resources are over-allocated or not allocated according to priority needs, resulting in water waste in certain areas.

[0086] Target optimization: Through rewards and penalties, continuously optimize water resource allocation strategies and adjust them towards goals such as water conservation, high efficiency, and sustainability.

[0087] Based on changes in water resource availability and population data, the DQN algorithm is used to dynamically map states and actions to optimize water conservation policies in the current region.

[0088] Q-value update formula:

[0089]

[0090] In the formula, s t As the current state, a t For the current action, a′ is the next state s. t+1 The action that brings the maximum Q value, r t+1 γ is the reward value, and γ is the discount factor;

[0091] In addition, the water resource carrying capacity threshold for each region is dynamically updated by taking into account climate change, population growth, and regional water demand.

[0092] Dynamic threshold adjustment: The water resource carrying capacity threshold for each region is dynamically updated based on factors such as climate change, population growth, and regional water demand. For example:

[0093] Population growth: If the population growth in a certain area accelerates, the carrying capacity threshold will be adjusted according to the water resource supply situation to ensure that water resources are not overloaded;

[0094] Climate change: In the event of drought or reduced precipitation, adjust carrying capacity thresholds, provide early warnings, and optimize water resource allocation.

[0095] Regional water demand changes: When regional water demand changes, it will trigger a real-time adjustment of the carrying capacity threshold to ensure the rational allocation of water resources for every resident in the region.

[0096] When a region experiences water shortage, a game theory-based water resource scheduling strategy can be used to achieve dynamic and optimized allocation of water resources across regions and mobilize surplus water resources from neighboring regions to supplement the water supply.

[0097] By combining water resources management expert systems, the effectiveness of water resources allocation can be evaluated in real time, and water resources management strategies can be adjusted within a preset time period through feedback mechanisms.

[0098] Preferably, the method also uses an adaptive neurofuzzy inference system to establish a dynamic adjustment model for the carrying capacity threshold, and automatically adjusts the water resource carrying capacity threshold by inputting water resource availability, climate change, population data and historical water use data.

[0099] Compared with the prior art, the beneficial effects of the present invention are:

[0100] This invention enables water resource management to flexibly respond to changes in supply and demand by dynamically adjusting thresholds, introducing intelligent algorithms and reinforcement learning, and optimizing water resource allocation and water-saving policies to ensure optimal utilization of water resources under various scenarios. It comprehensively considers regional characteristics and socio-economic factors, providing support for the formulation of personalized water-saving policies and long-term implementation strategies for different regions. Simultaneously, it optimizes policy implementation in real time through an intelligent management platform, enhancing the personalization and effectiveness of water-saving policies. Based on cutting-edge technologies such as deep learning, it scientifically predicts potential water shortage problems, triggers early warnings, and proposes solutions, providing effective guarantees for the long-term sustainable use of water resources. Attached Figure Description

[0101] Figure 1 This is a flowchart of the water resource population carrying capacity threshold calculation method based on hierarchical water use assessment of the present invention. Detailed Implementation

[0102] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0103] Example 1:

[0104] Please see Figure 1 As shown, the method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment includes the following steps:

[0105] Step 1: Deploy a sensor network in the target area to monitor water resources, climate change, socio-economic indicators, and residents' water use behavior in real time through edge computing units, and synchronize the collected multi-dimensional data to the cloud.

[0106] Step 2: Using big data analysis technology and probability density method, residential water use is subdivided into eight aspects according to residents' water use behavior, and residents' water demand is divided into three levels and water use intervals are set.

[0107] Step 3: Combining historical water use data, real-time sensor information, and economic and social indicators, use deep learning algorithms to predict water demand at various levels, and calculate the threshold range of water resource carrying capacity based on water resource availability.

[0108] Step 4: Real-time monitoring and prediction of the changing trend of water resource population carrying capacity through deep learning algorithms, and calculation of population carrying limits under different scenarios, dynamically adjusting the threshold range of water resource population carrying capacity;

[0109] Step 5: Based on the dynamically adjusted water resource carrying capacity threshold range, formulate personalized water conservation policies, and monitor water resource use efficiency in real time through an intelligent management platform to optimize the implementation of water conservation policies;

[0110] Step 6: Introduce reinforcement learning algorithms to automatically adjust water resource allocation and usage strategies, and update the water resource carrying capacity threshold range in real time to optimize water resource management strategies.

[0111] Deploy multiple types of high-precision intelligent sensors within the target area to form a fully covered Internet of Things system, and monitor water resources, climate change, economic and social indicators and residents' water use behavior in real time.

[0112] The smart sensor transmits the collected real-time data to the edge computing unit within the target area via wireless communication technology;

[0113] The edge computing unit performs preliminary analysis and processing on real-time data, filters out noisy data, and performs time-window weighted averaging to obtain multidimensional data.

[0114] A differentiated transmission algorithm is used to synchronously upload multidimensional data to the cloud platform through a high-speed network channel, and the cloud platform is used to store, analyze and mine the data.

[0115] Furthermore, by deploying multiple types of high-precision smart sensors within the target area, a comprehensive Internet of Things (IoT) system is formed, enabling real-time monitoring of water resources, climate change, socio-economic indicators, and residents' water usage behavior. Differentiated transmission algorithms and high-speed network channels ensure efficient data transmission and synchronous uploading to the cloud platform, thereby enabling the storage, analysis, and mining of large amounts of multi-dimensional data.

[0116] Combining time-series data-based clustering algorithms, and through big data analysis and clustering of residents' water use behavior, it is subdivided into eight aspects, including drinking water, toilet flushing water, personal hygiene water, cooking water, laundry water, household cleaning water, and aquaculture water.

[0117] Based on the characteristics of the target area, the water use situation of each aspect is calculated using the formula for calculating the additional water consumption, and the proportion of each aspect in the total water consumption of residents is analyzed.

[0118] By using the probability density method to conduct frequency analysis on water consumption for each water use category, residential water demand is divided into three levels: rigid water demand, elastic water demand, and luxury water demand.

[0119] The water usage distribution at each level was calculated using the probability density method, and a probability density distribution map was generated.

[0120] Water usage zones are defined for each level using probability density distribution maps, and adjustments are made as needed based on the characteristics of the target area.

[0121] Furthermore, by combining time-series data with clustering algorithms, big data analysis and clustering of residential water use behavior are conducted, subdividing complex water use patterns into eight categories to provide more accurate water demand analysis. Probability density methods are used to perform frequency analysis on each water use category, and by dividing water demand into three levels—rigid demand, elastic demand, and luxury demand—different water needs of residents are identified in greater detail, thus providing a scientific basis for water resource management and optimization. By constructing probability density distribution maps and setting reasonable water use intervals, water use strategies in target areas can be adjusted to achieve efficient allocation and conservation of water resources.

[0122] Based on historical water use data, an LSTM network is used to conduct a preliminary analysis of changes in water demand. A spatiotemporal relationship model is constructed by combining real-time sensor information for real-time monitoring of fluctuations in water demand.

[0123] By integrating economic and social indicators and fusing spatiotemporal information with nonlinear variation characteristics through deep learning algorithms, water demand at various levels can be predicted within a preset time period.

[0124] Based on the established water use intervals for each level, historical water use data is analyzed and calculated to obtain the per capita water use for each water use interval.

[0125] A water resource supply and demand balance model is introduced to compare the available water resources with the predicted water demand at various levels and calculate the water resource population carrying capacity range.

[0126] Monte Carlo simulation was used to sample multiple times under different water supply and demand scenarios to predict the water carrying capacity under various scenarios and optimize the water supply and demand balance model.

[0127] Based on the water resource carrying capacity range and per capita water consumption, the carrying capacity limit, warning line, moderate line and surplus line are derived, and the rational allocation, scheduling and optimization schemes of water resources are provided for decision-makers.

[0128] Furthermore, by combining LSTM networks to analyze historical water use data and utilizing real-time sensor information to construct a spatiotemporal relationship model, fluctuations in water demand can be monitored in real time, providing accurate dynamic water demand forecasts. By integrating economic and social indicators with deep learning algorithms to fuse spatiotemporal information and nonlinear variation characteristics, water demand at various levels can be effectively predicted. Introducing a water resource supply and demand balance model compares available water resources with predicted demand, accurately calculating the water resource carrying capacity range for the population, and providing a scientific basis for decision-makers.

[0129] A water resource carrying capacity prediction model based on LSTM was constructed using multidimensional data, and cross-validation and hyperparameter optimization were employed to improve the model's prediction accuracy.

[0130] Based on the characteristics of the target area, a prediction model is used to conduct simulation analysis under different scenarios, and the corresponding water resource supply and demand balance is calculated to further estimate the population carrying capacity limit.

[0131] By calculating and predicting the water resources carrying capacity for the population, and combining the changing trends under various scenarios, the threshold range of the water resources carrying capacity for the population is dynamically adjusted.

[0132] By utilizing multi-source data fusion technology, data from different types of high-precision sensors are combined to automatically generate customized water-saving suggestions through real-time analysis of water use efficiency and in combination with regional characteristics.

[0133] Multiple optimization algorithms are introduced to optimally allocate water resources and population carrying capacity based on the current situation, and relevant optimization suggestions are automatically adjusted.

[0134] When it is predicted that the water resource carrying capacity is about to reach a critical value, or when the predicted changes in the scenario may lead to water shortage, an early warning mechanism will be automatically triggered and a policy adjustment will be requested.

[0135] Furthermore, an LSTM-based water resource carrying capacity prediction model is used to accurately simulate water supply and demand balance under different scenarios and to estimate population carrying capacity limits. Multi-source data fusion technology is employed to analyze water use efficiency in real time and, combined with regional characteristics, to provide customized water-saving recommendations for different regions. Simultaneously, by dynamically adjusting the threshold range of water resource and population carrying capacity and applying various optimization algorithms, optimal allocation of water resources and population carrying capacity is achieved, enhancing decision-making flexibility and the ability to respond to emergencies.

[0136] Based on the water resource carrying capacity of different regions and their dynamic adjustment thresholds, nonlinear programming or genetic algorithms can be used to formulate regional personalized water conservation policies. For example, strict water quota management can be implemented in areas with relatively scarce water resources, while incentive measures, such as water fee discounts, can be adopted in areas with relatively abundant water resources.

[0137] The LCA method is used to assess the long-term impact of water conservation policies on water resources, the economy and society, and to balance the short-term benefits and long-term sustainability of water conservation policies.

[0138] Through a real-time feedback and decision support mechanism, water-saving policies are automatically adjusted based on feedback data, and the effectiveness of policy implementation is monitored and adjusted.

[0139] By introducing a social behavior model, we can assess the social acceptance and implementation effectiveness of different water conservation policies and optimize the implementation methods of these policies.

[0140] Furthermore, employing nonlinear programming or genetic algorithms to develop personalized water-saving policies ensures these policies are more precise and practical. Using life cycle analysis (LCA) to assess the long-term impacts of water-saving policies on water resources, the economy, and society helps find a balance between short-term benefits and long-term sustainability, avoiding excessive water conservation or resource waste. In addition, using social behavioral models to assess the social acceptance of water-saving policies further optimizes policy implementation, increases public participation and effectiveness, thereby promoting the long-term and effective implementation of water-saving measures.

[0141] Based on reinforcement learning algorithms, a reward and punishment mechanism is constructed according to the current water conservation policies of the region to continuously optimize water resource allocation and use strategies;

[0142] Based on changes in water resource availability and population data, the DQN algorithm is used to dynamically map states and actions to optimize water conservation policies in the current region.

[0143] In addition, the water resource carrying capacity threshold for each region is dynamically updated by taking into account climate change, population growth, and regional water demand.

[0144] When a region experiences water shortage, a game theory-based water resource scheduling strategy can be used to achieve dynamic and optimized allocation of water resources across regions and mobilize surplus water resources from neighboring regions to supplement the water supply.

[0145] By combining water resources management expert systems, the effectiveness of water resources allocation can be evaluated in real time, and water resources management strategies can be adjusted within a preset time period through feedback mechanisms.

[0146] Furthermore, by using the DQN algorithm for dynamic state-action mapping, policies can be adjusted in real time based on changes in water availability and population data, improving water conservation effectiveness. In addition, by combining climate change, population growth, and regional water demand, the water resource carrying capacity threshold is dynamically updated, making policies more adaptable and forward-looking. When a region experiences water shortages, game theory-based water resource allocation strategies can achieve optimal cross-regional water resource allocation, mobilizing surplus resources from neighboring regions to supplement water resources and avoid regional water crises.

[0147] This method also uses an adaptive neurofuzzy inference system to establish a dynamic adjustment model for the carrying capacity threshold. By inputting water resource availability, climate change, population data, and historical water use data, the method automatically adjusts the water resource population carrying capacity threshold.

[0148] Example 2:

[0149] Application Example: Threshold Calculation of Population Carrying Capacity for Domestic Water Use in Inland River Basins of Arid Regions

[0150] I. Background Introduction

[0151] Inland river basins in arid regions suffer from water scarcity and a prominent supply-demand imbalance. To rationally assess the population carrying capacity of water resources in these areas, this paper uses a hierarchical water use assessment method to refine the classification and calculation of residential water use, aiming to provide a scientific basis for water resource management and water conservation policy formulation.

[0152] II. Method Overview

[0153] (1) Data Acquisition and Preprocessing

[0154] Deploying a sensor network: High-precision smart sensors are deployed in urban areas of inland river basins in arid regions, covering multiple aspects such as water resources, climate, socio-economic indicators, and residents' water use behavior. The sensors transmit data in real time to edge computing units via wireless communication technology for preliminary processing and noise filtering, and then synchronize them to the cloud platform for storage and analysis via high-speed network channels.

[0155] Data Integration and Processing: Utilizing big data analytics, we integrate and process the collected multidimensional data, including historical water usage data, real-time sensor information, and socio-economic indicators. This ensures the accuracy, completeness, and timeliness of the data.

[0156] (2) Hierarchical water use assessment

[0157] Water use behavior classification: Residential water use is classified into eight categories according to water use behavior: drinking water, toilet flushing water, personal hygiene water (including washing and bathing), cooking water, laundry water, household cleaning water, and aquaculture water.

[0158] Water demand hierarchy classification:

[0159] Rigid water demand: This includes water for drinking, cooking, and other uses necessary for residents to maintain basic survival. It is not affected by economic and social conditions and is irreplaceable and incompressible.

[0160] Flexible water demand includes water for toilet flushing (partial), laundry, personal hygiene (water for washing and bathing beyond basic needs), and household cleaning. These water demands can vary with changes in external conditions and have moderate substitutability.

[0161] Extra water consumption includes water used for flushing toilets exceeding the prescribed standards, unreasonable personal hygiene water use, and additional water use caused by excessive cleaning. These water demands can be effectively reduced through technological improvements and increased water conservation awareness.

[0162] Water usage tiers are defined: Based on the probability density method, frequency analysis is performed on the water consumption of each water use behavior to define tiers for rigid, flexible, and luxury water use. Based on each defined tier, historical water usage data is analyzed and calculated to obtain the per capita water consumption for each tier. For example, toilet flushing water can be categorized into rigid, flexible, and luxury water use based on the toilet's water efficiency rating; washing and bathing water can be categorized based on daily usage frequency and temperature.

[0163] (3) Threshold calculation of water resources carrying capacity for population

[0164] Water demand forecasting: Combining historical water use data, real-time sensor information, and socio-economic indicators, deep learning algorithms are used to predict water demand at various levels. The non-linear characteristics of complex climate patterns, water supply and demand relationships, population changes, economic activities, social behaviors, and policy factors are taken into account.

[0165] Water supply and demand balance analysis: A water supply and demand balance model is introduced to compare the available water resources with the predicted water demand at various levels. Monte Carlo simulation is used to sample multiple times under different water supply and demand scenarios to predict the water carrying capacity under various scenarios.

[0166] Threshold range calculation: Based on the water resource carrying capacity range and per capita water consumption, the carrying capacity limit, warning line, moderate line, and surplus line are derived. These values ​​reflect the range of population that the region can support under different water resource supply and demand conditions.

[0167] Dynamic adjustment and optimization:

[0168] By using deep learning algorithms, the changing trends of water resource carrying capacity are monitored and predicted in real time. Based on different scenarios, the population carrying capacity limit is calculated, and the threshold range of water resource carrying capacity is dynamically adjusted.

[0169] By introducing reinforcement learning algorithms, water resource allocation and usage strategies are automatically adjusted, and the water resource carrying capacity threshold range is updated in real time to optimize water resource management strategies.

[0170] (4) Formulation and implementation of water conservation policies

[0171] Personalized water conservation policies: Based on dynamically adjusted water resource carrying capacity threshold ranges, regional personalized water conservation policies are formulated using nonlinear programming or genetic algorithms.

[0172] Intelligent management platform monitoring: The intelligent management platform monitors water resource utilization efficiency in real time, integrating technologies such as IoT sensors, data acquisition modules, and cloud computing analysis to collect water usage data from residents in various regions in real time.

[0173] Policy Effectiveness Evaluation and Optimization: The LCA (Limited Course Assessment) method is used to evaluate the long-term impacts of water conservation policies on water resources, the economy, and society, balancing the short-term benefits with long-term sustainability. Through real-time feedback and decision support mechanisms, water conservation policies are automatically adjusted, and the effectiveness of policy implementation is monitored and adjusted.

[0174] III. Detailed Calculation Process

[0175] Table 1 shows the definition of rigid, flexible, and luxury water use behaviors of residents in inland river basins in arid areas.

[0176] Table 1. Definitions of Rigid, Flexible, and Luxury Water Use Behaviors

[0177]

[0178]

[0179] Drinking water: It is used to sustain human survival, therefore it is entirely a rigid demand, with no elasticity or luxury demand;

[0180] Cooking water: Like drinking water, it is used to meet the physiological needs of human survival, therefore it is neither inelastic nor a luxury demand;

[0181] Toilet flushing water: According to the provisions of GB25502-2017 Water Efficiency Limits and Water Efficiency Grades for Toilets, Grades 1, 2 and 3 are classified into rigid, flexible and luxury water use according to the amount of water used.

[0182] Laundry water usage: Select the number of times you wash clothes per week;

[0183] Water for washing and rinsing: Water used for washing and rinsing twice or less per day is considered rigid water use, while water used more than twice per day is considered flexible water use.

[0184] Bath water: 25℃ is the starting temperature for heat sensation, therefore 25℃ is used as the dividing line;

[0185] Water used for environmental cleaning: The number of times the floor is mopped per week is used as the criterion.

[0186] Water for aquaculture: The number of flower pots is used as the basis for classification.

[0187] Water consumption data collection: Real-time water consumption data of various water use behaviors of urban residents are collected through sensor networks.

[0188] Water demand forecasting: Using deep learning algorithms such as LSTM networks, combined with historical data and real-time sensor information, water demand at various levels can be predicted for a period of time in the future.

[0189] Calculation of water resource carrying capacity threshold:

[0190] Based on the rigid, flexible, and luxury water use ranges defined in Table 1, let [Q] ai Q bi [Q] represents the rigid water usage range. bi Q ci [Q] represents the flexible water usage range. ci Q di [] represents the luxury water use range, where i = 1, 2, ..., 8, representing drinking water, toilet flushing water, personal hygiene water, cooking water, laundry water, household cleaning water, and aquaculture water, respectively.

[0191] Based on the characteristics of inland river basins in arid regions, the water consumption and rigid, flexible, and extravagant water use ranges for different water use behaviors are calculated. Based on the established water use ranges at each level, historical water use data is analyzed and calculated to obtain the per capita water consumption for each level of water use range. Based on the available water for urban residents, the water consumption ranges are divided by the rigid, flexible, and extravagant water use ranges for each water use behavior to obtain the water resource population carrying capacity range.

[0192] Based on the water resource carrying capacity range and per capita water consumption, the following are determined: carrying limit value (obtained by dividing by the minimum value of the rigid range), warning line (obtained by dividing by the per capita water consumption value of the rigid range), moderate line (obtained by dividing by the sum of the boundary value between the rigid and flexible ranges and the per capita water consumption value of the flexible range), and surplus line (obtained by dividing by the boundary value between the flexible and luxury ranges).

[0193] IV. Application Results

[0194] Using this method, we obtained the water resource carrying capacity threshold range for a town in an inland river basin in an arid region. According to the calculations, the town has an upper limit on the number of people it can support under current water resource conditions. When the population approaches or exceeds the carrying capacity limit, an early warning mechanism should be activated, and water-saving measures and optimized water resource allocation strategies should be implemented.

[0195] At the same time, we can formulate personalized water conservation policies based on the rigidity, flexibility, and extravagance of different water use behaviors. For example, for toilet flushing, we encourage residents to use water-saving toilets; for laundry, we advocate for reasonable control of the number of times and the amount of water used; and for personal hygiene, we advocate for water conservation and reducing waste.

[0196] Example 3:

[0197] This invention also provides a computer-readable storage medium storing a program for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment, as described above. When executed by a processor, this program implements the various processes of the above-described calculation method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0198] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0199] The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to general designs. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0200] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0201] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for calculating the population carrying capacity threshold of water resources based on hierarchical water use assessment, characterized in that, Includes the following steps: Step 1: Deploy a sensor network in the target area to monitor water resources, climate change, socio-economic indicators, and residents' water use behavior in real time through edge computing units, and synchronize the collected multi-dimensional data to the cloud. Step 2: Using big data analysis technology and probability density method, residential water use is subdivided into eight aspects according to residents' water use behavior, and residents' water demand is divided into three levels and water use intervals are set. Step 3: Combining historical water use data, real-time sensor information, and economic and social indicators, use deep learning algorithms to predict water demand at various levels, and calculate the threshold range of water resource carrying capacity based on water resource availability. Step 4: Real-time monitoring and prediction of the changing trend of water resource population carrying capacity through deep learning algorithms, and calculation of population carrying limits under different scenarios, dynamically adjusting the threshold range of water resource population carrying capacity; Step 5: Based on the dynamically adjusted water resource carrying capacity threshold range, formulate personalized water conservation policies, and monitor water resource use efficiency in real time through an intelligent management platform to optimize the implementation of water conservation policies; Step 6: Introduce reinforcement learning algorithms to automatically adjust water resource allocation and usage strategies, and update the water resource population carrying capacity threshold range in real time to optimize water resource management strategies. Step two includes: Combining time-series data-based clustering algorithms, and through big data analysis and clustering of residents' water use behavior, it is subdivided into eight aspects; Based on the characteristics of the target area, the water use situation of each aspect is calculated using the formula for calculating the additional water consumption, and the proportion of each aspect in the total water consumption of residents is analyzed. By using the probability density method to conduct frequency analysis on water consumption for each water use category, residential water demand is divided into three levels: rigid water demand, elastic water demand, and luxury water demand. Step three includes: Based on historical water use data, an LSTM network is used to conduct a preliminary analysis of changes in water demand. A spatiotemporal relationship model is constructed by combining real-time sensor information for real-time monitoring of fluctuations in water demand. By integrating economic and social indicators and fusing spatiotemporal information with nonlinear variation characteristics through deep learning algorithms, water demand at various levels can be predicted within a preset time period. Based on the established water use intervals for each level, historical water use data is analyzed and calculated to obtain the per capita water use for each water use interval. A water resource supply and demand balance model is introduced to compare the available water resources with the predicted water demand at various levels and calculate the water resource population carrying capacity range. Step four includes: A water resource carrying capacity prediction model based on LSTM was constructed using multidimensional data, and cross-validation and hyperparameter optimization were employed to improve the model's prediction accuracy. Based on the characteristics of the target area, a prediction model is used to conduct simulation analysis under different scenarios, and the corresponding water resource supply and demand balance is calculated to further estimate the population carrying capacity limit. By calculating and predicting the water resources carrying capacity for population, and combining the changing trends under various scenarios, the threshold range of the water resources carrying capacity for population is dynamically adjusted. By utilizing multi-source data fusion technology, data from different types of high-precision sensors are integrated, and water use efficiency is analyzed in real time and combined with regional characteristics to automatically generate customized water-saving suggestions. Multiple optimization algorithms are introduced to optimally allocate water resources and population carrying capacity based on the current situation, and relevant optimization suggestions are automatically adjusted. When it is predicted that the water resources carrying capacity is about to reach a critical value, or when the predicted changes in the scenario may lead to water shortage, an early warning mechanism will be automatically triggered and a policy adjustment will be prompted. Step five includes: Based on the water resource carrying capacity of different regions and their dynamic adjustment thresholds, regional personalized water conservation policies are formulated using nonlinear programming or genetic algorithms. The LCA method is used to assess the long-term impact of water conservation policies on water resources, the economy and society, and to balance the short-term benefits and long-term sustainability of water conservation policies. Through a real-time feedback and decision support mechanism, water-saving policies are automatically adjusted based on feedback data, and the effectiveness of policy implementation is monitored and adjusted. By introducing a social behavior model, we can assess the social acceptance and implementation effectiveness of different water conservation policies and optimize the implementation methods of these policies. Step six includes: Based on reinforcement learning algorithms, a reward and punishment mechanism is constructed according to the current water conservation policies of the region to continuously optimize water resource allocation and use strategies; Based on changes in water resource availability and population data, the DQN algorithm is used to dynamically map states and actions to optimize water conservation policies in the current region. In addition, the water resource carrying capacity threshold for each region is dynamically updated by taking into account climate change, population growth, and regional water demand. When a region experiences water shortage, a game theory-based water resource scheduling strategy can be used to achieve dynamic and optimized allocation of water resources across regions and mobilize surplus water resources from neighboring regions to supplement the water supply. By combining water resources management expert systems, the effectiveness of water resources allocation can be evaluated in real time, and water resources management strategies can be adjusted within a preset time period through feedback mechanisms.

2. The method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment according to claim 1, characterized in that, Step one includes: Deploy multiple types of high-precision intelligent sensors within the target area to form a fully covered Internet of Things system, and monitor water resources, climate change, economic and social indicators and residents' water use behavior in real time. The smart sensor transmits the collected real-time data to the edge computing unit within the target area via wireless communication technology; The edge computing unit performs preliminary analysis and processing on real-time data, filters out noisy data, and performs time-window weighted averaging to obtain multidimensional data. A differentiated transmission algorithm is used to synchronously upload multidimensional data to the cloud platform through a high-speed network channel, and the cloud platform is used to store, analyze and mine the data.

3. The method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment according to claim 2, characterized in that, The probability density function used in the probability density method is: In the formula, f(x) is the probability density of a certain water use behavior, K is the kernel function, h is the bandwidth parameter, and x i For sample points; The water usage distribution at each level was calculated using the probability density method, and a probability density distribution map was generated. Water usage zones are defined for each level using probability density distribution maps, and adjustments are made as needed based on the characteristics of the target area.

4. The method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment according to claim 3, characterized in that, Water resource supply and demand balance model: In the formula, R j W represents the available water volume of the j-th water source. loss It is the amount of water resources lost, D l This is the l-th type of water demand, where m and L are the quantities of water sources and water demand types, respectively. Monte Carlo simulation was used to sample multiple times under different water supply and demand scenarios to predict the water carrying capacity under various scenarios and optimize the water supply and demand balance model. Based on the water resource carrying capacity range and per capita water consumption, the carrying capacity limit, warning line, moderate line and surplus line are derived, and the rational allocation, scheduling and optimization schemes of water resources are provided for decision-makers.

5. The method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment according to claim 4, characterized in that, The Q-value update formula in the DQN algorithm: In the formula, s t As the current state, a t For the current action, a′ is the next state s. t+1 The action that brings the maximum Q value, r t+1 γ is the reward value, and γ is the discount factor.

6. The method for calculating the water resource population carrying capacity threshold based on hierarchical water use assessment according to claim 5, characterized in that: The method also uses an adaptive neurofuzzy inference system to establish a dynamic adjustment model for the carrying capacity threshold. By inputting water resource availability, climate change, population data, and historical water use data, the water resource carrying capacity threshold is automatically adjusted.

Images