A water resources dynamic quantity verification method and system based on big data analysis

Abstract

The present invention relates to the technical field of dynamic water resource quantity verification based on big data, and in particular to a method and system for dynamic water resource quantity verification based on big data analysis. The method comprises the following steps: collecting real-time water resource usage data through sensors and performing spatiotemporal pattern recognition to obtain spatiotemporal pattern data of water resources; calculating burst increments of water consumption based on the spatiotemporal pattern data of water resource usage to obtain burst increment data of water consumption; quantifying water resource losses based on burst increment data of node water flow to obtain physical fault water resource loss data; evaluating equipment performance loss on monitoring equipment fault data and performing statistical loss compensation on water consumption burst increment data to obtain statistical loss compensation data of water consumption; formulating a dynamic water resource quantity verification strategy based on statistical loss compensation data of water consumption to obtain a dynamic water resource quantity verification strategy. The present invention makes the verification technology more perfect by improving the verification technology.

Description

Technical Field

[0001] The present invention relates to the technical field of water resource dynamic quantity verification based on big data, and in particular to a water resource dynamic quantity verification method and system based on big data analysis. Background Art

[0002] Through systematic processing and analysis of large-scale, multi-dimensional data, the status and changing trends of water resources can be accurately and real-time evaluated and predicted. In modern society, the management and allocation of water resources face unprecedented challenges, especially in the context of global climate change, population growth and rapid urbanization, the shortage and over-utilization of water resources are becoming increasingly serious. In urban water supply systems, real-time monitoring and data analysis can be used to optimize the operating efficiency of the water supply network and reduce water leakage and energy consumption. However, a traditional water resource dynamic nuclear method based on big data analysis has the problems of inaccurate analysis of urban spatiotemporal water supply distribution and large errors in water supply settlement under abnormal conditions. Summary of the invention

[0003] Based on this, it is necessary to provide a water resources dynamic quantity verification method and system based on big data analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a water resources dynamic quantity method based on big data analysis is provided, the method comprising the following steps:

[0005] Step S1: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data; performing spatiotemporal pattern recognition on the real-time water resource usage data to obtain spatiotemporal pattern data of water resources;

[0006] Step S2: Perform water consumption anomaly detection based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data;

[0007] Step S3: Identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; classify the abnormal node pipeline state data by abnormal mode to obtain pipeline abnormal mode classification data, wherein the pipeline abnormal mode classification data includes physical fault mode data and monitoring equipment fault data; quantify the water resource loss according to the physical fault mode data to obtain physical fault water resource loss data;

[0008] Step S4: Evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment monitoring accuracy loss data to obtain water consumption statistical loss compensation data;

[0009] Step S5: Formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

[0010] The present invention collects water resource usage data in real time through sensors and other equipment, including water consumption in different time periods and different regions, and uses advanced data analysis techniques, such as machine learning algorithms, to analyze and pattern recognize real-time water resource usage data. This step can help identify water use patterns in different time and space dimensions, such as daily water use patterns, holiday water use patterns, and seasonal changes. Based on the identified spatiotemporal pattern data, the system can automatically detect abnormal water use conditions. These anomalies are manifested as sudden increases or decreases, which are inconsistent with the expected water use pattern. The water consumption data between different regions are analyzed to find the spatial correlation between them. This can help determine whether there are concentrated abnormal areas, or whether the anomalies in certain areas affect adjacent areas. Based on the spatial correlation data, the system can calculate the sudden increase in water consumption, that is, the magnitude of the sudden and significant increase in water consumption in a certain area. This calculation can help predict potential water use emergencies or areas that need special attention. Through real-time data collection and analysis, the system can quickly respond to abnormal water use, help water resource managers adjust water resource allocation and control measures in a timely manner, accurately identify water use anomalies and sudden increases, and help optimize water resource utilization efficiency, avoid resource waste and imbalance between supply and demand. Especially in the case of natural disasters or man-made disasters, early warnings can be issued through data analysis and corresponding disaster relief and protection measures can be taken to ensure water resource security. By analyzing the sudden increase data of node water flow, the system can identify pipeline nodes with abnormalities. These anomalies indicate physical faults or equipment problems inside the pipeline. The pipeline status data of the identified abnormal nodes can help indicate which pipeline parts need further inspection and maintenance. The pipeline status of the identified abnormal nodes is further classified into physical failure mode data. This includes actual physical problems such as pipeline rupture and water leakage. On the other hand, the pipeline status of the abnormal nodes is also classified as monitoring equipment failure data, which indicates that the monitoring equipment has a failure, resulting in inaccurate or incomplete data collection. In-depth analysis of the monitoring equipment failure data is carried out to evaluate the actual degree of loss of equipment performance. This can help determine the impact of equipment failure on water resource management and its impact on data accuracy and reliability. Based on the equipment performance loss assessment data, the system can calculate the statistical loss of water consumption caused by the loss of equipment monitoring accuracy. These loss data are calculated based on the difference between actual water consumption and monitoring data. Based on the physical failure mode data and water consumption statistical loss compensation data, the system formulates a dynamic water resource verification strategy. This strategy includes how to deal with water resource losses caused by physical failures and how to compensate for statistical losses caused by performance losses of monitoring equipment. The strategy involves measures such as repairing physical failures in pipelines, replacing or repairing monitoring equipment, and adjusting data collection frequency or accuracy to ensure the accuracy and efficiency of water resource management. The formulated dynamic water resource verification strategy will be sent to the cloud platform for execution.This can be an automated process, with real-time verification and adjustment achieved through cloud computing and remote monitoring systems. The system can promptly identify anomalies in pipelines and monitoring equipment, which helps to respond and repair quickly, reduce water resource losses and management costs. Based on detailed anomaly classification and evaluation data, more accurate pipeline maintenance and equipment replacement strategies can be formulated to improve equipment utilization and service life. By accurately capturing and compensating for water consumption statistical losses, the system can effectively reduce resource waste caused by equipment failures and data inaccuracies. Integrating these steps can make water resource management more intelligent and efficient, and help achieve sustainable water resource utilization and management goals. Therefore, the present invention provides a water resource dynamic verification method based on big data analysis, which is an improved treatment of a traditional water resource dynamic verification method based on big data analysis, and solves the problem that a traditional water resource dynamic verification method based on big data analysis has inaccurate analysis of urban spatiotemporal water supply distribution and large errors in water supply settlement under abnormal conditions, improves the accuracy of urban spatiotemporal water supply distribution analysis and reduces the problem of large errors in water supply settlement under abnormal conditions.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data;

[0013] Step S12: Cleaning the real-time water resource usage data to obtain cleaned real-time water resource usage data;

[0014] Step S13: normalizing the real-time water resource usage cleaning data to obtain normalized real-time water resource usage data;

[0015] Step S14: performing spatiotemporal pattern recognition on the real-time water resource usage data according to the normalized real-time water resource usage data to obtain spatiotemporal pattern data of water resource usage.

[0016] The present invention obtains water resource usage data in real time through sensors, including specific water consumption in various time periods and different locations. Ensure the accuracy and timeliness of the data, help managers to understand the actual use of water resources in a timely manner, and support real-time decision-making and response. Cleaning data can remove outliers, erroneous data or noise in the data to ensure that subsequent analysis and decision-making are based on high-quality data. The cleaned data is more reliable, reduces the analysis deviation caused by data quality problems, and improves the accuracy of management decisions. Through normalization processing, the real-time water resource usage data of different sensors and different data sources can be unified into a standard data format or numerical range. Normalized data is easier to compare and analyze, and can more effectively reveal the potential patterns and trends of water consumption. Spatiotemporal pattern recognition based on normalized data can reveal the usage patterns of water resources in different time and space dimensions, such as daily fluctuations, seasonal changes, etc. The identified patterns help predict future water use trends, support reasonable planning and allocation of water resources, and avoid imbalance between supply and demand or excessive consumption.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: Obtaining historical water consumption data of the region;

[0019] Step S22: performing water demand fluctuation analysis on the historical water consumption data of the region to obtain historical water consumption fluctuation data;

[0020] Step S23: Calculate seasonal demand differences for regional historical water consumption data based on historical water consumption fluctuation data to obtain seasonal demand difference data for water consumption;

[0021] Step S24: performing water consumption anomaly detection on the seasonal demand difference data of water resources consumption according to the spatiotemporal pattern data of water resources consumption to obtain regional water consumption anomaly data;

[0022] Step S25: performing spatial correlation analysis on the regional water consumption abnormal data according to the spatiotemporal pattern data of water resource consumption to obtain the regional water consumption spatial correlation data;

[0023] Step S26: Calculate the water consumption burst increment of the regional water consumption abnormal data according to the regional water consumption spatial correlation data to obtain the water consumption burst increment data.

[0024] The present invention obtains long-term historical water use data of a region, which is helpful to establish a comprehensive water use behavior model and trend analysis basis. Historical data provides a benchmark that can be used to compare and evaluate changes and anomalies in current water use. Analyzing historical water use data reveals the fluctuation pattern of water use in different time periods, such as daily fluctuations, weekly changes or seasonal fluctuations. By analyzing historical fluctuation data, future water use trends can be predicted to help planning and resource allocation. Based on historical water use fluctuation data, the difference in water demand between different seasons is calculated. This helps to understand and predict the degree of change in water resource demand in different seasons. Identifying seasonal demand differences can help water resource managers formulate optimized water supply plans to meet demand fluctuations in different seasons. Based on the spatiotemporal pattern data of water resource consumption, abnormal water use in the region, such as sudden increase or decrease in water use, is detected. Timely detection and response to abnormal water use can avoid water resource waste or imbalance between supply and demand and improve water resource utilization efficiency. Analyze the spatial correlation of water use between different regions to understand whether the changes in water use between regions are related. Assess whether the changes in water use between different regions affect each other, which helps identify water resource management strategies or adjustment measures. Based on the spatial correlation data of regional water consumption, the sudden increase in water consumption can be calculated. This helps to predict emergency water use or emergencies. Calculating and preparing for sudden increases in water consumption in advance can help water resource managers formulate and implement effective emergency response measures to ensure the safety and sustainability of water supply.

[0025] Preferably, step S25 comprises the following steps:

[0026] Step S251: performing a spatiotemporal distribution analysis of water consumption abnormality hotspots on the regional water consumption abnormality data according to the spatiotemporal pattern data of water resource consumption, and obtaining the spatiotemporal distribution data of water consumption abnormality hotspots;

[0027] Step S252: performing abnormal spatial domain effect analysis on the water consumption abnormal hotspot spatiotemporal distribution data to obtain water consumption abnormal spatial domain effect data;

[0028] Step S253: performing hotspot abnormal trend identification on the water consumption abnormal hotspot spatiotemporal distribution data according to the water consumption abnormal spatial domain effect data to obtain abnormal hotspot trend identification data;

[0029] Step S254: Perform spatial correlation analysis based on the water consumption abnormal spatial domain effect data and abnormal hot spot trend identification data to obtain regional water consumption spatial correlation data.

[0030] By analyzing the spatiotemporal pattern data of water resource consumption, the present invention can identify abnormal water consumption hotspots in the region, that is, those areas where water consumption is significantly higher or lower than normal levels. Understanding the distribution of these abnormal hotspots in time and space helps to determine whether there are concentrated or dispersed abnormal phenomena, and then take corresponding management measures. Analyzing the spatiotemporal distribution data of abnormal hotspots can reveal the mutual influence and spatial correlation of water use anomalies between different regions. This helps to understand the diffusion or aggregation trend of water use anomalies between different regions. Identify the spatial domain effects that cause anomalies, such as specific geographical conditions, population density changes, or infrastructure differences, to provide a basis for further management and prediction. Based on the data of abnormal spatial domain effects of water consumption, identify the trend changes of abnormal hotspots, such as abnormal persistence, growth or decline trends. Timely identification of the trend changes of abnormal hotspots helps to take measures in advance to adjust the water supply plan or implement emergency response to cope with water consumption fluctuations. Combined with the abnormal hotspot trend identification data, the spatial correlation of water consumption between different regions is further analyzed. This helps to understand the degree of correlation between water consumption changes between different regions and provide a basis for cross-regional resource allocation and emergency response. Spatial correlation analysis provides decision support for water resource managers, helping them to develop more effective resource allocation strategies and response measures to cope with complex water use changes and challenges among regions.

[0031] Preferably, step S26 comprises the following steps:

[0032] Step S261: Perform spatial heterogeneity assessment on the regional water consumption spatial correlation data to obtain regional water consumption spatial heterogeneity data;

[0033] Step S262: performing water consumption abnormality spatial regression analysis on the regional water consumption abnormality data according to the regional water consumption spatial heterogeneity data to obtain water consumption abnormality spatial regression data;

[0034] Step S263: performing cluster analysis on the water consumption abnormal spatial regression data to obtain water consumption abnormal regression cluster data;

[0035] Step S264: Calculate the water consumption burst increment for the regional water consumption abnormal data according to the water consumption abnormal spatial regression data and the water consumption abnormal regression clustering data to obtain the water consumption burst increment data.

[0036] The present invention reveals the spatial heterogeneity of water consumption between different regions by evaluating the spatial correlation data of regional water consumption. This includes the differences in water consumption caused by factors such as geography, population distribution, and economic development between regions. The obtained regional water consumption spatial heterogeneity data helps to establish the basis for subsequent analysis and help identify water use behavior patterns with significant differences in space. Using the regional water consumption spatial heterogeneity data, spatial regression analysis is performed on water consumption anomalies in the region. This analysis can identify the degree of influence of spatial factors on water use anomalies and help understand the causes and driving forces of anomalies in different regions. Through regression analysis, future water use anomalies can be predicted and measures can be taken in advance to manage and adjust them. Based on the water consumption anomaly spatial regression data, cluster analysis is performed to identify similar water consumption anomaly patterns or behavior types. This analysis helps to classify water consumption anomaly data in the region and understand the common characteristics and differences between different categories. Corresponding management strategies and countermeasures are formulated for each category to more accurately solve various water consumption anomaly problems. Based on the water consumption anomaly spatial regression data and regression clustering data, the sudden increase in water consumption is calculated. This helps to quantify and evaluate the increase in water use in emergencies or abnormal situations, and helps plan and manage resource allocation in response to emergencies. It provides forecasts and assessments of sudden water use increases, enabling managers to prepare and respond to emergencies more effectively and ensure the stable supply and management security of water resources.

[0037] Preferably, step S3 comprises the following steps:

[0038] Step S31: marking the regional nodes of the burst increment data of water consumption to obtain the burst increment node marking data;

[0039] Step S32: Calculate the burst increment of water flow between different nodes based on the burst increment node mark data according to the burst increment data of water consumption, and obtain the burst increment data of node water flow;

[0040] Step S33: Identify the abnormal node pipeline state according to the node water flow burst increment data and the burst increment node mark data to obtain the abnormal node pipeline state data;

[0041] Step S34: using the node abnormal pattern classification model to perform abnormal pattern classification on the abnormal node pipeline state data to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data;

[0042] Step S35: locating the physical fault node according to the physical fault mode data to obtain physical fault node locating data; evaluating the fault impact range of the physical fault node locating data to obtain node fault impact range evaluation data;

[0043] Step S36: quantify the water resource loss according to the node fault impact range assessment data and the node water flow burst increment data to obtain the physical fault water resource loss data.

[0044] The present invention determines the regional nodes with abnormalities based on the burst increment data of water consumption and marks them. These node marks provide the objects to focus on in subsequent analysis, which helps to concentrate resources and attention to deal with potential problems. Based on the burst increment node mark data, the burst increment of water flow between different nodes is calculated. This step reveals the change of water flow under abnormal conditions and provides basic data for subsequent abnormal identification and analysis. The state of the abnormal node pipeline is identified using the node water flow burst increment data. This includes detecting and confirming whether the node has abnormal water flow behavior, such as flow or pressure changes beyond the normal range. The abnormal node pipeline state data is classified through the abnormal pattern classification model. These classifications can cover physical failure modes and monitoring equipment failure modes, which help to distinguish different types of abnormal events and formulate corresponding countermeasures. According to the physical failure mode data, the specific node that causes the abnormality is accurately located. This helps to respond quickly and accurately repair to minimize service interruption and resource waste. Evaluate the impact range of the faulty node on the surrounding area. These data provide an understanding of the affected area and provide a basis and priority for planning repair work. Combined with the node failure impact range assessment data and the node water flow burst increment data, the water resource loss caused by the physical failure is quantified. These data provide important references for cost-benefit analysis and resource allocation, helping to optimize management and maintenance strategies.

[0045] Preferably, the step of constructing the node abnormal pattern classification model includes the following steps:

[0046] Obtain pipeline structure design data and sensor structure design data;

[0047] Carry out water flow scour limit bearing calculation on pipeline structure design data to obtain pipeline structure limit bearing data;

[0048] Calculate the stress limit of the pipeline structure design data according to the pipeline structure limit bearing data to obtain the pipeline structure stress bearing limit data;

[0049] Perform humidity adaptability simulation on sensor structure design data to obtain sensor humidity adaptability simulation data;

[0050] According to the sensor humidity adaptability simulation data, the sensor structure design data is subjected to sensitivity condition impairment analysis to obtain the sensor sensitivity condition impairment data;

[0051] Based on the PolicyGradient algorithm, feature learning is performed on the pipeline structure stress bearing limit data and sensor sensitivity condition reduction data to obtain abnormal mode characteristic learning data;

[0052] Based on the convolutional neural network, an abnormal pattern analysis model is constructed for the abnormal pattern characteristic learning data to obtain a node abnormal pattern classification model.

[0053] The present invention obtains the structural design data of the pipeline, including information such as the geometric shape and material properties of the pipeline, and the structural design data of the sensor, such as the location, type, and technical specifications of the sensor, and determines the ultimate bearing capacity of the pipeline under different water flow conditions by calculating the scouring effect of the water flow on the pipeline. These data reflect the mechanical stress faced by the pipeline structure in actual operation. Based on the ultimate bearing data of the pipeline structure, the ultimate capacity of the pipeline structure in terms of stress is calculated. These data take into account the mechanical properties and stress environment of the pipeline material and evaluate the safety and reliability of the pipeline. By simulating the response and performance of the sensor under different humidity conditions, the data of the sensor under different environmental conditions are obtained. The performance changes and sensitivity loss of the sensor under different environmental conditions are analyzed, and these analyses help understand the performance of the sensor under abnormal environments. Using the PolicyGradient algorithm, effective feature representations are learned from the stress bearing limit data of the pipeline structure and the sensitivity condition loss data of the sensor. These features can capture the key attributes of the pipeline and the sensor for the construction of subsequent models. Using the learned abnormal mode characteristic data, a convolutional neural network model is constructed. This model can identify and classify abnormal modes of pipelines and sensors, including problems caused by physical failures and environmental changes.

[0054] Preferably, step S4 comprises the following steps:

[0055] Step S41: extracting the equipment fault status code from the monitoring equipment fault data to obtain the equipment fault status code;

[0056] Step S42: performing equipment performance loss assessment on the monitoring equipment failure data according to the equipment failure status code to obtain equipment performance loss assessment data;

[0057] Step S43: Calculate the equipment monitoring precision loss according to the equipment performance loss evaluation data and the monitoring equipment failure data to obtain the equipment monitoring precision loss data;

[0058] Step S44: Compensate the water consumption statistics loss for the sudden increase in water consumption data according to the equipment monitoring accuracy loss data to obtain water consumption statistics loss compensation data.

[0059] The present invention extracts the equipment fault status code from the monitoring equipment fault data, which helps to standardize the fault information so that it can be effectively identified and processed by the system. Through the status code, the specific equipment fault type can be quickly located, which helps to quickly respond and repair, and reduce system downtime and production losses. According to the equipment fault status code, the performance loss of the monitoring equipment fault data is evaluated, and the performance impact of the equipment during the fault period or under abnormal operation can be quantified. By evaluating the equipment performance loss, resource allocation and maintenance strategies can be optimized to ensure efficient use of maintenance resources and reduce maintenance costs. Based on the equipment performance loss evaluation data, the equipment monitoring precision loss calculation is performed, which helps to evaluate the accuracy and reliability of the monitoring system. By accurately calculating the monitoring precision loss, the accuracy of the monitoring system for the actual situation can be improved, thereby improving the effect of data analysis and prediction. Using the equipment monitoring precision loss data, the water consumption burst increment data is statistically compensated for the loss, which can correct the problem of incomplete data caused by monitoring errors. Ensure the accuracy and completeness of water consumption data, and provide a reliable basis for subsequent data analysis, resource planning and decision making.

[0060] Preferably, step S44 includes the following steps:

[0061] Step S441: performing precision loss quantification processing on the equipment monitoring precision loss data to obtain precision loss quantification data;

[0062] Step S442: performing precision loss gradient sorting on the equipment monitoring precision loss data according to the precision loss quantification data to obtain precision loss gradient sorting data;

[0063] Step S443: using Lagrange multipliers to perform precision loss gradient constraints on the precision loss gradient sorting data to obtain precision loss gradient constrained data;

[0064] Step S444: performing quasi-Newton condition constraint calculation on the precision loss gradient sorting data according to the precision loss gradient constraint data to obtain device precision loss gradient approximation data;

[0065] Step S445: Perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment accuracy loss gradient approximation data to obtain water consumption statistical loss compensation data.

[0066] The present invention processes the equipment monitoring precision loss data to convert unstructured or complex data into a quantifiable form, which is convenient for subsequent data analysis and processing. The precision loss data of different devices or different time periods can be compared and analyzed, which is helpful to identify and optimize the problem points in the monitoring system. According to the precision loss quantification data, the precision loss of the monitoring equipment is sorted, and it can be clear which equipment or which data sets have the most serious precision impact. It helps managers to allocate resources according to the severity of the loss, and give priority to the equipment or data with the greatest impact, so as to improve the efficiency and reliability of the overall monitoring system. By using the Lagrange multiplier method to constrain the precision loss gradient, the distribution of precision loss can be effectively controlled and optimized to avoid excessive fluctuations and abnormal situations. Ensuring that the precision loss data fluctuates within an acceptable range helps to maintain the stability and long-term reliability of the monitoring system. According to the constraints, the quasi-Newton method is used to calculate the precision loss gradient, and the actual situation of the equipment precision loss is further optimized and approximated. While improving the equipment monitoring accuracy, the data inaccuracy caused by monitoring errors is reduced, providing a more reliable basis for subsequent data analysis and decision-making. Based on the equipment accuracy loss gradient approximation data, statistical loss compensation is performed on the sudden increase in water consumption data to correct the incomplete or inaccurate water consumption data caused by monitoring errors. The integrity and accuracy of water consumption data are ensured to provide reliable support for subsequent resource planning, decision-making and data analysis.

[0067] Preferably, the present invention further provides a water resources dynamic quantity verification system based on big data analysis, which is used to execute the water resources dynamic quantity verification method based on big data analysis as described above, and the water resources dynamic quantity verification system based on big data analysis includes:

[0068] The spatiotemporal pattern recognition module is used to collect real-time water resource usage data through sensors to obtain real-time water resource usage data; perform spatiotemporal pattern recognition on the real-time water resource usage data to obtain water resource spatiotemporal pattern data;

[0069] The incremental analysis module is used to detect water consumption anomalies based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data;

[0070] The fault classification module is used to identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; to classify the abnormal node pipeline state data into abnormal patterns to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data; to quantify water resource loss according to the physical fault pattern data to obtain physical fault water resource loss data;

[0071] The equipment accuracy loss compensation module is used to evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; to compensate for the water consumption statistical loss of the sudden increase data of water consumption according to the equipment monitoring accuracy loss data to obtain the water consumption statistical loss compensation data;

[0072] The strategy formulation module is used to formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

[0073] The beneficial effect of the present invention is that water resource usage data, including water consumption in different time periods and different regions, is collected in real time through sensors and other equipment, and advanced data analysis techniques, such as machine learning algorithms, are used to analyze and pattern recognize real-time water resource usage data. This step can help identify water use patterns in different time and space dimensions, such as daily water use patterns, holiday water use patterns, and seasonal changes. Based on the identified spatiotemporal pattern data, the system can automatically detect abnormal water use conditions. These anomalies are manifested as sudden increases or decreases that are inconsistent with the expected water use pattern. The water consumption data between different regions are analyzed to find the spatial correlation between them. This can help determine whether there are concentrated abnormal areas, or whether the anomalies in certain areas affect adjacent areas. Based on the spatial correlation data, the system can calculate the sudden increase in water consumption, that is, the magnitude of a sudden and significant increase in water consumption in a certain area. This calculation can help predict potential water use emergencies or areas that need special attention. Through real-time data collection and analysis, the system can quickly respond to abnormal water use, help water resource managers adjust water resource allocation and control measures in a timely manner, accurately identify water use anomalies and sudden increases, and help optimize water resource utilization efficiency, avoid resource waste and imbalance between supply and demand. Especially in the case of natural disasters or man-made disasters, early warnings can be issued through data analysis and corresponding disaster relief and protection measures can be taken to ensure water resource security. By analyzing the sudden increase data of node water flow, the system can identify pipeline nodes with abnormalities. These anomalies indicate physical faults or equipment problems inside the pipeline. The pipeline status data of the identified abnormal nodes can help indicate which pipeline parts need further inspection and maintenance. The pipeline status of the identified abnormal nodes is further classified into physical failure mode data. This includes actual physical problems such as pipeline rupture and water leakage. On the other hand, the pipeline status of the abnormal nodes is also classified as monitoring equipment failure data, which indicates that the monitoring equipment has a failure, resulting in inaccurate or incomplete data collection. In-depth analysis of the monitoring equipment failure data is carried out to evaluate the actual degree of loss of equipment performance. This can help determine the impact of equipment failure on water resource management and its impact on data accuracy and reliability. Based on the equipment performance loss assessment data, the system can calculate the statistical loss of water consumption caused by the loss of equipment monitoring accuracy. These loss data are calculated based on the difference between actual water consumption and monitoring data. Based on the physical failure mode data and water consumption statistical loss compensation data, the system formulates a dynamic water resource verification strategy. This strategy includes how to deal with water resource losses caused by physical failures and how to compensate for statistical losses caused by performance losses of monitoring equipment. The strategy involves measures such as repairing physical failures in pipelines, replacing or repairing monitoring equipment, and adjusting data collection frequency or accuracy to ensure the accuracy and efficiency of water resource management. The formulated dynamic water resource verification strategy will be sent to the cloud platform for execution.This can be an automated process, with real-time verification and adjustment achieved through cloud computing and remote monitoring systems. The system can promptly identify anomalies in pipelines and monitoring equipment, which helps to respond and repair quickly, reduce water resource losses and management costs. Based on detailed anomaly classification and evaluation data, more accurate pipeline maintenance and equipment replacement strategies can be formulated to improve equipment utilization and service life. By accurately capturing and compensating for water consumption statistical losses, the system can effectively reduce resource waste caused by equipment failures and data inaccuracies. Integrating these steps can make water resource management more intelligent and efficient, and help achieve sustainable water resource utilization and management goals. Therefore, the present invention provides a water resource dynamic verification method based on big data analysis, which is an improved treatment of a traditional water resource dynamic verification method based on big data analysis, and solves the problem that a traditional water resource dynamic verification method based on big data analysis has inaccurate analysis of urban spatiotemporal water supply distribution and large errors in water supply settlement under abnormal conditions, improves the accuracy of urban spatiotemporal water supply distribution analysis and reduces the problem of large errors in water supply settlement under abnormal conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic diagram of the steps of a water resources dynamic quantity method based on big data analysis;

[0075] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0076] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0077] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0078] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0079] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0080] To achieve this, please refer to Figure 1 to Figure 2 A water resources dynamic quantity method based on big data analysis, the method comprising the following steps:

[0081] Step S1: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data; performing spatiotemporal pattern recognition on the real-time water resource usage data to obtain spatiotemporal pattern data of water resources;

[0082] Step S2: Perform water consumption anomaly detection based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data;

[0083] Step S3: Identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; classify the abnormal node pipeline state data by abnormal mode to obtain pipeline abnormal mode classification data, wherein the pipeline abnormal mode classification data includes physical fault mode data and monitoring equipment fault data; quantify the water resource loss according to the physical fault mode data to obtain physical fault water resource loss data;

[0084] Step S4: Evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment monitoring accuracy loss data to obtain water consumption statistical loss compensation data;

[0085] Step S5: Formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

[0086] In the embodiment of the present invention, reference Figure 1The above is a schematic diagram of the steps of a water resources dynamic quantity method based on big data analysis of the present invention. In this example, the water resources dynamic quantity method based on big data analysis includes the following steps:

[0087] Step S1: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data; performing spatiotemporal pattern recognition on the real-time water resource usage data to obtain spatiotemporal pattern data of water resources;

[0088] In the embodiment of the present invention, suitable sensor devices, such as flow meters, pressure sensors, etc., are deployed at water resource use nodes or key locations to ensure that the sensor devices can accurately collect water resource use data in real time, including relevant parameters such as water flow, pressure, and temperature, and set up data transmission channels to transmit the real-time data collected by the sensors to a data center or cloud platform through a secure and reliable network. Ensure the security and stability of the data transmission process to ensure data integrity and real-time performance. Set up a storage system on the data center or cloud platform to store real-time water resource usage data for subsequent analysis and processing. Use a spatiotemporal pattern recognition algorithm to analyze and process the collected real-time water resource usage data. The algorithm includes but is not limited to a machine learning algorithm, a deep learning algorithm, etc., which is used to identify water resource usage patterns in different time and space. Ensure that the algorithm can effectively process large-scale data and can automatically identify and classify different spatiotemporal patterns. Generate spatiotemporal pattern data of water resources based on the results of algorithm processing. Including detailed description and classification of water resource usage patterns in different time periods, different locations, or different usage scenarios. Ensure that the generated spatiotemporal pattern data is visual and interpretable, which is convenient for subsequent analysis and decision support.

[0089] Step S2: Perform water consumption anomaly detection based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data;

[0090] In an embodiment of the present invention, a predefined algorithm or model is used to perform anomaly detection on the spatiotemporal pattern data of water resource consumption in each region. Common methods include statistical methods (such as mean, standard deviation), machine learning methods (such as isolated forest, abnormal factor detection), etc. Determine the threshold or standard for anomaly detection in order to identify those abnormal situations that exceed the normal water use pattern. The detected abnormal data is recorded and analyzed in detail. Including key information such as the time, location, and water consumption changes of the abnormality. Use spatial analysis methods (such as spatial autocorrelation, Kriging interpolation, etc.) to analyze the spatial correlation of water consumption between different locations in the region. Determine the propagation or correlation pattern of abnormal water consumption events in space. Based on the spatial correlation data, further calculate the abnormal water consumption data detected in the region. Calculate the sudden increase in water consumption, that is, the significant increase in abnormal events compared to the average. Ensure that the calculation process is accurate and in accordance with statistical principles for subsequent decision analysis and response measures.

[0091] Step S3: Identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; classify the abnormal node pipeline state data by abnormal mode to obtain pipeline abnormal mode classification data, wherein the pipeline abnormal mode classification data includes physical fault mode data and monitoring equipment fault data; quantify the water resource loss according to the physical fault mode data to obtain physical fault water resource loss data;

[0092] In an embodiment of the present invention, data related to pipeline nodes are screened out from the regional water consumption burst increment data obtained in step S2. The specific pipeline node or equipment location, as well as the time when the abnormality occurred and the abnormal water flow increment are determined. The state of the detected abnormal node is identified using real-time data provided by monitoring equipment (such as sensors). The node water flow burst increment data is compared and identified with the known pipeline state pattern using a machine learning model or professional domain knowledge. The specific state of the abnormal node is determined, such as abnormal pressure, abnormal flow, etc. The pipeline state data of the identified abnormal node is classified and analyzed. There are two main classifications: physical fault mode data and monitoring equipment fault data. Physical fault mode data refers to problems with the pipeline itself, such as rupture, blockage, etc.; monitoring equipment fault data refers to problems such as inaccurate information or inability to collect information caused by monitoring equipment. Based on physical fault mode data, a quantitative analysis of water resource loss is performed. The actual amount of water resources lost is estimated based on the specific fault type of the abnormal node and the detected water flow increment. A statistical method or model is used to calculate the approximate value of the actual amount of water lost.

[0093] Step S4: Evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment monitoring accuracy loss data to obtain water consumption statistical loss compensation data;

[0094] In an embodiment of the present invention, the monitoring equipment failure data identified in step S3 is evaluated in detail. The impact of the monitoring equipment failure on the monitoring accuracy and reliability is analyzed to determine the degree of performance loss of each faulty device. Problems such as inaccurate data, reduced collection frequency or inability to collect data caused by the failure of the monitoring equipment are considered. Based on the equipment performance loss evaluation data, the water consumption burst increment data obtained in step S2 is corrected and compensated. The water consumption statistical loss caused by each faulty device is determined, that is, the actual water consumption increment that cannot be accurately captured due to the equipment failure. Using a compensation model or algorithm, the original water consumption burst increment data is adjusted and corrected accordingly according to the specific values ​​in the equipment performance loss evaluation data. Data processing and analysis are performed on the compensated water consumption statistical loss data. Ensure the accuracy and rationality of the compensation data, as well as the accurate reflection of the actual water consumption increment. Statistical methods or models can be used to quantify the monitoring error caused by the failure of the monitoring equipment into a specific value of the actual water consumption statistical loss.

[0095] Step S5: Formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

[0096] In the embodiment of the present invention, the physical fault water resource loss data obtained in step S3 and the water consumption statistical loss compensation data obtained in step S4 are integrated. The actual water resource loss of each pipeline node or region is determined, including the direct loss caused by the physical fault and the indirect loss caused by the monitoring equipment failure. Based on the integrated water resource loss data, a dynamic water resource verification strategy is formulated. The verification frequency, verification method (such as on-site detection, remote monitoring, etc.), verification period and other specific details of each pipeline node or region are determined. Considering the difference in actual loss conditions, it is necessary to adjust the frequency and method of the verification strategy to minimize water resource loss and ensure the accuracy of the data. The formulated water resource dynamic verification strategy is sent to the cloud platform or a dedicated water resource management system. Ensure that the strategy can be effectively executed and implemented on the cloud platform and can automatically perform verification operations. This includes ensuring that the strategy can be seamlessly connected with the existing data acquisition system, monitoring equipment and operating procedures to achieve comprehensive water resource management and monitoring. Execute the formulated water resource dynamic verification strategy on the cloud platform. Perform real-time or regular water resource verification operations according to the verification frequency and period defined in the strategy. Ensure that the verification results can be fed back in real time during the operation, and can compare the actual loss data and compensation data, as well as the performance evaluation data of the monitoring equipment, to make necessary adjustments and optimizations.

[0097] Preferably, step S1 comprises the following steps:

[0098] Step S11: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data;

[0099] Step S12: Cleaning the real-time water resource usage data to obtain cleaned real-time water resource usage data;

[0100] Step S13: normalizing the real-time water resource usage cleaning data to obtain normalized real-time water resource usage data;

[0101] Step S14: performing spatiotemporal pattern recognition on the real-time water resource usage data according to the normalized real-time water resource usage data to obtain spatiotemporal pattern data of water resource usage.

[0102] In an embodiment of the present invention, a water resource monitoring system is configured and installed, including sensors and data acquisition equipment. Water resource usage data is collected in real time through sensors to ensure the real-time and accuracy of the data. The acquisition frequency and data format are determined to meet the needs of subsequent data processing and analysis. The raw data collected from the sensor is preliminarily cleaned and preprocessed. It includes steps such as removing or repairing outliers, processing missing data, and eliminating duplicate records. The quality and integrity of the cleaned data are ensured to prepare for subsequent normalization processing. The cleaned real-time water resource usage data is normalized. A normalization method is determined, such as linear normalization, standardization, etc., so as to unify data from different sensors or different time periods to the same scale. Ensure the consistency and comparability of the data to facilitate subsequent spatiotemporal pattern recognition and analysis. Based on the normalized real-time water resource usage data, the spatiotemporal pattern is recognized and analyzed. Use data mining, machine learning, or statistical analysis methods to identify the spatiotemporal variation pattern of water resource usage. Determine the water use pattern in different time periods and spatial locations, including periodic changes, trend changes, etc. Generate spatiotemporal pattern data of water resource usage as the basic data for anomaly detection and analysis in the subsequent step S2.

[0103] Preferably, step S2 comprises the following steps:

[0104] Step S21: Obtaining historical water consumption data of the region;

[0105] Step S22: performing water demand fluctuation analysis on the historical water consumption data of the region to obtain historical water consumption fluctuation data;

[0106] Step S23: Calculate seasonal demand differences for regional historical water consumption data based on historical water consumption fluctuation data to obtain seasonal demand difference data for water consumption;

[0107] Step S24: performing water consumption anomaly detection on the seasonal demand difference data of water resources consumption according to the spatiotemporal pattern data of water resources consumption to obtain regional water consumption anomaly data;

[0108] Step S25: performing spatial correlation analysis on the regional water consumption abnormal data according to the spatiotemporal pattern data of water resource consumption to obtain the regional water consumption spatial correlation data;

[0109] Step S26: Calculate the water consumption burst increment of the regional water consumption abnormal data according to the regional water consumption spatial correlation data to obtain the water consumption burst increment data.

[0110] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0111] Step S21: Obtaining historical water consumption data of the region;

[0112] In the embodiment of the present invention, the source of the regional historical water use data to be collected is determined. This includes local water departments and water resource monitoring agencies, ensuring that the data contains a sufficiently long time span and a sufficiently high time resolution to support subsequent demand analysis and anomaly detection. Confirm the source and quality of the data, ensure the integrity and accuracy of the data, and collect a large amount of unstructured or structured data from various sources, which meets the conditions of big data.

[0113] Step S22: performing water demand fluctuation analysis on the historical water consumption data of the region to obtain historical water consumption fluctuation data;

[0114] In the embodiment of the present invention, the historical water consumption data obtained in step S21 is subjected to fluctuation analysis. A statistical method or a time series analysis method is used to identify the periodicity and trend changes of historical water consumption. Water consumption fluctuation data is obtained, including information such as long-term trends, seasonal changes, and short-term fluctuations.

[0115] Step S23: Calculate seasonal demand differences for regional historical water consumption data based on historical water consumption fluctuation data to obtain seasonal demand difference data for water consumption;

[0116] In the embodiment of the present invention, the water demand difference in different seasons or time periods in the region is calculated based on the water consumption fluctuation data. The typical water consumption range and fluctuation of each season are determined. The seasonal variation factor is taken into account so that subsequent abnormal detection can be adjusted and analyzed according to the seasonal demand difference.

[0117] Step S24: performing water consumption anomaly detection on the seasonal demand difference data of water resources consumption according to the spatiotemporal pattern data of water resources consumption to obtain regional water consumption anomaly data;

[0118] In the embodiment of the present invention, based on the spatiotemporal pattern data of water resource usage, the seasonal demand difference data of water consumption obtained in step S23 is subjected to anomaly detection. A statistical method, a machine learning model or a rule-based method is used to identify water consumption conditions that exceed the normal range. The regional water consumption abnormality data is output to identify the abnormal water consumption conditions in each time period or season.

[0119] Step S25: performing spatial correlation analysis on the regional water consumption abnormal data according to the spatiotemporal pattern data of water resource consumption to obtain the regional water consumption spatial correlation data;

[0120] In the embodiment of the present invention, the spatial correlation of the regional water consumption anomaly data obtained in step S24 is analyzed based on the spatiotemporal pattern data of water resource consumption. The degree of correlation of water consumption anomalies between different geographical locations or pipeline nodes is determined. The specific pattern and trend of spatial correlation are determined by considering the geographical characteristics of the pipeline network and water resource distribution.

[0121] Step S26: Calculate the water consumption burst increment of the regional water consumption abnormal data according to the regional water consumption spatial correlation data to obtain the water consumption burst increment data.

[0122] In the embodiment of the present invention, the burst increment of water consumption is calculated based on the regional water consumption spatial correlation data obtained in step S25. The definition and calculation method of the burst increment are determined to identify unexpected or abnormal large water consumption increments. The burst increment data of water consumption is output to provide a basis for subsequent abnormal pipeline status identification and water resource loss quantification.

[0123] Preferably, step S25 comprises the following steps:

[0124] Step S251: performing a spatiotemporal distribution analysis of water consumption abnormality hotspots on the regional water consumption abnormality data according to the spatiotemporal pattern data of water resource consumption, and obtaining the spatiotemporal distribution data of water consumption abnormality hotspots;

[0125] Step S252: performing abnormal spatial domain effect analysis on the water consumption abnormal hotspot spatiotemporal distribution data to obtain water consumption abnormal spatial domain effect data;

[0126] Step S253: performing hotspot abnormal trend identification on the water consumption abnormal hotspot spatiotemporal distribution data according to the water consumption abnormal spatial domain effect data to obtain abnormal hotspot trend identification data;

[0127] Step S254: Perform spatial correlation analysis based on the water consumption abnormal spatial domain effect data and abnormal hot spot trend identification data to obtain regional water consumption spatial correlation data.

[0128] In an embodiment of the present invention, the water resource consumption spatiotemporal pattern data is used to perform hot spot analysis on the regional water consumption abnormality data obtained in step S24. The distribution of abnormal water consumption hot spots in each time period or season is determined, that is, the spatial location and time range of abnormal water use. A spatiotemporal distribution map of abnormal water consumption hot spots is drawn to present the spatiotemporal distribution of abnormal hot spots in a visual manner. Spatial statistical analysis is performed on the spatiotemporal distribution data of abnormal water consumption hot spots obtained in step S251. Spatial statistical methods such as Moran's I index and Geary's C index are used to analyze the spatial correlation and aggregation degree of abnormal hot spot data. The distribution pattern and aggregation effect of abnormal hot spots in space are determined to identify whether there is a significant spatial aggregation phenomenon. According to the abnormal spatial field effect data obtained in step S252, the trend change of abnormal water consumption hot spots is identified. The changing trend of abnormal hot spots in different time periods or seasons is analyzed, including changes in duration, frequency and intensity. The development trend of abnormal hot spots is determined to provide timely information support for subsequent water resource management and control. Combined with the abnormal spatial domain effect data obtained in step S252 and the abnormal hotspot trend identification data in step S253, further spatial correlation analysis is performed to determine the spatial correlation degree and influence range between abnormal hotspots in different regions. The propagation path or mutual influence relationship between abnormal hotspots is analyzed to better understand and predict the diffusion and evolution process of water consumption anomalies.

[0129] Preferably, step S26 comprises the following steps:

[0130] Step S261: Perform spatial heterogeneity assessment on the regional water consumption spatial correlation data to obtain regional water consumption spatial heterogeneity data;

[0131] Step S262: performing water consumption abnormality spatial regression analysis on the regional water consumption abnormality data according to the regional water consumption spatial heterogeneity data to obtain water consumption abnormality spatial regression data;

[0132] Step S263: performing cluster analysis on the water consumption abnormal spatial regression data to obtain water consumption abnormal regression cluster data;

[0133] Step S264: Calculate the water consumption burst increment for the regional water consumption abnormal data according to the water consumption abnormal spatial regression data and the water consumption abnormal regression clustering data to obtain the water consumption burst increment data.

[0134] In an embodiment of the present invention, the spatial correlation data of regional water consumption obtained in step S25 is used to perform spatial heterogeneity assessment. The spatial distribution characteristics and differences of water consumption anomalies in different regions are analyzed. The degree of spatial heterogeneity, that is, the spatial changes and differences in water consumption anomalies in different regions, is determined. According to the spatial heterogeneity data of regional water consumption obtained in step S261, a spatial regression analysis is performed on the regional water consumption anomaly data in step S24. A spatial regression model is established, and the spatial regression relationship of abnormal water consumption is analyzed considering the spatial relationship and influencing factors of different regions. The spatial regression data of water consumption anomalies is obtained to describe the trend and pattern of abnormal water consumption with spatial variation. The spatial regression data of water consumption anomalies obtained in step S262 is clustered. Using a clustering algorithm, such as K-means clustering or hierarchical clustering, abnormal water consumption areas with similar regression characteristics are divided into different categories or groups. The characteristics and commonalities of each cluster are determined, and the typical spatial patterns and trends of abnormal water consumption in different categories are identified. The water consumption burst increment is calculated based on the abnormal spatial regression data and abnormal regression cluster data of water consumption obtained in step S263. The burst increment in each abnormal water consumption category or cluster is analyzed, that is, the significant increase in abnormal water consumption under specific spatial conditions. The water consumption burst increment data is output to provide accurate quantitative indicators and basis for the control and management of abnormal water consumption.

[0135] Preferably, step S3 comprises the following steps:

[0136] Step S31: marking the regional nodes of the burst increment data of water consumption to obtain the burst increment node marking data;

[0137] Step S32: Calculate the burst increment of water flow between different nodes based on the burst increment node mark data according to the burst increment data of water consumption, and obtain the burst increment data of node water flow;

[0138] Step S33: Identify the abnormal node pipeline state according to the node water flow burst increment data and the burst increment node mark data to obtain the abnormal node pipeline state data;

[0139] Step S34: using the node abnormal pattern classification model to perform abnormal pattern classification on the abnormal node pipeline state data to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data;

[0140] Step S35: locating the physical fault node according to the physical fault mode data to obtain physical fault node locating data; evaluating the fault impact range of the physical fault node locating data to obtain node fault impact range evaluation data;

[0141] Step S36: quantify the water resource loss according to the node fault impact range assessment data and the node water flow burst increment data to obtain the physical fault water resource loss data.

[0142] In the embodiment of the present invention, the burst increment data of water consumption obtained in step S26 is used to mark the burst increment nodes in the region. Determine which regional nodes have a significant increase in water consumption in a specific time period, and mark these nodes as burst increment nodes. Output the burst increment node marking data, including the node ID and the relevant time period information. According to the burst increment node marking data, calculate the burst increment of water flow between different nodes. Analyze the change of water flow between each burst increment node to obtain the water flow increment data of each node. Output the node water flow burst increment data to describe the change of water flow between the burst increment nodes. Combine the node water flow burst increment data and the burst increment node marking data obtained in step S32 to identify the state of the abnormal node pipeline. According to the size and frequency of the water flow increment, determine whether the node is in an abnormal state. Output the abnormal node pipeline state data to identify which nodes have abnormal conditions of pipelines or equipment. Use the pre-established node abnormal pattern classification model to classify the abnormal node pipeline state data. The abnormal nodes are divided into different categories such as physical failure mode and monitoring equipment failure mode according to their abnormal modes. Output pipeline abnormal pattern classification data to describe the characteristics and classification results of each abnormal mode. Based on the physical fault mode data, determine the specific node location of the physical fault. Analyze the impact range of the physical fault node on the surrounding pipelines and equipment. Output the physical fault node location data and the fault impact range assessment data to describe the faulty node and its potential impact area. Combine the node fault impact range assessment data and the node water flow burst increment data in step S35 to quantify the water resource loss. Estimate the amount of water resource loss caused by the physical fault, taking into account the water flow changes and impact range around the abnormal node. Output the physical fault water resource loss data and provide indicators and data for quantifying the actual loss.

[0143] Preferably, the step of constructing the node abnormal pattern classification model includes the following steps:

[0144] Obtain pipeline structure design data and sensor structure design data;

[0145] Carry out water flow scour limit bearing calculation on pipeline structure design data to obtain pipeline structure limit bearing data;

[0146] Calculate the stress limit of the pipeline structure design data according to the pipeline structure limit bearing data to obtain the pipeline structure stress bearing limit data;

[0147] Perform humidity adaptability simulation on sensor structure design data to obtain sensor humidity adaptability simulation data;

[0148] According to the sensor humidity adaptability simulation data, the sensor structure design data is subjected to sensitivity condition impairment analysis to obtain the sensor sensitivity condition impairment data;

[0149] Based on the PolicyGradient algorithm, feature learning is performed on the pipeline structure stress bearing limit data and sensor sensitivity condition reduction data to obtain abnormal mode characteristic learning data;

[0150] Based on the convolutional neural network, an abnormal pattern analysis model is constructed for the abnormal pattern characteristic learning data to obtain a node abnormal pattern classification model.

[0151] In an embodiment of the present invention, the design data of the pipeline system is collected, including information such as pipeline materials, dimensions, and layout. The sensor design data is collected, including detailed information such as sensor type, location, and working principle. The water flow scouring limit bearing calculation is performed according to the pipeline structure design data. The ultimate bearing capacity of the pipeline under different water flow conditions is determined, taking into account the impact force and stress distribution of the water flow on the pipeline. The ultimate bearing data of the pipeline structure is obtained. Stress analysis and calculation are performed based on the ultimate bearing data of the pipeline structure. The bearing limit of the pipeline under different stress conditions is determined, taking into account the strength of the material and the safety of the structure. The stress bearing limit data of the pipeline structure is obtained. The humidity adaptability simulation is performed on the sensor structure design data. The performance of the sensor under different humidity conditions is simulated, taking into account the influence of humidity on materials and electronic components. Based on the humidity adaptability simulation data, the sensitivity condition impairment analysis of the sensor is performed. The influence of different humidity on the sensitivity and response speed of the sensor is analyzed, and the performance stability under different working environments is evaluated. The PolicyGradient algorithm is used to perform feature learning on the pipeline structure stress bearing limit data and the sensor sensitivity condition impairment data. The association and weight between different features are learned to capture the characteristics of abnormal patterns. Based on the aforementioned learned abnormal pattern characteristic data, a convolutional neural network model is constructed. Design network structures and hierarchies to identify and classify abnormal patterns of nodes. Determine the characteristics and identification criteria of each abnormal pattern to support subsequent abnormal pattern classification and diagnosis.

[0152] Preferably, step S4 comprises the following steps:

[0153] Step S41: extracting the equipment fault status code from the monitoring equipment fault data to obtain the equipment fault status code;

[0154] Step S42: performing equipment performance loss assessment on the monitoring equipment failure data according to the equipment failure status code to obtain equipment performance loss assessment data;

[0155] Step S43: Calculate the equipment monitoring precision loss according to the equipment performance loss evaluation data and the monitoring equipment failure data to obtain the equipment monitoring precision loss data;

[0156] Step S44: Compensate the water consumption statistics loss for the sudden increase in water consumption data according to the equipment monitoring accuracy loss data to obtain water consumption statistics loss compensation data.

[0157] In an embodiment of the present invention, monitoring equipment fault data is collected, including information such as timestamp and fault type. The data is preprocessed to ensure the integrity and accuracy of the data, and the equipment fault status code is extracted from the monitoring equipment fault data. The equipment fault status code usually indicates the specific type or classification of the fault, such as battery failure, sensor failure, etc. According to the equipment fault status code, the performance loss degree of the monitoring equipment is evaluated. The impact of each fault state on the normal operation of the monitoring equipment is analyzed, such as reduced data acquisition frequency, reduced accuracy, etc. Further analysis is performed in combination with the equipment performance loss evaluation data and the monitoring equipment fault data. The specific value of the equipment monitoring accuracy loss is calculated to quantify the reduction in monitoring accuracy of the equipment under different fault conditions. The water consumption data related to the monitoring equipment is collected, especially the burst increment data during the fault period. The actual change in water consumption during the fault period is determined and compared with the expected water consumption. The burst increment data of water consumption is corrected according to the equipment monitoring accuracy loss data. The purpose of the correction is to compensate for the error caused by the reduction in water consumption statistical accuracy caused by the failure of the monitoring equipment, so as to ensure the accuracy and reliability of the water consumption data.

[0158] Preferably, step S44 includes the following steps:

[0159] Step S441: performing precision loss quantification processing on the equipment monitoring precision loss data to obtain precision loss quantification data;

[0160] Step S442: performing precision loss gradient sorting on the equipment monitoring precision loss data according to the precision loss quantification data to obtain precision loss gradient sorting data;

[0161] Step S443: using Lagrange multipliers to perform precision loss gradient constraints on the precision loss gradient sorting data to obtain precision loss gradient constrained data;

[0162] Step S444: performing quasi-Newton condition constraint calculation on the precision loss gradient sorting data according to the precision loss gradient constraint data to obtain device precision loss gradient approximation data;

[0163] Step S445: Perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment accuracy loss gradient approximation data to obtain water consumption statistical loss compensation data.

[0164] In an embodiment of the present invention, equipment monitoring accuracy loss data is obtained, which generally includes the error or deviation of the monitoring data under different fault states. The equipment monitoring accuracy loss data is quantified and converted into numerical accuracy loss data. The quantification process can convert the loss data into a comparable and analyzable form based on a mathematical model or a statistical method. According to the accuracy loss quantification data, the monitoring accuracy loss under different fault states is sorted. The purpose of the sorting is to determine which fault states have the greatest or least impact on the monitoring accuracy. The Lagrange multiplier is used to constrain the accuracy loss gradient sorting data. The purpose of this step is to further optimize or constrain the gradient information of the accuracy loss based on the previous sorting results, so as to more accurately quantify the degree of impact under different fault states. According to the accuracy loss gradient constraint data, the quasi-Newton condition constraint calculation method is applied. This step helps to further optimize the accuracy loss gradient data to approximate the monitoring accuracy changes of the equipment under different fault states. According to the equipment accuracy loss gradient approximation data, the burst increment data of water consumption during the monitoring period is corrected. The purpose of the correction is to compensate for the statistical error of water consumption caused by the failure of the monitoring equipment and ensure the accuracy and reliability of the water consumption data.

[0165] Preferably, the present invention further provides a water resources dynamic quantity verification system based on big data analysis, which is used to execute the water resources dynamic quantity verification method based on big data analysis as described above, and the water resources dynamic quantity verification system based on big data analysis includes:

[0166] The spatiotemporal pattern recognition module is used to collect real-time water resource usage data through sensors to obtain real-time water resource usage data; perform spatiotemporal pattern recognition on the real-time water resource usage data to obtain water resource spatiotemporal pattern data;

[0167] The incremental analysis module is used to detect water consumption anomalies based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data;

[0168] The fault classification module is used to identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; to classify the abnormal node pipeline state data into abnormal patterns to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data; to quantify water resource loss according to the physical fault pattern data to obtain physical fault water resource loss data;

[0169] The equipment accuracy loss compensation module is used to evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; to compensate for the water consumption statistical loss of the sudden increase data of water consumption according to the equipment monitoring accuracy loss data to obtain the water consumption statistical loss compensation data;

[0170] The strategy formulation module is used to formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

[0171] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0172] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A water resources dynamic quantity method based on big data analysis, characterized in that: The following steps are involved: Step S1: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data; Perform spatiotemporal pattern recognition on real-time water resource usage data to obtain spatiotemporal pattern data of water resources; Step S2: Perform water consumption anomaly detection based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; Perform spatial correlation analysis on the regional water consumption abnormal data to obtain the regional water consumption spatial correlation data; perform water consumption sudden increment calculation on the regional water consumption abnormal data based on the regional water consumption spatial correlation data to obtain the water consumption sudden increment data; Step S3: Identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; classify the abnormal node pipeline state data by abnormal mode to obtain pipeline abnormal mode classification data, wherein the pipeline abnormal mode classification data includes physical fault mode data and monitoring equipment fault data; quantify the water resource loss according to the physical fault mode data to obtain physical fault water resource loss data; Step S4: Evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment monitoring accuracy loss data to obtain water consumption statistical loss compensation data; Step S5: Formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

2. The water resources dynamic quantity method based on big data analysis according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: collecting real-time water resource usage data through sensors to obtain real-time water resource usage data; Step S12: Cleaning the real-time water resource usage data to obtain cleaned real-time water resource usage data; Step S13: normalizing the real-time water resource usage cleaning data to obtain normalized real-time water resource usage data; Step S14: performing spatiotemporal pattern recognition on the real-time water resource usage data according to the normalized real-time water resource usage data to obtain spatiotemporal pattern data of water resource usage.

3. The water resources dynamic quantity method based on big data analysis according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: Obtaining historical water consumption data of the region; Step S22: performing water demand fluctuation analysis on the historical water consumption data of the region to obtain historical water consumption fluctuation data; Step S23: Calculate seasonal demand differences for regional historical water consumption data based on historical water consumption fluctuation data to obtain seasonal demand difference data for water consumption; Step S24: performing water consumption anomaly detection on the seasonal water demand difference data according to the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; Step S25: performing spatial correlation analysis on the regional water consumption abnormal data according to the spatiotemporal pattern data of water resource consumption to obtain the regional water consumption spatial correlation data; Step S26: Calculate the water consumption burst increment of the regional water consumption abnormal data according to the regional water consumption spatial correlation data to obtain the water consumption burst increment data.

4. The water resources dynamic quantity method based on big data analysis according to claim 3 is characterized in that: Step S25 includes the following steps: Step S251: performing a spatiotemporal distribution analysis of water consumption abnormality hotspots on the regional water consumption abnormality data according to the spatiotemporal pattern data of water resource consumption, and obtaining the spatiotemporal distribution data of water consumption abnormality hotspots; Step S252: performing abnormal spatial domain effect analysis on the water consumption abnormal hotspot spatiotemporal distribution data to obtain water consumption abnormal spatial domain effect data; Step S253: performing hotspot abnormal trend identification on the water consumption abnormal hotspot spatiotemporal distribution data according to the water consumption abnormal spatial domain effect data to obtain abnormal hotspot trend identification data; Step S254: Perform spatial correlation analysis based on the water consumption abnormal spatial domain effect data and abnormal hot spot trend identification data to obtain regional water consumption spatial correlation data.

5. The water resources dynamic quantity method based on big data analysis according to claim 3 is characterized in that: Step S26 includes the following steps: Step S261: Perform spatial heterogeneity assessment on the regional water consumption spatial correlation data to obtain regional water consumption spatial heterogeneity data; Step S262: performing water consumption abnormality spatial regression analysis on the regional water consumption abnormality data according to the regional water consumption spatial heterogeneity data to obtain water consumption abnormality spatial regression data; Step S263: performing cluster analysis on the water consumption abnormal spatial regression data to obtain water consumption abnormal regression cluster data; Step S264: Calculate the water consumption burst increment of the regional water consumption abnormal data according to the water consumption abnormal spatial regression data and the water consumption abnormal regression clustering data to obtain the water consumption burst increment data.

6. The water resources dynamic quantity method based on big data analysis according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: marking the regional nodes of the burst increment data of water consumption to obtain the burst increment node marking data; Step S32: Calculate the burst increment of water flow between different nodes based on the burst increment node mark data according to the burst increment data of water consumption, and obtain the burst increment data of node water flow; Step S33: Identify the abnormal node pipeline state according to the node water flow burst increment data and the burst increment node mark data to obtain the abnormal node pipeline state data; Step S34: using the node abnormal pattern classification model to perform abnormal pattern classification on the abnormal node pipeline state data to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data; Step S35: locating the physical fault node according to the physical fault mode data to obtain physical fault node locating data; evaluating the fault impact range of the physical fault node locating data to obtain node fault impact range evaluation data; Step S36: quantify the water resource loss according to the node fault impact range assessment data and the node water flow burst increment data to obtain the physical fault water resource loss data.

7. The water resources dynamic quantity method based on big data analysis according to claim 6 is characterized in that: The steps for building the node abnormal pattern classification model include the following steps: Obtain pipeline structure design data and sensor structure design data; Carry out water flow scour limit bearing calculation on pipeline structure design data to obtain pipeline structure limit bearing data; Calculate the stress limit of the pipeline structure design data according to the pipeline structure limit bearing data to obtain the pipeline structure stress bearing limit data; Perform humidity adaptability simulation on sensor structure design data to obtain sensor humidity adaptability simulation data; According to the sensor humidity adaptability simulation data, the sensor structure design data is subjected to sensitivity condition impairment analysis to obtain the sensor sensitivity condition impairment data; Based on the PolicyGradient algorithm, feature learning is performed on the pipeline structure stress bearing limit data and sensor sensitivity condition reduction data to obtain abnormal mode characteristic learning data; Based on the convolutional neural network, an abnormal pattern analysis model is constructed for the abnormal pattern characteristic learning data to obtain a node abnormal pattern classification model.

8. The water resources dynamic quantity method based on big data analysis according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: extracting the equipment fault status code from the monitoring equipment fault data to obtain the equipment fault status code; Step S42: performing equipment performance loss assessment on the monitoring equipment failure data according to the equipment failure status code to obtain equipment performance loss assessment data; Step S43: Calculate the equipment monitoring precision loss according to the equipment performance loss evaluation data and the monitoring equipment failure data to obtain the equipment monitoring precision loss data; Step S44: Compensate the water consumption statistics loss for the sudden increase in water consumption data according to the equipment monitoring accuracy loss data to obtain water consumption statistics loss compensation data.

9. The water resources dynamic quantity method based on big data analysis according to claim 8 is characterized in that: Step S44 includes the following steps: Step S441: performing precision loss quantification processing on the equipment monitoring precision loss data to obtain precision loss quantification data; Step S442: performing precision loss gradient sorting on the equipment monitoring precision loss data according to the precision loss quantification data to obtain precision loss gradient sorting data; Step S443: using Lagrange multipliers to perform precision loss gradient constraints on the precision loss gradient sorting data to obtain precision loss gradient constrained data; Step S444: performing quasi-Newton condition constraint calculation on the precision loss gradient sorting data according to the precision loss gradient constraint data to obtain device precision loss gradient approximation data; Step S445: Perform water consumption statistical loss compensation on the water consumption sudden increment data according to the equipment accuracy loss gradient approximation data to obtain water consumption statistical loss compensation data.

10. A water resources dynamic quantity system based on big data analysis, characterized in that: Used to execute the water resources dynamic quantity verification method based on big data analysis as claimed in claim 1, the water resources dynamic quantity verification system based on big data analysis includes: The spatiotemporal pattern recognition module is used to collect real-time water resource usage data through sensors to obtain real-time water resource usage data; perform spatiotemporal pattern recognition on the real-time water resource usage data to obtain water resource spatiotemporal pattern data; The incremental analysis module is used to detect water consumption anomalies based on the spatiotemporal pattern data of water resource consumption to obtain regional water consumption anomaly data; perform spatial correlation analysis on the regional water consumption anomaly data to obtain regional water consumption spatial correlation data; perform water consumption burst increment calculation on the regional water consumption anomaly data based on the regional water consumption spatial correlation data to obtain water consumption burst increment data; The fault classification module is used to identify the abnormal node pipeline state according to the node water flow burst increment data to obtain the abnormal node pipeline state data; to classify the abnormal node pipeline state data into abnormal patterns to obtain pipeline abnormal pattern classification data, wherein the pipeline abnormal pattern classification data includes physical fault pattern data and monitoring equipment fault data; to quantify water resource loss according to the physical fault pattern data to obtain physical fault water resource loss data; The equipment accuracy loss compensation module is used to evaluate the equipment performance loss of the monitoring equipment failure data to obtain equipment performance loss evaluation data; to compensate for the water consumption statistical loss of the sudden increase data of water consumption according to the equipment monitoring accuracy loss data to obtain the water consumption statistical loss compensation data; The strategy formulation module is used to formulate a dynamic water resource quantity verification strategy based on the water resource loss data caused by physical faults and the water consumption statistical loss compensation data, obtain the dynamic water resource quantity verification strategy, and send the dynamic water resource quantity verification strategy to the cloud platform to execute the dynamic water resource quantity verification operation.

Images