A smart community management method based on big data

Abstract

The present invention relates to the technical field of smart community management based on big data, and discloses a smart community management method based on big data. A three-in-one sensor is installed every 50 meters of a water supply pipe and every 100 meters of a drainage pipe, and a unique number is assigned to each sensor. A coordinate system is established with the midpoint of the community surface as the origin, and data is collected every 4 minutes to form a data point. After running for 48 hours, the baseline value of each sensor is calculated to form a baseline value data set. The high-density sensor network captures subtle changes and improves the accuracy of hidden danger detection. The unique number and coordinate system solve the problems of data confusion and unclear positioning, making it easier to lock the problem area. Data is collected every 4 minutes and the baseline value is calculated to overcome the defects of insufficient static standards, establish a dynamic benchmark, and enhance the adaptability of anomaly detection. The baseline value data set analysis solves the problem of lack of systematic basis for hidden danger detection and improves scientificity and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of smart community management based on big data, and specifically to a smart community management method based on big data. Background Art

[0002] As an important part of modern urban management, smart communities aim to improve the quality of life of residents and the efficiency of community management through information technology. The operation and management of water supply and drainage pipelines are key links in the construction of smart communities. The pipeline system is directly related to the daily water use and environmental sanitation of residents, and its safety and stability are particularly important. However, with the expansion of community scale and aging of pipelines, problems such as leakage and blockage have become increasingly prominent, posing severe challenges to water resource utilization and community safety. Traditional pipeline management relies heavily on manual inspections and regular maintenance, which is not only inefficient, but also difficult to achieve real-time monitoring and rapid response. Therefore, how to use big data and intelligent technology to optimize pipeline management has become an urgent problem to be solved in the development of smart communities. Although existing pipeline management technologies have made certain progress, there are still many shortcomings, such as limited monitoring range, delayed abnormal detection, and inaccurate leakage positioning. These problems limit the further improvement of the management level of smart communities.

[0003] In the prior art, pipeline monitoring usually uses a single type of sensor, such as monitoring only water pressure or flow, with a low installation density, making it difficult to fully cover the pipeline network, resulting in blind spots in operation status monitoring. Community pipeline systems are complex and widely distributed. If multi-dimensional data cannot be collected in real time, it is difficult to capture subtle changes in pipeline status, which in turn affects the accuracy and coverage of hidden danger inspections. In addition, the prior art often lacks systematicity in data management, and sensor data is not uniformly numbered or located, resulting in confusion in data sources and unclear spatial distribution. It is difficult for managers to quickly lock in problem areas, reducing the efficiency of inspections. These problems make traditional methods appear powerless when facing large-scale pipeline networks. Anomaly detection is an important part of pipeline management, but the prior art also has obvious shortcomings in this regard. Traditional anomaly judgments are mostly based on static thresholds or manual experience, lacking scientific dynamic benchmarks, resulting in low accuracy in anomaly identification. For example, pipeline operation is affected by environmental factors such as temperature. A single static standard cannot reflect individual differences, and the basis for anomaly judgment is insufficient, which is prone to false alarms or missed reports. At the same time, the existing systems are mostly limited to simple statistics in data analysis, and fail to make full use of big data technology for systematic analysis. There is a lack of objective basis for hidden danger inspection, and omissions or misjudgments are prone to occur when relying on manual judgment, which reduces the reliability of safety management. Leakage location is another difficult problem in pipeline management. Existing leakage detection technology is usually achieved through acoustic wave detection or manual inspection, with low positioning accuracy and long time consumption. Due to the lack of support from intelligent prediction models, it is difficult for existing methods to quickly calculate the coordinates of leakage points through sensor data, especially in complex pipeline networks. The fuzzy spatial relationship and insufficient parameter correlation analysis make it difficult to accurately locate the leakage location. In addition, the existing technology lacks a confidence assessment mechanism in leakage prediction, and the reliability of the prediction results is difficult to verify, resulting in low trust in the system output by managers, affecting decision-making efficiency.

[0004] In summary, this case aims to propose a smart community management method based on big data. Through high-density sensor deployment, multi-parameter real-time analysis, intelligent prediction model, automated response and data feedback optimization, it can comprehensively improve the efficiency and reliability of pipeline management and provide technical support for the construction of smart communities. Summary of the invention

[0005] The present invention provides a smart community management method based on big data, which promotes solving the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: a smart community management method based on big data, comprising: S1. Deployment of sensors and data collection: Install three-in-one sensors every 50 meters in the water supply pipe and every 100 meters in the drainage pipe; The three-in-one sensor includes: water pressure sensor , Flow Sensor and temperature sensor ; Assign a unique number to each 3-in-1 sensor ,in ; Get the total length of the water supply pipeline, recorded as ; Get the total length of the drainage pipe, recorded as ; Calculate the total number of three-in-one sensors, recorded as : ; in, To round up; Get the area covered by the community on the ground, recorded as the community surface; Get the midpoint of the community surface, take the midpoint of the community surface as the vertical intersection point, make two mutually perpendicular rays on the community surface, and use one of the rays as axis coordinates, and the other ray as Axis coordinates, a coordinate system with the midpoint of the community surface as the origin and two mutually perpendicular rays as coordinate axes is obtained; Data is collected every 4 minutes. The collected data includes: , three-in-one sensor number , water pressure value , flow value , temperature value , three-in-one sensor Axis coordinates , three-in-one sensor Axis coordinates , and form data points , ; Among them, the subscript Indicates 3-in-1 sensor; Run the system for 48 hours and calculate the baseline value for each 3-in-1 sensor: ; ; ; in, For the The reference water pressure of a three-in-one sensor; For the Base flow of a three-in-one sensor; For the The reference temperature of a three-in-one sensor; For the Three-in-one sensor The water pressure collected at the time; For the Three-in-one sensor The amount of traffic collected; For the Three-in-one sensor The temperature of the first collection; Will , , and Composition of benchmark data set ; get ; S2, analyze the collected data in real time and detect anomalies; S3, intelligent prediction model development and leakage location; S4, automated response and maintenance scheduling; S5. Data feedback and system optimization.

[0007] Optionally, the real-time analysis of the collected data and detection of anomalies specifically includes: Set the water pressure deviation rate algorithm to calculate the The water pressure deviation rate after temperature correction of the three-in-one sensor is expressed as : ; Set the flow deviation rate algorithm to calculate the The flow deviation rate after temperature correction of the three-in-one sensor is recorded as : ; Set the temperature deviation rate algorithm to calculate the The temperature deviation rate of the three-in-one sensor is expressed as : ; Calculate the water pressure weight, denoted as : ; like ,but ; Calculate the traffic weight, recorded as : ; like ,but ; Calculate the temperature weight, denoted as : ; like ,but .

[0008] Optionally, the real-time analysis of the collected data and detection of anomalies specifically includes: Set up a real-time anomaly detection algorithm to calculate the Abnormal index of three-in-one sensors: ; in, For the Abnormal index of three-in-one sensors; Set the anomaly index threshold, denoted as ; like , then the The data collected by the three-in-one sensor is considered abnormal and Stored in the alarm data table; Set the abnormal sensor set, denoted as , ; in, For the The number of the abnormal sensor; is the total number of abnormal sensors.

[0009] Optionally, the intelligent prediction model development and leakage location specifically include: For each , calculate the The minimum value of the Euclidean distance between a three-in-one sensor and all other three-in-one sensors is determined : ; Set the water pressure gradient factor algorithm to calculate the The water pressure gradient factor of the three-in-one sensor is expressed as : ; Set the flow gradient factor algorithm to calculate the The flow gradient factor of a three-in-one sensor is expressed as : .

[0010] Optionally, the intelligent prediction model development and leakage location specifically include: Set the leakage location algorithm and locate the leakage point coordinates : ; in, To predict the leakage point coordinate; To predict the leakage point coordinate; For the search process coordinate; For the search process coordinate; The confidence is calculated by the confidence algorithm, which is recorded as : ; in, is the total number of abnormal sensors; Will Stored in the leakage prediction data table.

[0011] Optionally, the automated response and maintenance scheduling specifically includes: like , generating an alert; The alert format is: , pushed to the maintenance team through the system; Get the real-time location of the maintenance personnel; Calculate the distance from the maintenance worker to the leakage point, recorded as : ; in, For maintenance personnel coordinate; For maintenance personnel coordinate; Select The youngest maintenance worker and assigns the maintenance task to this maintenance worker; Check the location of the smart valve near the leakage point; Calculate the distance from the smart valve to the leakage point, recorded as : ; choose The smallest smart valve is instructed to reduce water pressure to 40%; Get the list of materials used in the repair, recorded as ; Get the maintenance person's number, recorded as ; The maintenance tasks are grouped into a maintenance task data set, which is specifically: ; Store the maintenance task dataset in a data table middle.

[0012] Optionally, the data feedback and system optimization specifically include: Maintenance personnel submit actual leakage points and leakage causes; Calculate the prediction error, denoted as : ; in, The actual leakage coordinate; The actual leakage coordinate; like m, adjust the weights: ; ; ; Calculate the false alarm rate, denoted as : ; in, The number of false alarms in the last 100 alarms; like ,but ; The optimized parameters Store to data table middle.

[0013] The present invention has the following beneficial effects: 1. By deploying three-in-one sensors every 50 meters in the water supply pipeline and every 100 meters in the drainage pipeline, the problem of blind spots in pipeline operation status monitoring is solved, and the effect of full coverage and accurate perception is achieved. In community management, the safe operation of water supply and drainage pipelines is crucial, but traditional methods often make it difficult to achieve real-time monitoring of the entire pipeline section. By installing three-in-one sensors containing water pressure, flow and temperature every 50 meters in the water supply pipeline and every 100 meters in the drainage pipeline, the system can fully cover the pipeline network and avoid the occurrence of monitoring blind spots. This high-density deployment combined with the multi-parameter acquisition function of the three-in-one sensor enables subtle changes in the pipeline operation status to be captured, providing a rich data foundation for subsequent analysis, thereby significantly improving the coverage and accuracy of hidden danger detection. By assigning a unique number to each three-in-one sensor and establishing a coordinate system, the problems of confusing data sources and unclear spatial positioning are solved, and the effect of data traceability and spatial analysis is achieved. In large-scale pipeline networks, confusion in data sources and lack of location information may lead to inefficient hidden danger detection. By assigning a unique number to each sensor and establishing a coordinate system with the midpoint of the community surface as the origin, the sensor position is accurately determined to the X-axis and Y-axis coordinates, and the system can clearly track the source of each set of data and locate its spatial position. This method not only solves the problem of multi-sensor data confusion, but also lays the foundation for the spatial distribution analysis of hidden dangers, allowing managers to quickly lock in problem areas and improve the pertinence of investigation and response. By collecting data every 4 minutes and running for 48 hours to calculate the benchmark value, the problem of insufficient basis for judging the abnormal operation status of the pipeline is solved, and the basic effect of dynamic benchmark establishment and abnormal detection is achieved. The abnormal judgment of the pipeline operation status requires a scientific benchmark as a reference, while traditional methods often rely on static standards and are difficult to adapt to actual changes. By collecting data containing time, number, water pressure, flow, temperature and coordinates every 4 minutes, and calculating the benchmark water pressure, benchmark flow and benchmark temperature of each sensor within 48 hours, the system can generate dynamic benchmark values ​​based on actual operation data. This method overcomes the defect that a single standard cannot reflect the individual differences of pipelines, and provides a reliable personalized reference standard for subsequent abnormal detection, thereby enhancing the accuracy and adaptability of the early warning system. By forming a benchmark data set and analyzing it, the problem of lack of systematic basis for hidden danger detection is solved, and data-driven safety management effect is achieved. If hidden danger detection relies solely on manual experience, omissions or misjudgments are prone to occur. By forming a benchmark data set with the benchmark water pressure, benchmark flow and benchmark temperature of each sensor, the system provides a comprehensive digital description of the safety status of the pipeline. This data-driven approach makes hidden danger detection no longer rely on subjective judgment, but is based on objective benchmark value analysis, which can systematically identify potential risks. The ultimate effect is to improve the scientificity and reliability of safety management.

[0014] 2. By setting the water pressure, flow and temperature deviation rate algorithms and performing temperature correction, the problem of inaccurate judgment of pipeline operation parameter anomalies is solved, and a more accurate anomaly detection effect is achieved. In pipeline monitoring, water pressure, flow and temperature anomalies may be important signals of safety hazards, but the deviation of a single parameter may be interfered with by environmental factors, such as temperature fluctuations, leading to misjudgment. By setting the water pressure deviation rate algorithm, flow deviation rate algorithm and temperature deviation rate algorithm for the i-th three-in-one sensor, and introducing temperature correction in the calculation of water pressure and flow deviation rates, the system can eliminate the interference of temperature changes on water pressure and flow. This method solves the problem of inaccurate anomaly judgment caused by confusion of environmental variables in traditional detection, thereby significantly improving the accuracy and reliability of anomaly detection and providing a more scientific basis for hidden danger investigation. By introducing weight calculation and setting conditional judgment, the problem of single anomaly severity assessment is solved, and a dynamic priority management effect is achieved. In pipeline operation, the importance of anomalies of different parameters may vary with specific circumstances. Traditional methods often use fixed thresholds, which are difficult to respond flexibly. By calculating the water pressure weight, flow weight and temperature weight, and setting the conditional if ,but , the system dynamically adjusts the weight of each parameter according to the size of the deviation rate. This step solves the problem that the abnormal evaluation is too single and cannot distinguish between the important and the urgent, so that the system can give priority to the parameters with large deviations, and then realize dynamic priority management. The final effect is to improve the efficiency of resource allocation and ensure that serious hidden dangers are dealt with in a timely manner. By analyzing the collected data in real time and combining the deviation rate and weight, the problem of delayed hidden danger investigation is solved, and the effect of real-time warning and rapid response is achieved. In community management, if the hidden danger investigation fails to detect abnormalities in time, accidents may occur. By analyzing the collected data in real time, calculating the water pressure deviation rate, flow deviation rate and temperature deviation rate of each sensor, and combining the weight for comprehensive evaluation, the system can identify and quantify the risk at the first time when the abnormality occurs. This real-time processing method overcomes the defect of strong lag of traditional regular inspections, making the hidden danger investigation from passive to active. The final effect is to realize the real-time warning function, which buys valuable time for managers to take countermeasures quickly and reduces safety risks. Through the algorithm design of comprehensive deviation rate and weight, the problem of insufficient correlation analysis of multi-parameter abnormalities is solved, and the systematic safety status evaluation effect is achieved. During pipeline operation, water pressure, flow rate and temperature are often interrelated, and analyzing a single parameter alone may ignore the overall risk. By combining the deviation rates of water pressure, flow rate and temperature with the corresponding weights, the system can comprehensively evaluate the degree of abnormality of multiple parameters and their mutual influence. This method solves the limitation of insufficient multi-parameter correlation analysis in traditional detection, making the safety status assessment more systematic and comprehensive.

[0015] 3. By setting a real-time anomaly detection algorithm and calculating the anomaly index, the problem of the difficulty in quantifying the abnormality of the pipeline operation status is solved, and the accurate evaluation effect of the degree of abnormality is achieved. In pipeline safety monitoring, the abnormality of the operation status is often difficult to judge intuitively. Traditional methods rely on experience or simple thresholds and lack scientific basis. By setting a real-time anomaly detection algorithm for the i-th three-in-one sensor and calculating the anomaly index, the system can integrate the deviation degree of multiple parameters such as water pressure, flow and temperature into a quantitative indicator. This step solves the problem that the judgment of abnormality is highly subjective and difficult to quantify, so that the degree of abnormality of the pipeline operation status can be clearly expressed by numerical values. The final effect is to achieve the precision and standardization of abnormal evaluation, providing a reliable basis for subsequent decision-making. By setting the anomaly index threshold and filtering abnormal data, the problem of low efficiency of abnormal identification is solved, and the effect of quickly locking hidden dangers is achieved. In the real-time analysis of a large amount of sensor data, it will take a lot of time and resources to check the anomalies one by one. By setting the anomaly index threshold and stipulating that when the threshold is exceeded, the data of the i-th sensor is identified as abnormal and stored in the alarm data table, the system can automatically filter normal data and quickly focus on abnormal points. This method overcomes the defects of low efficiency of traditional full-volume analysis and greatly reduces the interference of invalid data. The final effect is to quickly lock the hidden dangers, so that managers can quickly locate the problem sensors and improve the efficiency of troubleshooting. By establishing an abnormal sensor set and recording the number, the problem of unclear traceability of abnormal data is solved, and the traceability of abnormal locations is achieved. In a complex pipeline network, if the source cannot be accurately traced after the abnormality occurs, it will increase the difficulty of handling. By setting up an abnormal sensor set and recording the number of each abnormal sensor and its total number, the system can clearly mark the specific location and scope of the abnormality. This step solves the problem of confusing sources and difficult positioning of abnormal data, and provides clear guidance for subsequent maintenance or emergency response. The final effect is to achieve traceability of abnormal locations and efficient management, ensuring that hidden dangers are handled in a targeted manner. By analyzing and storing alarm data in real time, the problem of delayed warning of hidden dangers is solved, and the effect of real-time monitoring and timely response is achieved. During pipeline operation, if hidden dangers cannot be discovered and warned in time, accidents may expand. By analyzing the collected data in real time, calculating the abnormal index, and storing it in the alarm data table immediately when the threshold is exceeded, the system can issue an alarm as soon as the abnormality occurs. This real-time processing method overcomes the problem of delayed traditional periodic inspections, turning hidden danger inspections from passive to active. The ultimate effect is to achieve real-time monitoring and timely response capabilities, significantly reducing safety risks.

[0016] 4. By calculating the Euclidean distance between each three-in-one sensor and other sensors and taking the minimum value, the problem of fuzzy spatial relationship in leakage positioning is solved, and the accurate identification effect of the neighboring relationship of potential leakage points is achieved. In the pipeline system, leakage is often closely related to the spatial relationship between sensors, but traditional methods are difficult to accurately determine the spatial correlation between abnormal points and the surrounding environment. By calculating the Euclidean distance between the i-th three-in-one sensor and all other sensors and taking the minimum value to determine, the system can identify the position of the nearest neighboring sensor to the sensor. This step solves the problem of fuzzy spatial relationship in leakage positioning and establishes the topological connection between sensors by quantifying the distance. The final effect is to achieve accurate identification of the neighboring relationship of potential leakage points, providing a key spatial basis for subsequent positioning and analysis. By setting the water pressure gradient factor algorithm and calculating the water pressure gradient factor, the problem of difficulty in capturing the abnormal propagation characteristics of water pressure is solved, and the prediction effect of the water pressure change trend of the leakage point is achieved. Pipeline leakage is usually accompanied by abnormal changes in water pressure, but a single water pressure value is difficult to reflect its dynamic propagation characteristics. By setting the water pressure gradient factor algorithm and calculating the water pressure gradient factor of the i-th sensor, the system can analyze the changing trend of water pressure in space and time. This method overcomes the limitation of traditional detection that only focuses on static water pressure values ​​and ignores gradient changes, so that the propagation law of pressure anomalies caused by leakage can be clearly presented. The final effect is to predict the trend of water pressure changes at the leakage point, providing dynamic data support for the intelligent model to determine the leakage location. By setting the flow gradient factor algorithm and calculating the flow gradient factor, the problem of insufficient analysis of the correlation between flow anomalies and leakage is solved, and the accurate capture effect of flow changes at the leakage point is achieved. Flow anomaly is an important representation of leakage, but traditional methods often cannot effectively associate flow changes with leakage locations. By setting the flow gradient factor algorithm and calculating the flow gradient factor of the i-th sensor, the system can quantify the distribution and change characteristics of flow in the pipeline network. This step solves the defect of insufficient analysis of the correlation between flow anomalies and leakage points, and captures the significant fluctuations of flow caused by leakage through the calculation of gradient factors. The final effect is to accurately capture the flow changes at the leakage point and enhance the ability of the intelligent prediction model to identify the leakage location. By combining Euclidean distance, water pressure gradient and flow gradient to develop an intelligent prediction model, the problem of leakage positioning relying on manual experience is solved, and the data-driven automatic positioning effect is achieved. Traditional leakage location mostly relies on manual investigation, which is inefficient and prone to errors. By integrating the data of Euclidean distance, water pressure gradient factor and flow gradient factor, an intelligent prediction model is developed. The system can automatically analyze the possibility of leakage and its location based on multi-dimensional parameters. This method overcomes the problem of strong subjectivity and low efficiency of manual experience judgment, and realizes the automation of the whole process from data collection to leakage location. The final effect is to significantly improve the accuracy and efficiency of leakage location, providing an intelligent solution for the safe maintenance of pipeline systems.

[0017] 5. By setting the leakage location algorithm and calculating the coordinates of the leakage point, the problem of the difficulty in accurately determining the leakage location is solved, and the spatial positioning effect of the leakage point is achieved. In the pipeline system, the exact location of the leakage point is often difficult to directly determine through a single sensor data, and traditional methods rely on rough estimation or manual investigation. By setting the leakage location algorithm, the coordinates of the predicted leakage point are calculated based on the abnormal sensor data, and the coordinates are optimized during the search process. The system can integrate multi-source data into a specific spatial location. This step solves the problem of fuzzy leakage location and insufficient positioning accuracy, and achieves accurate spatial positioning of the leakage point through mathematical modeling and search optimization. The final effect is to improve the pertinence of leakage investigation and provide clear target coordinates for subsequent maintenance. The confidence of leakage prediction is calculated through the confidence algorithm, which solves the problem of difficulty in evaluating the reliability of leakage location results and achieves the quantitative effect of prediction credibility. If the leakage location lacks reliability assessment, managers may question the prediction results, resulting in decision delays. Through the confidence algorithm, the confidence is calculated, and combined with parameters such as the total number of abnormal sensors, the system can provide a quantitative credibility indicator for each leakage location. This method overcomes the defect of traditional positioning results that lacks verification basis, so that the reliability of the prediction can be clearly presented. The final effect is to achieve scientific quantification of prediction credibility, enhance managers' trust in positioning results, and improve decision-making efficiency. By storing the leakage positioning results in the leakage prediction data table, the problem of scattered leakage information that is difficult to manage is solved, and the effect of data centralization and historical traceability is achieved. In the pipeline system, if the leakage prediction data is stored in a scattered manner, it may lead to information loss or difficulty in systematic analysis. By storing the coordinates and confidence of the predicted leakage point in the leakage prediction data table, the system can record and save the results of each positioning in a centralized manner. This step solves the problem of chaotic management and difficulty in tracing leakage information, and provides complete data support for subsequent analysis and optimization. The final effect is to achieve the centralization and historical traceability of leakage data, and improve the systematic and long-term management capabilities of pipeline maintenance. By integrating positioning algorithms and confidence calculations to develop intelligent prediction models, the inefficiency of leakage investigation relying on manual experience is solved, and intelligent leakage prediction and positioning effects are achieved. Traditional leakage investigation relies more on manual judgment and on-site inspection, which is inefficient and error-prone. By combining the leakage location algorithm and the confidence algorithm to develop an intelligent prediction model, the system can automatically analyze abnormal data and output the leakage location and credibility. This method overcomes the limitations of manual investigation, which is time-consuming and highly subjective, and realizes the intelligence of the entire process from data analysis to location prediction.

[0018] 6. By generating an alarm and pushing it to the maintenance team, the problem of delayed response to leakage is solved, and the effect of rapid notification and timely processing is achieved. If the relevant personnel are not notified of pipeline leakage in time, it may lead to waste of water resources and expansion of safety accidents. By generating an alarm when the conditions are met and pushing it to the maintenance team through the system, the system can pass the leakage information to the responsible person as soon as possible. This step solves the problem of slow response of traditional manual discovery and notification methods, ensuring that the maintenance team is quickly informed of the situation. The final effect is to achieve rapid notification and timely processing, shorten the leakage response time, and reduce potential losses. By calculating the distance from the maintenance worker to the leakage point and selecting the nearest maintenance worker, the problem of inefficient maintenance resource allocation is solved, and the effect of optimizing task allocation is achieved. In maintenance scheduling, if the location of the maintenance worker is not considered, it may lead to unreasonable task allocation and extend the maintenance time. By obtaining the real-time location of the maintenance worker, calculating the distance from the leakage point, and selecting the maintenance worker with the smallest distance, the system can optimize task allocation according to spatial location. This method overcomes the defects of low efficiency of random assignment or experience allocation in traditional scheduling. The final effect is to achieve the optimal allocation of maintenance tasks, improve maintenance efficiency, and speed up the repair of leakage. By querying the position of the smart valve and sending a water pressure adjustment command, the problem of water flow control when leakage occurs is solved, and the effect of emergency control and resource protection is achieved. When leakage occurs, if the water flow is not controlled in time, it will aggravate the waste of water resources and pipeline damage. By querying the position of the smart valve close to the leakage point, calculating its distance to the leakage point, selecting the nearest valve and sending a command to reduce the water pressure to 40%, the system can quickly reduce the water pressure at the leakage point. This step solves the urgent problem of water flow control when leakage occurs, and reduces water flow loss through automatic control. The final effect is to achieve emergency control and resource protection, and reduce the economic and environmental impact caused by leakage. By forming a maintenance task data set and storing it in a data table, the problem of scattered maintenance information being difficult to trace is solved, and the systematic management effect of the maintenance process is achieved. If the maintenance task lacks a unified record, it may lead to information loss or difficulty in subsequent analysis. By obtaining the material list and the maintenance worker number, forming a maintenance task data set, and storing it in the data table, the system centrally manages the task-related information. This method solves the defects of scattered maintenance information and difficulty in tracking, and provides a complete record for the maintenance process. The final effect is to achieve systematization and traceability of the maintenance process, facilitate subsequent evaluation and optimization, and improve the management level of community management.

[0019] 7. The problem of insufficient positioning accuracy of the prediction model is solved by the maintenance personnel submitting the actual leakage point and cause and calculating the prediction error, and the self-verification effect of the model accuracy is achieved. If the leakage positioning of the intelligent prediction model deviates greatly from the actual situation, it may lead to a waste of maintenance resources. By requiring the maintenance personnel to submit the coordinates of the actual leakage point and the leakage cause, and calculate the prediction error, the system can quantify the deviation between the predicted leakage point and the actual position. This step solves the problem that the positioning accuracy of the model is difficult to verify, and the reliability of the prediction result is clarified through error calculation. The final effect is to achieve self-verification of the model accuracy, provide data basis for subsequent optimization, and improve the accuracy of positioning. By adjusting the weight according to the prediction error, the problem that the model parameter fixation cannot adapt to the actual situation is solved, and the adaptive optimization effect of the prediction model is achieved. If the prediction model uses fixed weights, it may not be able to cope with the dynamic changes in the operating environment of the pipeline system. By setting the calculation method of the water pressure weight, flow weight and temperature weight to adjust if the prediction error exceeds 5 meters, the system can dynamically optimize the parameters according to the actual error. This method overcomes the defect of increased prediction deviation caused by parameter fixation, making the model more in line with the actual situation. The final effect is to achieve adaptive optimization of the prediction model, improve its adaptability to complex scenarios and prediction accuracy. By calculating the false alarm rate of the last 100 alarms and adjusting the threshold, the problem of frequent false alarms affecting the credibility of the system is solved, and the reliability of the alarm system is improved. Frequent false alarms may cause the maintenance team to lose trust in the alarm and reduce the response efficiency. By calculating the false alarm rate, that is, the proportion of leaks not found in the last 100 alarms, and adjusting the abnormal index threshold when it exceeds 0.2, the system can optimize the alarm triggering conditions according to the actual false alarm situation. This step solves the problem of high false alarm rate leading to reduced system credibility, and reduces invalid alarms through dynamic threshold adjustment. The final effect is to improve the reliability of the alarm system and enhance the team's trust in the system and response enthusiasm. By storing the optimized parameters in the data table, the problem of the difficulty of continuous application of the optimization results is solved, and the effect of long-term improvement of the system is achieved. If the system optimization results are not saved, repeated adjustments may result in the inability to accumulate improvement results. By storing the adjusted weights and thresholds in the data table, the system can record the parameters optimized each time and directly apply them in subsequent operations. This method solves the problem of scattered optimization results and difficulty in continuous use, and provides data support for the continuous improvement of the system. The net effect is to achieve the system’s long-term improvement capabilities, ensuring that hazard detection and prediction models are continuously improved over time, thereby improving the overall safety management level of community management. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Example, see Figure 1 , a smart community management method based on big data, comprising: S1. Deployment of sensors and data collection: Install three-in-one sensors every 50 meters in the water supply pipe and every 100 meters in the drainage pipe; The three-in-one sensor includes: water pressure sensor , Flow Sensor and temperature sensor ; Assign a unique number to each 3-in-1 sensor ,in ; Get the total length of the water supply pipeline, recorded as ; Get the total length of the drainage pipe, recorded as ; Calculate the total number of three-in-one sensors, recorded as : ; in, To round up; Get the area covered by the community on the ground, recorded as the community surface; Get the midpoint of the community surface, take the midpoint of the community surface as the vertical intersection point, make two mutually perpendicular rays on the community surface, and use one of the rays as axis coordinates, and the other ray as Axis coordinates, a coordinate system with the midpoint of the community surface as the origin and two mutually perpendicular rays as coordinate axes is obtained; Data is collected every 4 minutes. The collected data includes: , three-in-one sensor number , water pressure value , flow value , temperature value , three-in-one sensor Axis coordinates , three-in-one sensor Axis coordinates , and form data points , ; Among them, the subscript Indicates 3-in-1 sensor; Run the system for 48 hours and calculate the baseline value for each 3-in-1 sensor: ; ; ; in, For the The reference water pressure of a three-in-one sensor; For the Base flow of a three-in-one sensor; For the The reference temperature of a three-in-one sensor; For the Three-in-one sensor The water pressure collected at the time; For the Three-in-one sensor The amount of traffic collected; For the Three-in-one sensor The temperature of the first collection; Will , , and Composition of benchmark data set ; get ; S2, analyze the collected data in real time and detect anomalies; S3, intelligent prediction model development and leakage location; S4, automated response and maintenance scheduling; S5. Data feedback and system optimization.

[0023] By deploying three-in-one sensors every 50 meters in the water supply pipeline and every 100 meters in the drainage pipeline, the problem of blind spots in pipeline operation status monitoring is solved, and the effect of full coverage and accurate perception is achieved. In community management, the safe operation of water supply and drainage pipelines is crucial, but traditional methods often make it difficult to achieve real-time monitoring of the entire pipeline section. By installing three-in-one sensors containing water pressure, flow and temperature every 50 meters in the water supply pipeline and every 100 meters in the drainage pipeline, the system can fully cover the pipeline network and avoid the occurrence of monitoring blind spots. This high-density deployment combined with the multi-parameter acquisition function of the three-in-one sensor enables subtle changes in the pipeline operation status to be captured, providing a rich data foundation for subsequent analysis, thereby significantly improving the coverage and accuracy of hidden danger detection. By assigning a unique number to each three-in-one sensor and establishing a coordinate system, the problems of confusing data sources and unclear spatial positioning are solved, and the effect of data traceability and spatial analysis is achieved. In large-scale pipeline networks, confusion in data sources and lack of location information may lead to inefficient hidden danger detection. By assigning a unique number to each sensor and establishing a coordinate system with the midpoint of the community surface as the origin, the sensor position is accurately determined to the X-axis and Y-axis coordinates, and the system can clearly track the source of each set of data and locate its spatial position. This method not only solves the problem of multi-sensor data confusion, but also lays the foundation for the spatial distribution analysis of hidden dangers, allowing managers to quickly lock in problem areas and improve the pertinence of investigation and response. By collecting data every 4 minutes and running for 48 hours to calculate the benchmark value, the problem of insufficient basis for judging the abnormal operation status of the pipeline is solved, and the basic effect of dynamic benchmark establishment and abnormal detection is achieved. The abnormal judgment of the pipeline operation status requires a scientific benchmark as a reference, while traditional methods often rely on static standards and are difficult to adapt to actual changes. By collecting data containing time, number, water pressure, flow, temperature and coordinates every 4 minutes, and calculating the benchmark water pressure, benchmark flow and benchmark temperature of each sensor within 48 hours, the system can generate dynamic benchmark values ​​based on actual operation data. This method overcomes the defect that a single standard cannot reflect the individual differences of pipelines, and provides a reliable personalized reference standard for subsequent abnormal detection, thereby enhancing the accuracy and adaptability of the early warning system. By forming a benchmark data set and analyzing it, the problem of lack of systematic basis for hidden danger detection is solved, and data-driven safety management effect is achieved. If hidden danger detection relies solely on manual experience, omissions or misjudgments are prone to occur. By forming a benchmark data set with the benchmark water pressure, benchmark flow and benchmark temperature of each sensor, the system provides a comprehensive digital description of the safety status of the pipeline. This data-driven approach makes hidden danger detection no longer rely on subjective judgment, but is based on objective benchmark value analysis, which can systematically identify potential risks. The ultimate effect is to improve the scientificity and reliability of safety management.

[0024] The real-time analysis of collected data and detection of anomalies specifically includes: Set the water pressure deviation rate algorithm to calculate the The water pressure deviation rate after temperature correction of the three-in-one sensor is expressed as : ; Set the flow deviation rate algorithm to calculate the The flow deviation rate after temperature correction of the three-in-one sensor is expressed as : ; Set the temperature deviation rate algorithm to calculate the The temperature deviation rate of the three-in-one sensor is expressed as : ; Calculate the water pressure weight, denoted as : ; like ,but ; Calculate the traffic weight, recorded as : ; like ,but ; Calculate the temperature weight, denoted as : ; like ,but .

[0025] By setting the water pressure, flow and temperature deviation rate algorithms and performing temperature correction, the problem of inaccurate judgment of pipeline operation parameter anomalies is solved, and a more accurate anomaly detection effect is achieved. In pipeline monitoring, water pressure, flow and temperature anomalies may be important signals of safety hazards, but the deviation of a single parameter may be interfered by environmental factors, such as temperature fluctuations, leading to misjudgment. By setting the water pressure deviation rate algorithm, flow deviation rate algorithm and temperature deviation rate algorithm for the i-th three-in-one sensor, and introducing temperature correction in the calculation of water pressure and flow deviation rates, the system can eliminate the interference of temperature changes on water pressure and flow. This method solves the problem of inaccurate anomaly judgment caused by confusion of environmental variables in traditional detection, thereby significantly improving the accuracy and reliability of anomaly detection and providing a more scientific basis for hidden danger investigation. By introducing weight calculation and setting conditional judgment, the problem of single anomaly severity assessment is solved, and a dynamic priority management effect is achieved. In pipeline operation, the importance of anomalies of different parameters may vary with specific circumstances. Traditional methods often use fixed thresholds, which are difficult to respond flexibly. By calculating the water pressure weight, flow weight and temperature weight, and setting conditional judgment, the system can eliminate the interference of temperature changes on water pressure and flow. ,but , the system dynamically adjusts the weight of each parameter according to the size of the deviation rate. This step solves the problem that the abnormal evaluation is too single and cannot distinguish between the important and the urgent, so that the system can give priority to the parameters with large deviations, and then realize dynamic priority management. The final effect is to improve the efficiency of resource allocation and ensure that serious hidden dangers are dealt with in a timely manner. By analyzing the collected data in real time and combining the deviation rate and weight, the problem of delayed hidden danger investigation is solved, and the effect of real-time warning and rapid response is achieved. In community management, if the hidden danger investigation fails to detect abnormalities in time, accidents may occur. By analyzing the collected data in real time, calculating the water pressure deviation rate, flow deviation rate and temperature deviation rate of each sensor, and combining the weight for comprehensive evaluation, the system can identify and quantify the risk at the first time when the abnormality occurs. This real-time processing method overcomes the defect of strong lag of traditional regular inspections, making the hidden danger investigation from passive to active. The final effect is to realize the real-time warning function, which buys valuable time for managers to take countermeasures quickly and reduces safety risks. Through the algorithm design of comprehensive deviation rate and weight, the problem of insufficient correlation analysis of multi-parameter abnormalities is solved, and the systematic safety status evaluation effect is achieved. During pipeline operation, water pressure, flow rate and temperature are often interrelated, and analyzing a single parameter alone may ignore the overall risk. By combining the deviation rates of water pressure, flow rate and temperature with the corresponding weights, the system can comprehensively evaluate the degree of abnormality of multiple parameters and their mutual influence. This method solves the limitation of insufficient multi-parameter correlation analysis in traditional detection, making the safety status assessment more systematic and comprehensive.

[0026] The real-time analysis of collected data and detection of anomalies specifically includes: Set up a real-time anomaly detection algorithm to calculate the The abnormality index of the three-in-one sensor is used to measure the degree to which the pipeline operation status deviates from the normal value. The larger the value, the higher the possibility of abnormality: ; in, For the Abnormal index of three-in-one sensors; Set the anomaly index threshold, denoted as ; like , then the The data collected by the three-in-one sensor is considered abnormal and Stored in the alarm data table; Set the abnormal sensor set, denoted as , ; in, For the The number of the abnormal sensor; is the total number of abnormal sensors.

[0027] By setting a real-time anomaly detection algorithm and calculating the anomaly index, the problem of pipeline operation status anomaly being difficult to quantify is solved, and the effect of accurate assessment of the degree of anomaly is achieved. In pipeline safety monitoring, the anomaly of the operation status is often difficult to judge intuitively. Traditional methods rely on experience or simple thresholds and lack scientific basis. By setting a real-time anomaly detection algorithm for the i-th three-in-one sensor and calculating the anomaly index, the system can integrate the deviation degree of multiple parameters such as water pressure, flow and temperature into a quantitative indicator. This step solves the problem that the judgment of anomalies is highly subjective and difficult to quantify, so that the degree of anomaly of the pipeline operation status can be clearly expressed by numerical values. The final effect is to achieve the precision and standardization of anomaly evaluation, providing a reliable basis for subsequent decision-making. By setting the anomaly index threshold and filtering the abnormal data, the problem of low efficiency of anomaly identification is solved, and the effect of quickly locking hidden dangers is achieved. In the real-time analysis of a large amount of sensor data, it will take a lot of time and resources to check the anomalies one by one. By setting the anomaly index threshold and stipulating that when the threshold is exceeded, the data of the i-th sensor is identified as abnormal and stored in the alarm data table, the system can automatically filter normal data and quickly focus on abnormal points. This method overcomes the defect of low efficiency of traditional full-volume analysis and greatly reduces the interference of invalid data. The final effect is to quickly lock the hidden dangers, so that managers can quickly locate the problem sensors and improve the efficiency of troubleshooting. By establishing an abnormal sensor set and recording the number, the problem of unclear traceability of abnormal data is solved, and the traceability of abnormal locations is achieved. In a complex pipeline network, if the source cannot be accurately traced after the abnormality occurs, it will increase the difficulty of handling. By setting up an abnormal sensor set and recording the number of each abnormal sensor and its total number, the system can clearly mark the specific location and scope of the abnormality. This step solves the problem of confusing sources and difficult positioning of abnormal data, and provides clear guidance for subsequent maintenance or emergency response. The final effect is to achieve traceability of abnormal locations and efficient management, ensuring that hidden dangers are handled in a targeted manner. By analyzing and storing alarm data in real time, the problem of delayed warning of hidden dangers is solved, and the effect of real-time monitoring and timely response is achieved. During pipeline operation, if hidden dangers cannot be discovered and warned in time, accidents may expand. By analyzing the collected data in real time, calculating the abnormal index, and storing it in the alarm data table immediately when the threshold is exceeded, the system can issue an alarm as soon as the abnormality occurs. This real-time processing method overcomes the problem of delayed traditional regular inspections, making hidden danger troubleshooting from passive to active. The end result is the ability to monitor in real time and respond promptly, significantly reducing security risks.

[0028] The intelligent prediction model development and leakage location specifically include: For each , calculate the The minimum value of the Euclidean distance between a three-in-one sensor and all other three-in-one sensors is determined : ; Set the water pressure gradient factor algorithm to calculate the The water pressure gradient factor of the three-in-one sensor is expressed as : ; Set the flow gradient factor algorithm to calculate the The flow gradient factor of a three-in-one sensor is expressed as : .

[0029] By calculating the Euclidean distance between each three-in-one sensor and other sensors and taking the minimum value, the problem of fuzzy spatial relationship in leakage positioning is solved, and the accurate identification effect of the neighboring relationship of potential leakage points is achieved. In the pipeline system, leakage is often closely related to the spatial relationship between sensors, but traditional methods are difficult to accurately determine the spatial correlation between abnormal points and the surrounding environment. By calculating the Euclidean distance between the i-th three-in-one sensor and all other sensors and taking the minimum value to determine, the system can identify the position of the nearest neighboring sensor to the sensor. This step solves the problem of fuzzy spatial relationship in leakage positioning and establishes the topological connection between sensors by quantifying the distance. The final effect is to achieve accurate identification of the neighboring relationship of potential leakage points, providing a key spatial basis for subsequent positioning and analysis. By setting the water pressure gradient factor algorithm and calculating the water pressure gradient factor, the problem of difficulty in capturing the abnormal propagation characteristics of water pressure is solved, and the prediction effect of the water pressure change trend of the leakage point is achieved. Pipeline leakage is usually accompanied by abnormal changes in water pressure, but a single water pressure value is difficult to reflect its dynamic propagation characteristics. By setting the water pressure gradient factor algorithm and calculating the water pressure gradient factor of the i-th sensor, the system can analyze the changing trend of water pressure in space and time. This method overcomes the limitation of traditional detection that only focuses on static water pressure values ​​and ignores gradient changes, so that the propagation law of pressure anomalies caused by leakage can be clearly presented. The final effect is to predict the trend of water pressure changes at the leakage point, providing dynamic data support for the intelligent model to determine the leakage location. By setting the flow gradient factor algorithm and calculating the flow gradient factor, the problem of insufficient analysis of the correlation between flow anomalies and leakage is solved, and the accurate capture effect of flow changes at the leakage point is achieved. Flow anomaly is an important representation of leakage, but traditional methods often cannot effectively associate flow changes with leakage locations. By setting the flow gradient factor algorithm and calculating the flow gradient factor of the i-th sensor, the system can quantify the distribution and change characteristics of flow in the pipeline network. This step solves the defect of insufficient analysis of the correlation between flow anomalies and leakage points, and captures the significant fluctuations of flow caused by leakage through the calculation of gradient factors. The final effect is to accurately capture the flow changes at the leakage point and enhance the ability of the intelligent prediction model to identify the leakage location. By combining Euclidean distance, water pressure gradient and flow gradient to develop an intelligent prediction model, the problem of leakage positioning relying on manual experience is solved, and the data-driven automatic positioning effect is achieved. Traditional leakage location mostly relies on manual investigation, which is inefficient and prone to errors. By integrating the data of Euclidean distance, water pressure gradient factor and flow gradient factor, an intelligent prediction model is developed. The system can automatically analyze the possibility of leakage and its location based on multi-dimensional parameters. This method overcomes the problem of strong subjectivity and low efficiency of manual experience judgment, and realizes the automation of the whole process from data collection to leakage location. The final effect is to significantly improve the accuracy and efficiency of leakage location, providing an intelligent solution for the safe maintenance of pipeline systems.

[0030] The intelligent prediction model development and leakage location specifically include: Set the leakage location algorithm and locate the leakage point coordinates : ; in, To predict the leakage point coordinate; To predict the leakage point coordinate; For the search process coordinate; For the search process coordinate; The confidence is calculated by the confidence algorithm, which is recorded as : ; in, is the total number of abnormal sensors; Will Stored in the leakage prediction data table.

[0031] By setting the leakage location algorithm and calculating the coordinates of the leakage point, the problem of the difficulty in accurately determining the leakage location is solved, and the spatial positioning effect of the leakage point is achieved. In the pipeline system, the exact location of the leakage point is often difficult to directly determine through a single sensor data, and traditional methods rely on rough estimation or manual investigation. By setting the leakage location algorithm, the coordinates of the predicted leakage point are calculated based on the abnormal sensor data, and the coordinates are optimized during the search process. The system can integrate multi-source data into a specific spatial location. This step solves the problem of fuzzy leakage location and insufficient positioning accuracy, and achieves accurate spatial positioning of the leakage point through mathematical modeling and search optimization. The final effect is to improve the pertinence of leakage investigation and provide clear target coordinates for subsequent maintenance. The confidence of leakage prediction is calculated through the confidence algorithm, which solves the problem of difficulty in evaluating the reliability of leakage location results and achieves the quantitative effect of prediction credibility. If the leakage location lacks reliability assessment, managers may question the prediction results, resulting in decision delays. Through the confidence algorithm, the confidence is calculated, and combined with parameters such as the total number of abnormal sensors, the system can provide a quantitative credibility indicator for each leakage location. This method overcomes the defect of traditional positioning results that lacks verification basis, so that the reliability of the prediction can be clearly presented. The final effect is to achieve scientific quantification of prediction credibility, enhance managers' trust in positioning results, and improve decision-making efficiency. By storing the leakage positioning results in the leakage prediction data table, the problem of scattered leakage information that is difficult to manage is solved, and the effect of data centralization and historical traceability is achieved. In the pipeline system, if the leakage prediction data is stored in a scattered manner, it may lead to information loss or difficulty in systematic analysis. By storing the coordinates and confidence of the predicted leakage point in the leakage prediction data table, the system can record and save the results of each positioning in a centralized manner. This step solves the problem of chaotic management and difficulty in tracing leakage information, and provides complete data support for subsequent analysis and optimization. The final effect is to achieve the centralization and historical traceability of leakage data, and improve the systematic and long-term management capabilities of pipeline maintenance. By integrating positioning algorithms and confidence calculations to develop intelligent prediction models, the inefficiency of leakage investigation relying on manual experience is solved, and intelligent leakage prediction and positioning effects are achieved. Traditional leakage investigation relies more on manual judgment and on-site inspection, which is inefficient and error-prone. By combining the leakage location algorithm and the confidence algorithm to develop an intelligent prediction model, the system can automatically analyze abnormal data and output the leakage location and credibility. This method overcomes the limitations of manual investigation, which is time-consuming and highly subjective, and realizes the intelligence of the entire process from data analysis to location prediction.

[0032] The automated response and maintenance scheduling specifically includes: like , generating an alert; The alert format is: , pushed to the maintenance team through the system; Get the real-time location of the maintenance personnel; Calculate the distance from the maintenance worker to the leakage point, recorded as : ; in, For maintenance personnel coordinate; For maintenance personnel coordinate; Select The youngest maintenance worker and assigns the maintenance task to this maintenance worker; Check the location of the smart valve near the leakage point; Calculate the distance from the smart valve to the leakage point, recorded as : ; choose The smallest smart valve is instructed to reduce water pressure to 40%; Get the list of materials used in the repair, recorded as ; Get the maintenance person's number, recorded as ; The maintenance tasks are grouped into a maintenance task data set, which is specifically: ; Store the maintenance task dataset in a data table middle.

[0033] By generating an alarm and pushing it to the maintenance team, the problem of delayed response to leakage is solved, and the effect of rapid notification and timely processing is achieved. If the relevant personnel are not notified of pipeline leakage in time, it may lead to waste of water resources and expansion of safety accidents. By generating an alarm when the conditions are met and pushing it to the maintenance team through the system, the system can pass the leakage information to the responsible person as soon as possible. This step solves the problem of slow response of traditional manual discovery and notification methods, ensuring that the maintenance team is quickly informed of the situation. The final effect is to achieve rapid notification and timely processing, shorten the leakage response time, and reduce potential losses. By calculating the distance from the maintenance worker to the leakage point and selecting the nearest maintenance worker, the problem of inefficient maintenance resource allocation is solved, and the effect of optimizing task allocation is achieved. In maintenance scheduling, if the location of the maintenance worker is not considered, it may lead to unreasonable task allocation and extend the maintenance time. By obtaining the real-time location of the maintenance worker, calculating the distance from the leakage point, and selecting the maintenance worker with the smallest distance, the system can optimize task allocation according to spatial location. This method overcomes the defects of low efficiency of random assignment or experience allocation in traditional scheduling. The final effect is to achieve the optimal allocation of maintenance tasks, improve maintenance efficiency, and speed up the leakage repair speed. By querying the position of the smart valve and sending a water pressure adjustment command, the problem of water flow control when leakage occurs is solved, and the effect of emergency control and resource protection is achieved. When leakage occurs, if the water flow is not controlled in time, it will aggravate the waste of water resources and pipeline damage. By querying the position of the smart valve close to the leakage point, calculating its distance to the leakage point, selecting the nearest valve and sending a command to reduce the water pressure to 40%, the system can quickly reduce the water pressure at the leakage point. This step solves the urgent problem of water flow control when leakage occurs, and reduces water flow loss through automatic control. The final effect is to achieve emergency control and resource protection, and reduce the economic and environmental impact caused by leakage. By forming a maintenance task data set and storing it in a data table, the problem of scattered maintenance information being difficult to trace is solved, and the systematic management effect of the maintenance process is achieved. If the maintenance task lacks a unified record, it may lead to information loss or difficulty in subsequent analysis. By obtaining the material list and the maintenance worker number, forming a maintenance task data set, and storing it in the data table, the system centrally manages the task-related information. This method solves the defects of scattered maintenance information and difficulty in tracking, and provides a complete record for the maintenance process. The final effect is to achieve systematization and traceability of the maintenance process, facilitate subsequent evaluation and optimization, and improve the management level of community management.

[0034] The data feedback and system optimization specifically include: Maintenance personnel submit actual leakage points and leakage causes; Calculate the prediction error, denoted as : ; in, The actual leakage coordinate; The actual leakage coordinate; like m, adjust the weights: ; ; ; Calculate the false alarm rate, the proportion of the last 100 alarms that did not find a leak, denoted as : ; in, The number of false alarms in the last 100 alarms; like ,but ; The optimized parameters Store to data table middle.

[0035] By requiring the maintenance personnel to submit the actual leakage point and cause and calculate the prediction error, the problem of insufficient positioning accuracy of the prediction model is solved, and the self-verification effect of the model accuracy is achieved. If the leakage positioning of the intelligent prediction model deviates greatly from the actual situation, it may lead to a waste of maintenance resources. By requiring the maintenance personnel to submit the coordinates of the actual leakage point and the leakage cause, and calculate the prediction error, the system can quantify the deviation between the predicted leakage point and the actual position. This step solves the problem that the positioning accuracy of the model is difficult to verify, and the reliability of the prediction result is clarified through error calculation. The final effect is to achieve self-verification of the model accuracy, provide data basis for subsequent optimization, and improve the accuracy of positioning. By adjusting the weight according to the prediction error, the problem that the model parameter fixation cannot adapt to the actual situation is solved, and the adaptive optimization effect of the prediction model is achieved. If the prediction model uses fixed weights, it may not be able to cope with the dynamic changes in the operating environment of the pipeline system. By setting the calculation method of the water pressure weight, flow weight and temperature weight to adjust if the prediction error exceeds 5 meters, the system can dynamically optimize the parameters according to the actual error. This method overcomes the defect of increased prediction deviation caused by parameter fixation, making the model more in line with the actual situation. The final effect is to achieve adaptive optimization of the prediction model, improving its adaptability to complex scenarios and prediction accuracy. By calculating the false alarm rate of the last 100 alarms and adjusting the threshold, the problem of frequent false alarms affecting the credibility of the system is solved, and the reliability of the alarm system is improved. Frequent false alarms may cause the maintenance team to lose trust in the alarm and reduce the response efficiency. By calculating the false alarm rate, that is, the proportion of leaks not found in the last 100 alarms, and adjusting the abnormal index threshold when it exceeds 0.2, the system can optimize the alarm triggering conditions according to the actual false alarm situation. This step solves the problem of low system credibility caused by high false alarm rate, and reduces invalid alarms through dynamic threshold adjustment. The final effect is to improve the reliability of the alarm system and enhance the team's trust in the system and response enthusiasm. By storing the optimized parameters in the data table, the problem of the difficulty of continuous application of the optimization results is solved, and the effect of long-term improvement of the system is achieved. If the system optimization results are not saved, repeated adjustments may fail to accumulate improvement results. By storing the adjusted weights and thresholds in the data table, the system can record the parameters optimized each time and directly apply them in subsequent operations. This method solves the problem of scattered optimization results and difficulty in continuous use, and provides data support for the continuous improvement of the system. The net effect is to achieve the system’s long-term improvement capabilities, ensuring that hazard detection and prediction models are continuously improved over time, thereby improving the overall safety management level of community management.

[0036] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0037] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A smart community management method based on big data, characterized in that: include: S1. Deployment of sensors and data collection: Install three-in-one sensors every 50 meters in the water supply pipe and every 100 meters in the drainage pipe; The three-in-one sensor includes: water pressure sensor , Flow Sensor and temperature sensor ; Assign a unique number to each 3-in-1 sensor ,in ; Get the total length of the water supply pipeline, recorded as ; Get the total length of the drainage pipe, recorded as ; Calculate the total number of three-in-one sensors, recorded as : ； in, To round up; Get the area covered by the community on the ground, recorded as the community surface; Get the midpoint of the community surface, take the midpoint of the community surface as the vertical intersection point, make two mutually perpendicular rays on the community surface, and use one of the rays as axis coordinates, and the other ray as Axis coordinates, a coordinate system with the midpoint of the community surface as the origin and two mutually perpendicular rays as coordinate axes is obtained; Data is collected every 4 minutes. The collected data includes: , three-in-one sensor number , water pressure value , flow value , temperature value , three-in-one sensor Axis coordinates , three-in-one sensor Axis coordinates , and form data points , ; Among them, the subscript Indicates 3-in-1 sensor; Run the system for 48 hours and calculate the baseline value for each 3-in-1 sensor: ； ； ； in, For the The reference water pressure of a three-in-one sensor; For the Base flow of a three-in-one sensor; For the The reference temperature of a three-in-one sensor; For the Three-in-one sensor The water pressure collected at the time; For the Three-in-one sensor The amount of traffic collected; For the Three-in-one sensor The temperature of the first collection; Will , , and Composition of benchmark data set ; get ; S2, analyze the collected data in real time and detect anomalies; S3, intelligent prediction model development and leakage location; S4, automated response and maintenance scheduling; S5. Data feedback and system optimization.

2. According to the big data-based smart community management method of claim 1, it is characterized in that: The real-time analysis of collected data and detection of anomalies specifically includes: Set the water pressure deviation rate algorithm to calculate the The water pressure deviation rate after temperature correction of the three-in-one sensor is expressed as : ； Set the flow deviation rate algorithm to calculate the The flow deviation rate after temperature correction of the three-in-one sensor is recorded as : ； Set the temperature deviation rate algorithm to calculate the The temperature deviation rate of the three-in-one sensor is expressed as : ； Calculate the water pressure weight, denoted as : ； like ,but ; Calculate the traffic weight, recorded as : ； like ,but ; Calculate the temperature weight, denoted as : ； like ,but .

3. According to the big data-based smart community management method of claim 2, it is characterized in that: The real-time analysis of collected data and detection of anomalies specifically includes: Set up a real-time anomaly detection algorithm to calculate the Abnormal index of three-in-one sensors: ； in, For the Abnormal index of three-in-one sensors; Set the anomaly index threshold, denoted as ; like , then the The data collected by the three-in-one sensor is considered abnormal and Stored in the alarm data table; Set the abnormal sensor set, denoted as , ; in, For the The number of the abnormal sensor; is the total number of abnormal sensors.

4. According to the big data-based smart community management method of claim 3, it is characterized in that: The intelligent prediction model development and leakage location specifically include: For each , calculate the The minimum value of the Euclidean distance between a three-in-one sensor and all other three-in-one sensors is determined : ； Set the water pressure gradient factor algorithm to calculate the The water pressure gradient factor of the three-in-one sensor is expressed as : ； Set the flow gradient factor algorithm to calculate the The flow gradient factor of a three-in-one sensor is expressed as : 。 5. The method for managing a smart community based on big data according to claim 4 is characterized in that: The intelligent prediction model development and leakage location specifically include: Set the leakage location algorithm and locate the leakage point coordinates : ； in, To predict the leakage point coordinate; To predict the leakage point coordinate; For the search process coordinate; For the search process coordinate; The confidence is calculated by the confidence algorithm, which is recorded as : ； in, is the total number of abnormal sensors; Will Stored in the leakage prediction data table.

6. The smart community management method based on big data according to claim 5 is characterized in that: The automated response and maintenance scheduling specifically includes: like , generating an alert; The alert format is: , pushed to the maintenance team through the system; Get the real-time location of the maintenance personnel; Calculate the distance from the maintenance worker to the leakage point, recorded as : ； in, For maintenance personnel coordinate; For maintenance personnel coordinate; Select The youngest maintenance worker and assigns the maintenance task to this maintenance worker; Check the location of the smart valve near the leakage point; Calculate the distance from the smart valve to the leakage point, recorded as : ； choose The smallest smart valve is instructed to reduce water pressure to 40%; Get the list of materials used in the repair, recorded as ; Get the maintenance person's number, recorded as ; The maintenance tasks are grouped into a maintenance task data set, which is specifically: ； Store the maintenance task dataset in a data table middle.

7. The method for managing a smart community based on big data according to claim 6 is characterized in that: The data feedback and system optimization specifically include: Maintenance personnel submit actual leakage points and leakage causes; Calculate the prediction error, denoted as : ； in, The actual leakage coordinate; The actual leakage coordinate; like m, adjust the weights: ； ； ； Calculate the false alarm rate, denoted as : ； in, The number of false alarms in the last 100 alarms; like ,but ; The optimized parameters Store to data table middle.

Images