Water-saving comprehensive service system and method based on big data

Abstract

The invention relates to the technical field of water-saving comprehensive service based on big data, and discloses a water-saving comprehensive service system and method based on big data. Collecting and standardizing water consumption data at continuous observation moments; adaptively determining a layering number and calculating a layering threshold value of each layer; performing hierarchical classification on all moment data according to a threshold value; calculating a sample number, an average value and a standard deviation of each layer of data; setting a self-adaptive peak clipping amount according to the maximum water consumption and the standard deviation; equivalently distributing the peak clipping amount to each moment and updating the water consumption data; and finally, through closed-loop verification and error correction, it is ensured that the expected reduction amount is consistent with the actual reduction amount. The data quality is improved through standardized processing, traditional interval division errors are avoided through self-adaptive layering, water consumption abnormity is recognized through statistical analysis, accurate peak clipping is achieved in combination with standard deviation, and the reliability of an execution result is guaranteed through a closed-loop mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of water-saving integrated services based on big data, and specifically to a water-saving integrated service system and method based on big data. Background Art

[0002] In recent years, with the acceleration of urbanization and the improvement of industrialization, water demand has increased significantly, while water resource supply capacity is relatively limited. Comprehensive water-saving services have become an important means of ensuring sustainable development. Traditional water-saving management relies heavily on manually set water quotas, tiered pricing, and timed quota control, which makes it difficult to meet the needs of refinement and dynamism. In existing water management systems, fixed thresholds or rules are usually used to control water consumption in segments, and the portion exceeding the threshold is uniformly reduced. However, the fixed threshold segmentation method cannot adaptively adjust the stratification interval according to the actual water consumption distribution, resulting in data at certain levels being too sparse or too dense, affecting the accuracy of statistical features. The unified reduction strategy often ignores the volatility and distribution differences of water consumption in different time periods, which can easily lead to excessive or insufficient peak shaving. In addition, there is a lack of subsequent verification and correction mechanisms after the one-time reduction, making it difficult to ensure closed-loop consistency throughout the entire process.

[0003] Existing big data-based water consumption analysis technologies primarily focus on identifying water consumption patterns and forecasting demand, estimating future water consumption trends through machine learning or time series models. However, they lack a systematic approach for implementing dynamic peak shaving and tiered management in real-time or near-real time. Some studies cluster water consumption data according to fixed time windows or fixed water consumption intervals and then apply the same peak shaving allocation strategy to each cluster. While this simplifies the computation to some extent, it ignores the statistical fluctuations within the clusters, resulting in peak shaving amounts often mismatching the actual fluctuations. Furthermore, these methods lack a comprehensive validation process to verify that the actual peak shaving effect matches the expected target, and lack a unified error allocation and correction strategy. This leads to a gradual increase in cumulative error, making it difficult to achieve strict closed-loop control. Regarding normalization, existing technologies often linearly normalize raw water consumption based on maximum and minimum values ​​or scale it by a fixed ratio, but do not use the normalization results for the dynamic calculation of adaptive tiered thresholds. Regarding the selection of the number of tiers, most solutions directly specify a preset number of tiers or determine it through empirical rules, lacking a mechanism to adaptively adjust the number of tiers based on data size. This makes it difficult to balance tiering accuracy with algorithmic complexity. In terms of calculating statistical characteristic quantities, there are few solutions that uniformly plan indicators such as sample size, mean value, and standard deviation within the layer and use them for subsequent adaptive peak-shaving strategies. They are based only on simple mean values ​​or maximum values, which easily ignores the discreteness information of the data distribution. In addition, in determining the amount of peak shaving, existing technologies either artificially set a fixed reduction ratio or make reductions based on the difference between a single peak value and a threshold value, making it difficult to simultaneously take into account the volatility of each layer and the differences between layers. What is more serious is that many methods lack a unified error closed-loop verification design. When it is found that the actual reduction amount does not meet the expected target, it is impossible to allocate errors according to reasonable rules and make quantitative corrections, which may ultimately lead to the overall water-saving effect deviating from the design target. The lack of a closed-loop consistency verification and error correction mechanism also causes the system to gradually deviate from the original design target during long-term operation, making it difficult to maintain a stable water reduction effect.

[0004] To this end, this project aims to propose a comprehensive water-saving service system and method based on big data. By collecting water consumption data, conducting statistical analysis, and implementing intelligent peak-shaving processing, this system dynamically adjusts and optimizes water consumption during different time periods. Based on continuously observed water consumption data, this method employs adaptive stratification and statistical feature modeling to implement peak shaving and error control, ultimately delivering a stable and reasonable water-saving solution. Its core concept is to utilize a data-driven approach to analyze user water usage behavior. Through data stratification and adaptive peak-shaving calculation, this system achieves peak reduction, reduces resource waste, and improves overall water efficiency. Summary of the Invention

[0005] The present invention provides a water-saving integrated service system and method based on big data, which promotes the solution of the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: a comprehensive water-saving service method based on big data, comprising: Carry out data collection and preprocessing, obtain water consumption data in the continuous observation time series, and perform standardization on it; Adaptively determine the number of data stratification layers based on the total number of observation moments; According to the index of each layer, calculate the corresponding stratification threshold of each layer; All observation moments are stratified and classified according to the preset stratification threshold; For the time sets within each layer, calculate statistical characteristics, including sample size, average water consumption and standard deviation; Determine the adaptive peak shaving amount for each layer based on the maximum water consumption and standard deviation within the layer; Distribute the peak shaving amount equally according to time and update the adjusted water consumption data; Perform closed-loop integrity check to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, perform error allocation and correction until peak shaving consistency is met.

[0007] Optionally, the data collection and preprocessing are performed to obtain water consumption data in a continuous observation time sequence and perform standardization processing on the data, specifically including: Collected in a continuous observation time sequence Water consumption data within ;in, For the Actual water consumption at the moment; is the total number of observation times; Calculate the minimum water consumption for the entire sequence: ;in, The minimum water consumption at all times; Calculate the maximum water consumption for the entire sequence: ;in, The maximum water consumption at all times.

[0008] Optionally, the adaptive determination of the number of data stratification layers based on the total number of observation moments specifically includes: According to the total time Calculate the number of layers: ;in, To round down.

[0009] Optionally, the step of calculating the layer threshold corresponding to each layer based on the index of each layer specifically includes: Index each layer , calculate the boundary threshold: ; in, is the current layer index; For the The boundary threshold of the layer; when hour, ;when hour, .

[0010] Optionally, the stratification classification of all observation moments according to a preset stratification threshold specifically includes: For each layer , build the moment index collection: ; in, For the The time index of the layer; like , then the subsequent calculation of this layer is skipped; among them, For collection Number of elements.

[0011] Optionally, for the time sets within each layer, statistical feature quantities are calculated, including sample size, average water consumption, and standard deviation, specifically including: For each Layer , perform the following steps: S1. Calculate the number of samples in the layer: ;in, For the Layer time number; S2. Calculate the average water consumption within the layer: ;in, For the Average water consumption per floor; S3. Calculate the standard deviation of water consumption within the layer: ;in, For the Standard deviation of water consumption per layer; S4. Calculate the global maximum standard deviation: ;in, is the maximum standard deviation among all strata.

[0012] Optionally, determining the adaptive peak shaving amount of each layer based on the maximum water consumption and standard deviation within the layer specifically includes: For each Layer , perform the following steps: S5. Calculate the peak value within the layer: ;in, For the Maximum water consumption per layer; S6. Calculate the total peak shaving amount within the layer: ;in, For the The total water consumption reduction planned for each level.

[0013] Optionally, distributing the peak shaving amount equally according to time and updating the adjusted water consumption data specifically includes: For each Layer , perform the following steps: S7, index at each moment Calculate the peak shaving amount at this moment: ;in, For the moment water consumption reductions; S8. Update the adjusted water consumption: ;in, For the Water consumption after adjustment at all times; Output adjusted water consumption sequence ; Output the total amount of peak shaving per layer .

[0014] Optionally, the closed-loop integrity check is performed to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, the error is allocated and corrected until the peak shaving consistency is met, specifically including: Calculate the actual total reduction: ;in, The total water consumption actually reduced during the entire process; Calculate the expected total reductions: ;in, is the sum of the total peak shaving amount of each layer; like , it indicates that the entire peak clipping process does not produce additional errors; like , then proceed to step S9 error correction, which specifically includes: S91. Calculate the total error ;in, The difference between the expected and actual peak shaving amount. A positive value indicates that further shaving is needed, while a negative value indicates that the peak shaving has been exceeded. S92. Set the error distribution amount at each moment: ;in, The water consumption that needs to be adjusted at each moment; S93, for all , unified: ;in, is the final water consumption after error correction; For the moment The layer index; S94, Order ; like , the correction is completed; like ,Will , repeat step S9 until consistency is met; The consistency includes the actual total reduction being equal to the expected total reduction.

[0015] A system for implementing the big data-based water-saving integrated service method, comprising: Data calculation module: used to perform data preprocessing, statistics calculation and peak shaving calculation; Layered processing module: used to adaptively determine the number of data layers, calculate layer thresholds, and perform time stratification based on the total number of observation moments; Verification and correction module: used to perform closed-loop integrity verification and error correction.

[0016] The present invention has the following beneficial effects: 1. Introducing full water usage data from continuous observations and standardizing it provides consistent input for subsequent analysis. Standardization reduces errors caused by inconsistent meter dimensions across regions and users, and supports large-scale deployment and comparative analysis. Traditional methods rely on single-point or daily statistics, ignoring temporal variations in water usage. This solution leverages the complete time series for more refined modeling.

[0017] 2. By taking the logarithm of the total observation time, the optimal number of strata is automatically calculated, eliminating the need for manual configuration. This reduces manual parameter selection and improves system deployment automation. The system dynamically adapts to data size and accommodates different granularity scenarios. Conventional methods often manually set stratification intervals or divide them into equally spaced intervals, lacking adaptability and making them difficult to handle unevenly distributed data.

[0018] 3. The entire data interval is linearly partitioned, and upper and lower thresholds are established for each layer. Each moment is then assigned to the corresponding layer based on the threshold, achieving a clearly structured and logically defined hierarchical system. Compared to existing methods based on fixed intervals or outlier processing, this method avoids subjective constraints and better reflects the natural distribution of data. Furthermore, an optimization strategy of skipping layers with no data is employed to reduce inefficient computing resource consumption and improve system efficiency.

[0019] 4. A hierarchical statistical mechanism is introduced to calculate the mean, standard deviation, and sample size of each layer, while extracting the global maximum standard deviation for subsequent normalization ratio setting. This design bases the peak shaving solution on quantitative analysis, specifically using standard deviation as a volatility metric. This allows the allocation of peak shaving amounts to be mathematically based, rather than empirically based. Compared to existing technologies, this invention's data processing is more interpretable and adaptable.

[0020] 5. By combining the peak value within a layer with the standard deviation, the planned total reduction is calculated based on a normalized ratio, achieving differentiated peak-shaving control at different levels. Compared to traditional methods of proportional compression or threshold truncation, this invention is more flexible and intelligent. Its peak reduction not only takes into account extreme values ​​but also considers the stability of the layer, making the overall control strategy both precise and stable, avoiding the uneven water supply or reduced user experience caused by aggressive reductions.

[0021] 6. Evenly distribute the total peak-shaving amount for each layer to the time of that layer, simplifying calculations while ensuring fair distribution. The updated data must still meet the constraint of not being lower than the lower bound of that layer to avoid damaging the integrity of the data structure. This strategy avoids unreasonable negative water consumption or excessive reduction issues, and has higher operational stability and user acceptance in actual engineering applications. Compared to individual strategies that overly focus on reducing single-point high values, the balanced processing of this invention is more in line with city-level water scheduling needs.

[0022] 7. Establish a comprehensive peak-shaving verification mechanism, comparing theoretical values ​​with actual adjustment results. If errors are detected, a fine-tuning mechanism is introduced to automatically correct the deviation until the expected target is met. An average error distribution strategy is used to make corrections simple and controllable. Traditional water-saving strategies rarely include closed-loop feedback mechanisms, and processing errors often require manual intervention. This method provides fully automated closed-loop control capabilities, significantly reducing operating and maintenance costs and improving its practicality and reliability. Its dynamic consistency judgment logic provides important support for big data control systems.

[0023] 8. Build a complete system architecture consisting of three modules: data calculation, hierarchical processing, and error correction, to achieve an end-to-end, closed-loop water control process. This modularization allows for rapid integration into smart water platforms or water-saving management and control systems, and interconnects with existing water supply scheduling and user-side equipment, demonstrating strong engineering feasibility and potential for widespread adoption. Compared to isolated algorithms, this invention provides a complete, executable implementation solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0026] Example, see Figure 1 , a comprehensive water-saving service method based on big data, including: Carry out data collection and preprocessing, obtain water consumption data in the continuous observation time series, and perform standardization on it; Adaptively determine the number of data stratification layers based on the total number of observation moments; According to the index of each layer, calculate the corresponding stratification threshold of each layer; All observation moments are stratified and classified according to the preset stratification threshold; For the time sets within each layer, calculate statistical characteristics, including sample size, average water consumption and standard deviation; Determine the adaptive peak shaving amount for each layer based on the maximum water consumption and standard deviation within the layer; Distribute the peak shaving amount equally according to time and update the adjusted water consumption data; Perform closed-loop integrity check to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, perform error allocation and correction until peak shaving consistency is met.

[0027] By building a complete set of data-driven water management processes, a series of problems, such as incomplete water use data analysis, inaccurate peak-shaving plans, and uncontrollable execution results, were systematically addressed. First, data quality was improved through standardized preprocessing, laying a solid foundation for subsequent modeling. Second, an adaptive hierarchical algorithm was used to segment water use data, effectively resolving the misclassification problem caused by traditional fixed interval divisions. Third, abnormal peaks in water use behavior were identified through hierarchical statistical analysis, and on this basis, standard deviation-driven adaptive peak-shaving volume setting was implemented, making the reduction measures targeted and flexible. Finally, a closed-loop error correction mechanism was used to ensure consistency between plan execution and theoretical design, resolving the problem of unverifiable results of traditional water-saving plans. Overall, this method integrates the advantages of data intelligence, algorithm adaptation, and closed-loop execution, significantly improving the scientific nature and practical operability of water-saving measures.

[0028] The data collection and preprocessing are to obtain the water consumption data in the continuous observation time series and perform standardization processing on it, specifically including: Collected in a continuous observation time sequence Water consumption data within ;in, For the Actual water consumption at the moment; is the total number of observation moments; obtain the original water consumption sequence to provide input data for subsequent stratification and statistics; Calculate the minimum water consumption for the entire sequence: ;in, is the minimum water consumption at all times; determine the lower bound of the stratification threshold; Calculate the maximum water consumption for the entire sequence: ;in, is the maximum water consumption at all times; determines the upper limit of the stratification threshold.

[0029] By constructing a unified water consumption data sequence and extracting the maximum and minimum values, the problem of incomparable and unprocessable data caused by scattered data sources and inconsistent dimensions in traditional water conservation management is solved. Especially in the context of different user groups or regions using different measurement equipment, standardization effectively improves the stability and consistency of subsequent stratification and statistical models. The maximum / minimum value calculation also provides a boundary reference for the subsequent stratification threshold setting, ensuring that all data points can be accurately classified into the corresponding layer. The introduction of this step enables the entire system to have the adaptability of "starting from the data", avoiding the impact of data bias on model results, and providing a foundation for achieving true data-driven water conservation.

[0030] The adaptive determination of the number of data stratification layers based on the total number of observation moments specifically includes: According to the total time Calculate the number of layers: ;in, Round down; adaptively set the number of layers to balance layer accuracy and computational complexity.

[0031] By setting the number of stratifications based on the total number of observation moments, the problem of "over-segmentation" or "extensive segmentation" in different scenarios of water use data volatility is solved. Traditional water-saving systems often rely on manually specified stratification levels, which can easily lead to situations where they are not adapted to the actual data distribution, thus affecting the accuracy of stratification. The present invention effectively balances model accuracy and computational complexity through adaptive calculation of the number of layers. It is not only suitable for scenarios with large amounts of data, but can also converge quickly under small sample conditions. At the same time, the growth rate of the number of layers is controlled through logarithmic transformation to ensure that the system still has good computing performance and resource consumption control capabilities when facing ultra-large-scale data, thereby improving the versatility and intelligence of the system.

[0032] The calculation of the layer threshold corresponding to each layer based on the index of each layer specifically includes: Index each layer , calculate the boundary threshold: ; in, is the current layer index; For the The boundary threshold of the layer; when hour, ;when hour, ; Linearly divide the upper and lower bounds of each layer and construct the hierarchical interval .

[0033] This method achieves precise demarcation of continuous water consumption data intervals, resolving the imbalanced distribution problem caused by the traditional use of fixed intervals or empirical thresholds. Linear interpolation evenly demarcates intervals between the maximum and minimum values, ensuring clear boundaries at each level and effectively improving the scientific nature and logical consistency of the stratification. This method ensures the stability and dispersion of data points across layers, particularly when dealing with multimodal or non-average water consumption curves. This provides more balanced data support for subsequent statistical calculations, thereby improving the accuracy and effectiveness of peak-shaving strategy design.

[0034] The stratification and classification of all observation moments according to the preset stratification threshold specifically includes: For each layer , build the moment index collection: ; in, For the The time index included in the layer; each moment is assigned to the corresponding layer according to the water consumption interval; like , then the subsequent calculation of this layer is skipped; among them, For collection The number of elements; avoid wasting computing resources on layers without data.

[0035] The system accurately maps each water usage moment to its corresponding layer, forming a clear data organization structure within the layer. This structure not only enhances data management efficiency but also provides a good grouping basis for hierarchical statistics. In particular, when certain intervals lack sufficient data, the logic of skipping empty layers avoids wasting computing resources on meaningless layers, improving overall algorithm execution efficiency and system responsiveness. This mechanism significantly reduces system resource waste, supports high-performance data processing and parallel computing, and provides a technical foundation for real-time water conservation decisions.

[0036] The statistical characteristics are calculated for the time sets in each layer, including the number of samples, average water consumption and standard deviation, specifically including: For each Layer , perform the following steps: S1. Calculate the number of samples in the layer: ;in, For the The number of stratum moments; the denominator for subsequent mean and standard deviation calculations; S2. Calculate the average water consumption within the layer: ;in, For the Average water consumption of the layer; measure the central trend of the layer; S3. Calculate the standard deviation of water consumption within the layer: ;in, For the Standard deviation of water consumption at each layer; measures the degree of fluctuation of water consumption at that layer and is used for adaptive peak shaving; S4. Calculate the global maximum standard deviation: ;in, is the maximum standard deviation among all layers; normalize the standard deviation of each layer to ensure that the subsequent ratio calculation is dimensionless.

[0037] Introducing standard deviation calculations to guide the differentiated setting of subsequent peak-shaving strategies is a key step in the intelligentization of the system. Compared to using only the average value as the peak-shaving benchmark, this solution further measures the volatility of data within a layer through standard deviation, which helps to accurately identify "unstable water use" layers. By calculating the global maximum standard deviation, the degree of fluctuation in each layer can be normalized, making the calculation of peak-shaving amounts more comparable and consistent across multiple layers, effectively solving the problem of peak-shaving strategy failure due to large differences in statistical characteristics between layers. This mechanism significantly improves the response flexibility and customization capabilities of the water-saving system.

[0038] Determining the adaptive peak shaving amount of each layer based on the maximum water consumption and standard deviation within the layer specifically includes: For each Layer , perform the following steps: S5. Calculate the peak value within the layer: ;in, For the Maximum water consumption per floor; determine peak water consumption for calculating peak shaving benchmark; S6. Calculate the total peak shaving amount within the layer: ;in, For the The total water consumption to be reduced by each level is planned; the peak reduction amount is adaptively set according to the degree of fluctuation and peak height.

[0039] Traditional water-saving solutions often rely on manually set percentages to reduce peak water use, lacking the ability to tailor strategies to actual data fluctuations. This solution, by introducing a dual-factor mechanism—maximum water use within a stratum + standard deviation—to determine peak water use reduction. This allows the reduction strategy to account for both extremes and fluctuations in water use, resolving the dilemma of "too little reduction is ineffective, while too much reduction triggers complaints." Furthermore, setting reductions based on the standard deviation ratio allows for flexible regulation during periods of high water use fluctuations, improving user experience and system controllability while avoiding the discomfort and social resistance often associated with one-size-fits-all solutions.

[0040] The peak shaving amount is distributed equally at each time, and the adjusted water consumption data is updated, specifically including: For each Layer , perform the following steps: S7, index at each moment Calculate the peak shaving amount at this moment: ;in, For the moment The total peak reduction amount is evenly distributed to each moment of this layer; S8. Update the adjusted water consumption: ;in, For the The water consumption after adjustment at all times; ensure that it is not lower than the lower limit of this layer after adjustment, and smooth the peak water consumption; Output adjusted water consumption sequence ; Output the total amount of peak shaving per layer .

[0041] The equal allocation mechanism improves the fairness and feasibility of peak-shaving strategies, avoiding concentrated reductions at a single moment that could cause user dissatisfaction or system disruption. By smoothly adjusting water consumption, this method effectively reduces the dramatic fluctuations between peaks and valleys, improving the stability of the overall water use curve. Furthermore, the output of the adjusted water consumption sequence provides an interface for subsequent monitoring systems or visualization interfaces, facilitating post-implementation evaluation and feedback, effectively enhancing the intelligence and visualization of water conservation management.

[0042] The closed-loop integrity check is performed to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, the error is allocated and corrected until the peak shaving consistency is met. Specifically, it includes: Calculate the actual total reduction: ;in, The total water consumption actually reduced in the whole process; measure the effectiveness of the implementation of the method; Calculate the expected total reductions: ;in, The sum of the total peak reduction of each layer; the sum of the theoretical reductions designed by the method; like , it indicates that the entire peak shaving process does not produce additional errors; verifying that the entire process is error-free and the closed loop is completed; like , then proceed to step S9 error correction, which specifically includes: S91. Calculate the total error ;in, The difference between the expected and actual peak shaving amount. A positive value indicates that further shaving is needed, while a negative value indicates that the peak shaving has been exceeded. It quantifies the amount of peak shaving that needs to be compensated. S92. Set the error distribution amount at each moment: ;in, The water consumption that needs to be readjusted at each moment; the overall error is evenly distributed to simplify the correction; S93, for all , unified: ;in, is the final water consumption after error correction; For the moment The layer index to which it belongs; fine-tune according to the error while not breaking the layer boundary; S94, Order ; like , the correction is completed; like ,Will , repeat step S9 until consistency is met; The consistency includes the actual total reduction being equal to the expected total reduction.

[0043] Traditional water-saving measures often lack feedback and correction of implementation results, resulting in a significant disconnect between theory and practice. This solution compares expected and actual water reductions, automatically quantifies errors, and makes gradual adjustments, effectively implementing closed-loop control strategies. This mechanism not only improves the reliability of water-saving systems but also offers excellent scalability and dynamic adaptability, making it suitable for deployment in complex scenarios such as urban water services, building water conservation, and industrial smart water use.

[0044] This embodiment also provides a system for a comprehensive water-saving service method based on big data, including: Data calculation module: used to perform data preprocessing, statistics calculation and peak shaving calculation; Layered processing module: used to adaptively determine the number of data layers, calculate layer thresholds, and perform time stratification based on the total number of observation moments; Verification and correction module: used to perform closed-loop integrity verification and error correction.

[0045] The system architecture clearly divides responsibilities: the data computing module handles underlying operations, the layered processing module implements logical division and decision-making, and the verification module provides closed-loop control, making it an ideal platform for intelligent water management. Through modular decoupling, the system can be expanded on demand, supporting algorithm upgrades and cloud deployment, significantly improving project implementation capabilities and commercial feasibility.

[0046] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0047] 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 within the scope of protection of the present invention.

Claims

1. A comprehensive water-saving service method based on big data, characterized in that: include: Carry out data collection and preprocessing, obtain water consumption data in the continuous observation time series, and perform standardization on it; Adaptively determine the number of data stratification layers based on the total number of observation moments; According to the index of each layer, calculate the corresponding stratification threshold of each layer; All observation moments are stratified and classified according to the preset stratification threshold; For the time sets within each layer, calculate statistical characteristics, including sample size, average water consumption and standard deviation; Determine the adaptive peak shaving amount for each layer based on the maximum water consumption and standard deviation within the layer; Distribute the peak shaving amount equally according to time and update the adjusted water consumption data; Perform closed-loop integrity check to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, perform error allocation and correction until peak shaving consistency is met.

2. A water-saving comprehensive service method based on big data according to claim 1, characterized in that: The data collection and preprocessing are to obtain the water consumption data in the continuous observation time series and perform standardization processing on it, specifically including: Collected in a continuous observation time sequence Water consumption data within ;in, For the Actual water consumption at the moment; is the total number of observation times; Calculate the minimum water consumption for the entire sequence: ;in, The minimum water consumption at all times; Calculate the maximum water consumption for the entire sequence: ;in, The maximum water consumption at all times.

3. A water-saving integrated service method based on big data according to claim 2, characterized in that: The adaptive determination of the number of data stratification layers based on the total number of observation moments specifically includes: According to the total time Calculate the number of layers: ;in, To round down.

4. A water-saving integrated service method based on big data according to claim 3, characterized in that: The calculation of the layer threshold corresponding to each layer based on the index of each layer specifically includes: Index each layer , calculate the boundary threshold: ； in, is the current layer index; For the The boundary threshold of the layer; when hour, ;when hour, .

5. A water-saving integrated service method based on big data according to claim 4, characterized in that: The stratification and classification of all observation moments according to the preset stratification threshold specifically includes: For each layer , build the moment index collection: ； in, For the The time index that the layer contains; like , then the subsequent calculation of this layer is skipped; among them, For collection Number of elements.

6. A water-saving integrated service method based on big data according to claim 5, characterized in that: The statistical characteristics are calculated for the time sets in each layer, including the number of samples, average water consumption and standard deviation, specifically including: For each Layer , perform the following steps: S1. Calculate the number of samples in the layer: ;in, For the Layer moment number; S2. Calculate the average water consumption within the layer: ;in, For the Average water consumption per floor; S3. Calculate the standard deviation of water consumption within the layer: ;in, For the Standard deviation of water consumption per layer; S4. Calculate the global maximum standard deviation: ;in, is the maximum standard deviation among all strata.

7. A water-saving integrated service method based on big data according to claim 6, characterized in that: Determining the adaptive peak shaving amount of each layer based on the maximum water consumption and standard deviation within the layer specifically includes: For each Layer , perform the following steps: S5. Calculate the peak value within the layer: ;in, For the Maximum water consumption per layer; S6. Calculate the total peak shaving amount within the layer: ;in, For the The total water consumption reduction planned for each level.

8. The water-saving integrated service method based on big data according to claim 7, characterized in that: The peak shaving amount is distributed equally at each time, and the adjusted water consumption data is updated, specifically including: For each Layer , perform the following steps: S7, index at each moment Calculate the peak shaving amount at this moment: ;in, For the moment water consumption reductions; S8. Update the adjusted water consumption: ;in, For the Water consumption after adjustment at all times; Output adjusted water consumption sequence ; Output the total amount of peak shaving per layer .

9. The water-saving integrated service method based on big data according to claim 8, characterized in that: The closed-loop integrity check is performed to verify whether the actual water consumption reduction is consistent with the expected value. If there is an error, the error is allocated and corrected until the peak shaving consistency is met. Specifically, it includes: Calculate the actual total reduction: ;in, The total water consumption actually reduced during the entire process; Calculate the expected total reductions: ;in, is the sum of the total peak shaving amount of each layer; like , it indicates that the entire peak clipping process does not produce additional errors; like , then proceed to step S9 error correction, which specifically includes: S91. Calculate the total error ;in, The difference between the expected and actual peak shaving amount. A positive value indicates that further shaving is needed, while a negative value indicates that the peak shaving has been exceeded. S92. Set the error distribution amount at each moment: ;in, The water consumption that needs to be adjusted at each moment; S93, for all , unified: ;in, is the final water consumption after error correction; For the moment The layer index; S94, Order ; like , the correction is completed; like ,Will , repeat step S9 until consistency is met; The consistency includes the actual total reduction being equal to the expected total reduction.

10. A system using the big data-based water-saving integrated service method according to claim 9, characterized in that: include: Data calculation module: used to perform data preprocessing, statistics calculation and peak shaving calculation; Layered processing module: used to adaptively determine the number of data layers, calculate layer thresholds, and perform time stratification based on the total number of observation moments; Verification and correction module: used to perform closed-loop integrity verification and error correction.

Images