Intelligent water conservancy storage device with automatic adjustment function and control method

Abstract

The application relates to the technical field of industrial control, in particular to an intelligent water conservancy storage device with an automatic adjusting function and a control method, which solves the technical problem that the prior art only establishes a prediction model according to water body data, ignores the equipment condition of a water gate, and leads to an unsatisfactory regulation effect. The method comprises the following steps: acquiring index data of multiple indexes of a monitored water gate at multiple acquisition moments of a monitoring period; mapping the index data to a multidimensional sample space, and analyzing the data distribution of the index data; comparing the data distribution of multiple monitoring periods, analyzing the recession degree of each index, and analyzing the index correlation degree between each index and other indexes; adjusting the water gate opening degree of the current monitoring period according to the index data, the data distribution, the recession degree of each index and the index correlation degree of the current monitoring period, obtaining a pre-regulation opening degree of the next monitoring period; and regulating the water gate opening degree of the next monitoring period based on the pre-regulation opening degree.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, specifically to an intelligent water storage and regulation device and control method with automatic adjustment function. Background Technology

[0002] Natural water resources are unevenly distributed across regions, with varying total water resources and supply-demand conditions. Therefore, water conservancy projects are necessary to regulate and manage water volume, ensuring the rational allocation of water resources. Sluice gates are common hydraulic structures on rivers or canals used to control flow and regulate water levels; water flow is managed by controlling the opening of the sluice gates.

[0003] To improve the accuracy and timeliness of water flow control, existing technologies typically establish predictive models based on monitoring water volume, water level, and other water body data in the area where the sluice gate is located. These models are then used to adjust the sluice gate opening to achieve water flow control. However, in actual operation, long-term exposure to pollution, water flow, load, and weather can cause sluice gates to age or suffer partial damage. This reduces their opening and closing capacity, affecting their ability to control water flow. Consequently, the control parameters obtained from predictive models based solely on water body data may not achieve the desired control effect. Summary of the Invention

[0004] To address the problem that existing technologies rely solely on water body data to build predictive models, neglecting the equipment conditions of sluice gates and resulting in unsatisfactory regulation effects, the present invention aims to provide an intelligent water conservancy regulation and storage device and control method with automatic adjustment function. The specific technical solution adopted is as follows:

[0005] Firstly, a smart water control method with automatic adjustment function is provided, comprising: acquiring indicator data of multiple indicators of a sluice gate at multiple acquisition times during a monitoring cycle; mapping the indicator data to a multi-dimensional sample space and analyzing the data distribution of the indicator data; each indicator corresponding to one dimension of the multi-dimensional sample space; comparing the data distribution of multiple monitoring cycles, analyzing the degree of decline of each indicator, and analyzing the correlation between each indicator and other indicators; adjusting the sluice gate opening degree of the current monitoring cycle based on the indicator data, data distribution, degree of decline of each indicator, and correlation of indicators to obtain the pre-controlled opening degree for the next monitoring cycle; and regulating the sluice gate opening degree for the next monitoring cycle based on the pre-controlled opening degree.

[0006] Based on the above technical solution, in the intelligent water conservancy control method with automatic adjustment function provided by this invention, the changes of multi-dimensional indicator data over time are analyzed by analyzing the data distribution of multi-dimensional sample space. The degree of indicator decay is used to quantify the aging degree of the sluice gate, achieving accurate identification of the decline in sluice gate control capability. This improves upon the shortcomings of existing technologies that ignore the decline in control capability caused by the aging of the sluice gate itself, and enhances the accuracy of sluice gate opening regulation. Then, by optimizing the weight allocation by combining multi-dimensional data with indicator correlation, the mutual influence of multiple indicators on the sluice gate control capability can be comprehensively considered, making the regulation allocation more in line with the actual operating logic, avoiding the one-sidedness of single indicator regulation, and further improving the regulation effect.

[0007] In conjunction with the first aspect above, in one possible implementation, the method for regulating the sluice gate opening in the next monitoring cycle based on the pre-regulation opening degree specifically includes: adjusting the pre-regulation opening degree according to the indicator data and indicator correlation degree obtained at each acquisition time in the next monitoring cycle to obtain the real-time sluice gate opening degree at each acquisition time.

[0008] In conjunction with the first aspect above, in one possible implementation, the aforementioned data distribution includes: the degree of clustering and the degree of outlier in the indicator data; the method for analyzing the data distribution of the indicator data specifically includes: mapping the indicator data to multidimensional data points in a multidimensional sample space, mapping to a single-dimensional sample space corresponding to each indicator, and clustering the single-dimensional data points in the single-dimensional sample space to obtain multiple clusters corresponding to each indicator; determining the clustering index of the single-dimensional data points within each cluster based on the number of single-dimensional data points in each cluster and the distance between the single-dimensional data points and the cluster center; the clustering index of the single-dimensional data points characterizes the degree of clustering of the indicator data; determining the outlier index of the single-dimensional data points within each cluster based on the distance between the single-dimensional data points and the cluster center and the degree of clustering of the single-dimensional data points; the outlier index of the single-dimensional data points characterizes the degree of outlier in the indicator data.

[0009] In conjunction with the first aspect above, in one possible implementation, the method of clustering single-dimensional data points in a single-dimensional sample space to obtain multiple clusters corresponding to each indicator specifically includes: clustering single-dimensional data points in a single-dimensional sample space to obtain multiple clusters by using an iterative self-organizing data analysis techniques algorithm (ISODATA) and preset clustering parameters.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the method for comparing the data distribution across multiple monitoring periods and analyzing the degree of decay of each indicator specifically includes: normalizing the outlier index of single-dimensional data points in each monitoring period, and marking single-dimensional data points whose normalization results are greater than a preset threshold as outliers; determining the degree of fluctuation of each indicator in the two consecutive monitoring periods based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in the two consecutive monitoring periods, the number of outliers, and the number of clusters of multiple clusters; and determining the degree of decay of each indicator based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in the multiple monitoring periods and the degree of fluctuation of each indicator.

[0011] In conjunction with the first aspect above, in one possible implementation, the method for analyzing the correlation between each indicator and other indicators specifically includes: determining the correlation between indicators based on the aggregation index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods, the degree of decay of each indicator, the number of indicators for multiple indicators, and the number of periods for multiple monitoring periods.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, the method of adjusting the sluice gate opening for the current monitoring period based on the indicator data, data distribution, the degree of decay of each indicator, and the correlation of indicators to obtain the pre-regulation opening for the next monitoring period specifically includes: determining the degree to be regulated for the sluice gate opening based on the indicator data of the current monitoring period, the aggregation index of single-dimensional data points within multiple clusters corresponding to each indicator, the degree of decay of each indicator, and the correlation of indicators; and determining the pre-regulation opening based on the sluice gate opening and the degree to be regulated for the current monitoring period.

[0013] In conjunction with the first aspect above, in one possible implementation, the method of adjusting the pre-regulation opening degree based on the indicator data and indicator correlation degree obtained at each collection time in the next monitoring cycle to obtain the real-time sluice gate opening degree at each collection time specifically includes: determining the real-time sluice gate opening degree at the target collection time based on the indicator data, indicator correlation degree, pre-regulation opening degree, and the real-time sluice gate opening degree at the previous collection time.

[0014] In conjunction with the first aspect mentioned above, in one possible implementation, the aforementioned multiple indicators include at least: upstream actual water level, downstream actual water level, sluice gate uplift pressure, upstream water flow velocity, and downstream water flow velocity data.

[0015] Secondly, an intelligent water storage and regulation device with automatic adjustment function is provided, comprising: a data acquisition module, a data processing module, and a regulation module; the data acquisition module is used to acquire indicator data of multiple indicators of the monitored sluice gate at multiple acquisition times during the monitoring cycle; the data processing module is used to map the indicator data to a multi-dimensional sample space and analyze the data distribution of the indicator data; each indicator corresponds to one dimension of the multi-dimensional sample space; the data processing module is also used to compare the data distribution of multiple monitoring cycles, analyze the degree of decline of each indicator, and analyze the correlation between each indicator and other indicators; the data processing module is also used to adjust the sluice gate opening of the current monitoring cycle according to the indicator data, data distribution, degree of decline of each indicator, and correlation of indicators in the current monitoring cycle to obtain the pre-regulated opening for the next monitoring cycle; the regulation module is used to regulate the sluice gate opening of the next monitoring cycle based on the pre-regulated opening.

[0016] Thirdly, an intelligent water storage and regulation device with automatic adjustment function is provided, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is used to execute the instructions to implement the method described in the first aspect and any possible implementation thereof. The intelligent water storage and regulation device with automatic adjustment function can be an electronic device or a chip within an electronic device.

[0017] Fourthly, a computer-readable storage medium is provided, which stores instructions that, when executed on an intelligent water storage device with automatic regulation function, cause the intelligent water storage device with automatic regulation function to perform the methods described in the first aspect and any possible implementation thereof.

[0018] Fifthly, a computer program product containing instructions is provided, which, when running on an intelligent water storage device with automatic adjustment function, causes the intelligent water storage device with automatic adjustment function to perform the methods described in the first aspect and any possible implementation thereof.

[0019] The present invention has the following beneficial effects:

[0020] By analyzing the data distribution in a multidimensional sample space, the changes of multidimensional indicator data over time are observed. The degree of indicator decay is used to quantify the aging of the sluice gate, enabling accurate identification of the decline in sluice gate control capability. This improves upon existing technologies that neglect the control capability decline caused by the aging of the sluice gate itself, thus enhancing the accuracy of sluice gate opening regulation. Furthermore, by optimizing weight allocation through multidimensional data and indicator correlation, the mutual influence of multiple indicators on sluice gate control capability can be comprehensively considered. This makes the regulation allocation more aligned with actual operational logic, avoiding the one-sidedness of single-indicator regulation and further improving the regulation effect. Attached Figure Description

[0021] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A device architecture diagram of an intelligent water storage device with automatic adjustment function provided in one embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a water gate structure for a control gate is provided in one embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of a gate regulating system according to an embodiment of the present invention;

[0025] Figure 4 A flowchart illustrating an intelligent water control method with automatic adjustment function provided in one embodiment of the present invention;

[0026] Figure 5 This is a schematic diagram of a multidimensional sample space provided in one embodiment of the present invention;

[0027] Figure 6 A flowchart of another intelligent water control method with automatic adjustment function provided in an embodiment of the present invention;

[0028] Figure 7 This is a schematic diagram of a clustering result provided in one embodiment of the present invention;

[0029] Figure 8 This is a schematic diagram of a water flow monitoring method according to an embodiment of the present invention;

[0030] Figure 9 This is a schematic diagram of the hardware structure of an intelligent water storage device with automatic adjustment function provided in one embodiment of the present invention. Detailed Implementation

[0031] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent water storage device and control method with automatic adjustment function proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] The following description, in conjunction with the accompanying drawings, details the specific scheme of an intelligent water storage device and control method with automatic adjustment function provided by the present invention.

[0034] Please see Figure 1 This diagram illustrates the device architecture of an intelligent water storage and regulation device with automatic adjustment function according to an embodiment of the present invention. The intelligent water storage and regulation device 100 with automatic adjustment function includes: a data acquisition module 101, a data processing module 102, and a control module 103. Figure 2 As shown, taking a control gate as an example, the opening and closing degree of the sluice gate is adjusted by the intelligent water conservancy regulation and storage device 100, thereby achieving precise control of water flow.

[0035] The data acquisition module 101 is used to collect key indicator data of the sluice gate's own status and water flow status in real time, providing raw data support for subsequent analysis.

[0036] In some implementations, the data acquisition module 101 includes at least one water level sensor (one upstream and one downstream), a vibrating wire lifter (on the sluice gate body), and a current meter (one upstream and one downstream). Specifically, the water level sensor (accuracy ±0.01m) is installed on the upstream and downstream riverbanks, within 50m of the sluice gate; the vibrating wire lifter (accuracy ±0.5kPa) is embedded in the concrete at the bottom of the sluice gate, with three evenly distributed measuring points to take the average value; the current meter (accuracy ±0.05m / s) is installed in the stable water flow sections upstream and downstream, at a depth of 0.5m above the water surface.

[0037] The data processing module 102 is used to integrate, analyze and calculate the collected data, including constructing the sample space, screening outliers, quantifying the degree of aging, and calculating the control coefficient. It is the core module of the device.

[0038] The control module 103 is used to drive the actuator to adjust the opening degree of the sluice gate according to the real-time opening command output by the data processing module 102, so as to realize water flow control.

[0039] In some implementations, such as Figure 3 The gate regulating system shown includes at least a gate opening controller, a gate hoist (such as a screw-type gate hoist), and a gate actuator in the regulating module 103.

[0040] In some implementations, the intelligent water storage device 100 with automatic adjustment function also includes a data transmission module 104 and an output module 105.

[0041] The data transmission module 104 is used to establish a real-time data channel between the acquisition end and the processing end to ensure that the original data is transmitted without delay or loss.

[0042] The output module 105 is used to intuitively display key information about the operation of the sluice gate, providing staff with a basis for status monitoring and maintenance. It may include display devices (such as industrial touch screens, remote monitoring terminals) and status indicator lights (optional, used for aging degree / fault warning).

[0043] In the intelligent water storage device 100 with automatic adjustment function, the data acquisition module 101 sends the acquired data to the data processing module 102 through the data transmission module 104 to ensure the timeliness of the data; the data processing module 102 converts the calculated real-time opening degree (such as opening percentage, gate lifting height) into an electrical signal command and sends it to the control module 103; after receiving the command, the control module 103 drives the hoist to operate, drives the gate to lift and lower, and precisely adjusts the opening degree of the sluice gate.

[0044] The data processing module 102 can also send key information such as the aging degree of the sluice gate (based on a comprehensive judgment of the decay index, such as mild / moderate / severe), the current real-time opening degree, and the pre-controlled opening degree to the output module 105 for real-time display, which is convenient for staff to monitor.

[0045] In some implementations, if the output module 105 supports interactive functions (such as a touch screen), staff can input manual intervention commands (such as emergency gate closure or adjustment of monitoring cycle) through the output module 105. The commands are transmitted back to the data processing module 102 via the data link, triggering the control module 103 to perform manual operations, thereby improving the system's flexibility.

[0046] Please see Figure 4 The diagram illustrates a flowchart of an intelligent water control method with automatic adjustment function according to an embodiment of the present invention. This intelligent water control method with automatic adjustment function includes:

[0047] S1. Acquire indicator data of multiple indicators of the monitoring sluice gate at multiple acquisition times during the monitoring cycle.

[0048] In some implementations, detection instruments are set up to acquire aging data of the sluice gate itself and water flow data that needs to be regulated. For example, a monitoring cycle of 24 hours (h) is used, with data recorded every 10 minutes (min), meaning each cycle contains 144 data collection points to ensure data timeliness and continuity, covering different states such as peak and off-peak water flow. Several indicators include at least:

[0049] The actual upstream and downstream water levels (unit: meters) collected by the water level sensor can reflect the water height upstream and downstream of the sluice gate;

[0050] The uplift pressure (unit: kPa) collected by the vibrating wire pressure gauge can reflect the upward pressure of the water flow on the sluice gate structure and affect its stability;

[0051] The upstream and downstream water flow velocity data (unit: meters per second) collected by the current meter can reflect the speed of water movement in the upstream and downstream areas.

[0052] S2. Map the indicator data to a multidimensional sample space and analyze the data distribution of the indicator data.

[0053] Among them, such as Figure 5 As shown, each indicator corresponds to one dimension (i.e., one coordinate axis) of the multidimensional sample space. By establishing a multidimensional sample space for each monitoring period and arranging the obtained data in the multidimensional sample space, the data distribution of each monitoring period can be intuitively displayed.

[0054] In some implementations, multiple coordinate axes are established based on multiple dimensions of the multidimensional sample space, and the coordinate axis scales are normalized (e.g., the index values ​​are mapped to the [0, 1] interval) to eliminate dimensional differences. To achieve the normalization of the coordinate axis scales in the multidimensional sample space and ensure the effectiveness of the analysis, the original values ​​of multiple indicators at each collection time must be normalized accordingly to obtain the index data, so that each indicator has equal weight in the space.

[0055] Each data acquisition moment represents a single data point in space after normalization of multiple index values. Specific dimensional distributions may include:

[0056] Upstream and downstream water level difference: The direction of the increase of the difference between the actual upstream water level and the actual downstream water level (upstream and downstream water level difference) is taken as the positive direction of the coordinate axis;

[0057] Uplift pressure: The direction of increasing uplift pressure at the sluice gate is taken as the positive direction of the coordinate axis;

[0058] Upstream water flow velocity: The direction of increasing upstream water flow velocity is taken as the positive direction of the coordinate axis;

[0059] Downstream flow velocity: The direction of increasing downstream flow velocity is the positive direction of the coordinate axis.

[0060] The control capability of a sluice gate and the state of water flow are the result of the synergistic effect of multiple indicators, rather than the independent influence of a single indicator. For example, an increase in uplift pressure not only directly reduces the gate's anti-sliding stability but also causes abnormal fluctuations in the water level difference between upstream and downstream, thus affecting the water flow velocity; conversely, sudden changes in flow velocity will also exacerbate the changes in uplift pressure. Moreover, the core manifestation of sluice gate aging is the decline in control capability. This decline is not an isolated change in a single indicator but a holistic deterioration in the distribution of multiple indicator data. For example, aging sluice gates will simultaneously exhibit phenomena such as an expanded range of water level difference fluctuations, dispersed uplift pressure data, and a shortened duration of flow velocity stability, and these changes will superimpose and amplify each other.

[0061] If each indicator is analyzed separately using a single-dimensional sample space, the inherent logic of the interrelationship and mutual influence between these indicators will be severed. For example, looking only at the flow velocity data, it is impossible to determine whether the fluctuation is caused by changes in the water flow itself or by a chain reaction caused by abnormal uplift pressure, which can easily lead to misjudgment of the actual operating status of the sluice gate.

[0062] The multidimensional sample space incorporates all indicators into the same analytical framework. Each data point is a collection of multidimensional information such as water level difference, uplift pressure, and flow velocity, which can intuitively present the linkage between indicators. If the data in one dimension is abnormal, it may be accompanied by synchronous fluctuations in other dimensions.

[0063] In some implementations, the spatial density (number of data points per unit volume) and average distance (average Euclidean distance from all points to the center of space) of the data points are calculated. The higher the density and the smaller the average distance, the more stable the distribution of the index data.

[0064] S3. Compare the data distribution across multiple monitoring periods, analyze the degree of decline of each indicator, and analyze the correlation between each indicator and other indicators.

[0065] Single-period data can only reflect the state of indicators at a certain moment / stage, and cannot distinguish whether data fluctuations are caused by short-term accidental factors (such as a single rainstorm) or long-term aging. However, by comparing the data distribution of multiple monitoring periods (such as the past month), the gradual degradation caused by aging can be captured in multiple dimensions, thereby quantifying the degree of decline.

[0066] In some implementations, data from the past 30 monitoring periods (1 month) are selected, and the distribution differences between the current period and historical periods are compared. The degree of decline can be quantified through the following dimensions:

[0067] Distribution offset: By comparing the mean (center) of the data distribution of each period with the Euclidean distance between the current period's data center and the historical average center, if the mean continuously deviates from the initial stable value (e.g., the average water level difference increases from 2m to 3.5m month by month), it indicates that the performance of the sluice gate structure has changed (e.g., the sluice chamber settlement causes the water level control to fail), which is a direct signal of aging.

[0068] Variation in dispersion: Calculate the standard deviation / variance of the index data for each period. If the dispersion continues to increase over time (e.g., the standard deviation of uplift pressure increases from 3 kPa to 8 kPa month by month), it indicates that the sluice gate's ability to control the index has decreased and the degree of decline has deepened.

[0069] Outlier percentage: The percentage of outliers in each period is statistically analyzed. Due to the decline in control capacity, aging sluice gates are more prone to extreme data (such as sudden increases in flow velocity), and the percentage of outliers will increase significantly over time.

[0070] Furthermore, the operation of a sluice gate is the result of the synergistic effect of multiple indicators (such as changes in uplift pressure affecting water level difference, and changes in water level difference affecting flow velocity). The correlation between indicators reflects this inherent logic of "mutual influence and mutual constraint." Its core function is to provide a scientific basis for weight allocation for "precise control of sluice gate opening" and avoid the one-sidedness of single indicator control.

[0071] In some implementations, the degree of linear correlation can be quantified by calculating the Pearson correlation coefficient between any two indicators; the higher the correlation coefficient, the stronger the linkage between the two indicators.

[0072] S4. Based on the indicator data, data distribution, degree of decline of each indicator, and correlation of indicators in the current monitoring period, adjust the sluice gate opening in the current monitoring period to obtain the pre-regulation opening for the next monitoring period.

[0073] Among these, the indicator data of the current monitoring period directly reflects the real-time water flow status and determines the basic direction of the opening adjustment; the data distribution reflects the stability of the water flow status and corrects the opening adjustment range; the degree of decline of each indicator quantifies the reduction in the sluice gate's control capability and compensates for opening adjustment deviations; the correlation between indicators determines the control weight of each indicator, avoiding the one-sidedness of a single indicator dominating. The combination of these four aspects can form a closed loop of real-time status, stability, aging compensation, and synergistic weights, enabling the pre-regulation opening to accurately respond to current water flow changes and adapt to the performance degradation of the sluice gate due to aging, thus achieving adaptive hydraulic control.

[0074] S5. Based on the pre-controlled opening degree, adjust the sluice gate opening degree for the next monitoring cycle.

[0075] In some implementation methods, the pre-controlled opening degree can be directly used as the sluice gate opening degree for the next monitoring cycle. Alternatively, the sluice gate opening degree can be further dynamically corrected based on the pre-controlled opening degree and real-time data.

[0076] Based on the above technical solution, the changes of multi-dimensional indicator data over time are analyzed by examining the data distribution in a multi-dimensional sample space. The degree of indicator decay is used to quantify the aging of the sluice gate, enabling accurate identification of the decline in sluice gate control capability. This improves upon existing technologies that neglect the control capability decline caused by the aging of the sluice gate itself, thus enhancing the accuracy of sluice gate opening control. Furthermore, by optimizing weight allocation through multi-dimensional data and indicator correlation, the mutual influence of multiple indicators on sluice gate control capability can be comprehensively considered. This makes the control allocation more aligned with actual operational logic, avoiding the one-sidedness of single-indicator control and further improving the control effect.

[0077] In one possible implementation, combining Figure 4 ,like Figure 6 As shown, the method in S5 above can be specifically implemented through the following S51, which will be explained in detail below:

[0078] S51. Based on the indicator data and correlation degree of each collection moment in the next monitoring cycle, adjust the pre-control opening degree to obtain the real-time sluice gate opening degree at each collection moment.

[0079] Based on the above technical solutions, secondary corrections can be made by combining real-time indicator data and correlation, taking into account both long-term trend predictions and short-term emergencies, avoiding the lag of single pre-control, and improving the timeliness and accuracy of water conservancy control.

[0080] In one possible implementation, the data distribution includes the degree of clustering and outlier of the indicator data; the method for analyzing the data distribution of the indicator data in S2 can be specifically implemented through S21 to S23, which are explained in detail below:

[0081] S21. Map the indicator data to multidimensional data points in a multidimensional sample space, map them to the single-dimensional sample space corresponding to each indicator, and cluster the single-dimensional data points in the single-dimensional sample space to obtain multiple clusters corresponding to each indicator.

[0082] Sluice gates regulate water level and flow by controlling their opening and closing. Due to natural factors such as rainfall, the water flow at a sluice gate location varies at different times. The sluice gate controls its opening based on these flow differences. During the regulation process, the water flow state also changes (for example, when the sluice gate is open during flood discharge, the downstream flow increases over time, causing the upstream water level to gradually decrease and the downstream water level to gradually rise, while the water flow velocity gradually stabilizes). Furthermore, the impact of different water flow states on the sluice gate varies, causing the uplift pressure on the sluice gate to change accordingly. Therefore, the distribution of data acquired at different times in the sample space differs, and the distribution is relatively similar when the water flow states are similar.

[0083] Based on the above analysis, in order to analyze the water passage capacity of the sluice gate in different periods, taking a single monitoring cycle as an example, the corresponding sample space is clustered, and the sluice gate control capacity of the current cycle is analyzed according to the data distribution.

[0084] It should be noted that the data is first mapped to a multi-dimensional sample space, and then the data points in the space are mapped to a single dimension for clustering, rather than directly clustering each single indicator in its single-dimensional space. The core purpose is to achieve data processing while preserving the synergistic relationship between multiple indicators. This can unify the data benchmark and avoid the failure of aging analysis and control logic caused by the severing of the linkage relationship between indicators by independent single-dimensional clustering.

[0085] In some implementations, the index data at each acquisition time are normalized and combined into a multidimensional coordinate system, i.e., data points (x1, x2, x3, ..., x4) in a multidimensional space. s ), where s is the number of indicators.

[0086] For example, if three indicators are monitored and their normalized values ​​at a certain moment are 0.5, 0.3, and 0.6 respectively, then the data point at that moment is the coordinates (0.5, 0.3, 0.6) in three-dimensional space.

[0087] By combining the initial values ​​of the correlation between indicators in each dimension (such as historical correlation between indicators), multidimensional data is projected onto a space of indicator correlation, which intuitively shows the clustering / dispersion of the data distribution in that dimension.

[0088] For example, if there are data points at three time points in the multidimensional sample space: time 1 (0.5, 0.3, 0.6), time 2 (0.6, 0.4, 0.7), and time 3 (0.4, 0.2, 0.5), and the correlation coefficient of the first dimension is 0.3, the correlation coefficient of the second dimension is 0.2, and the correlation coefficient of the third dimension is 0.5, then by combining the correlation coefficients, the coordinates of the data points in the multidimensional sample space are mapped to the coordinates of the data points in the corresponding single-dimensional sample space for each of the three dimensions: coordinates = (indicator of that dimension in the multidimensional sample space) × (1 + correlation coefficient of that dimension). Specifically, this includes:

[0089] First dimension: Single-dimensional coordinate at time 1 = 0.5 × (1 + 0.3) = 0.65; Single-dimensional coordinate at time 2 = 0.6 × (1 + 0.3) = 0.78; Single-dimensional coordinate at time 3 = 0.4 × (1 + 0.3) = 0.52;

[0090] Second dimension: Single-dimensional coordinate at time 1 = 0.3 × (1 + 0.2) = 0.36; Single-dimensional coordinate at time 2 = 0.4 × (1 + 0.2) = 0.48; Single-dimensional coordinate at time 3 = 0.2 × (1 + 0.2) = 0.24;

[0091] The third dimension: the single-dimensional coordinate of time 1 = 0.6 × (1 + 0.5) = 0.9; the single-dimensional coordinate of time 2 = 0.7 × (1 + 0.5) = 1.05; the single-dimensional coordinate of time 3 = 0.5 × (1 + 0.5) = 0.75.

[0092] S22. Determine the clustering index of the single-dimensional data points within each cluster based on the number of single-dimensional data points in each cluster and the distance between the single-dimensional data points and the cluster center.

[0093] Among them, the clustering index of single-dimensional data points characterizes the degree of clustering of indicator data, quantifies the degree of clustering of data points within each cluster, and the higher the clustering index, the more concentrated the data distribution and the better the stability of the indicator.

[0094] Specifically, the formula for calculating the clustering index is:

[0095]

[0096] in, Let be the clustering index of the j-th cluster;

[0097] Let be the number of data points i within the j-th cluster. The larger the numerator, the more data points the cluster covers.

[0098] is the farthest distance between a data point within the j-th cluster and the cluster center;

[0099] is the average distance between data points within the j-th cluster and the cluster center;

[0100] This indicates the uniformity of the distance from the data points in the j-th cluster to the cluster center. The more uniform the distribution, the higher the degree of clustering.

[0101] It is the product of the uniformity of distance and the spatial range of the cluster. The smaller the value, the smaller the spatial range of the cluster, or the higher the uniformity of data distribution, or both. Both of these situations indicate that the data points are more concentrated within the cluster.

[0102] The overall formula comprehensively quantifies the degree of clustering by using the ratio of the numerator (cluster size) to the denominator (distribution concentration). A larger clustering index indicates a more compact and uniform distribution of data points, indicating a higher degree of clustering; conversely, a smaller clustering index indicates a lower degree of clustering.

[0103] S23. Determine the outlier index of the single-dimensional data points in each cluster based on the distance between the single-dimensional data points and the cluster center, and the degree of aggregation of the single-dimensional data points.

[0104] Among them, the outlier index of a single-dimensional data point represents the degree of outlier in the index data. It can be used to identify outliers (such as abnormal data) within a cluster. The higher the outlier index, the more significantly the data point deviates from the cluster center, and the more likely it is to be noise or an outlier.

[0105] Specifically, the formula for calculating the outlier index is:

[0106]

[0107] Let be the outlier index of the i-th data point;

[0108] The distance between the i-th data point and the cluster center within its own cluster (the j-th cluster);

[0109] This represents the average distance between multiple data points within the cluster (the j-th cluster) to which the i-th data point belongs and the cluster center.

[0110] This is the clustering index of the cluster to which the i-th data point belongs (the j-th cluster);

[0111] This represents the absolute value of the deviation between the distance of the i-th data point from the cluster center and the average distance within the cluster. The larger this value is, the farther the data point deviates from the average distance within the cluster, reflecting the anomaly of the data point from the distance dimension.

[0112] The overall formula quantifies the outlier degree of a data point by dividing the deviation of the data point from the average distance within the cluster by the cluster's clustering index. The larger the outlier index, the more the data point deviates from the average distance within the cluster and is located in a highly clustered cluster, indicating a higher degree of outlierness; conversely, the smaller the outlier index, the lower the degree of outlierness.

[0113] Based on the above technical solution, the data is first mapped to a multi-dimensional space and then decomposed into one-dimensional clusters. This avoids the problem of directly severing the correlation between indicators in one-dimensional clusters, and allows for refined analysis of the characteristics of each indicator. Then, through clustering and outlier indices, the qualitative description of data distribution clustering / dispersion is transformed into calculable quantitative indicators, providing an objective basis for subsequent aging analysis and control.

[0114] In one possible implementation, the method in S21 above for clustering single-dimensional data points in a single-dimensional sample space to obtain multiple clusters corresponding to each indicator can be specifically implemented through the following S211, which will be explained in detail below:

[0115] S211. Using the iterative self-organizing data analysis techniques algorithm (ISODATA) and preset clustering parameters, cluster the single-dimensional data points in the single-dimensional sample space to obtain multiple clusters.

[0116] Water conservancy monitoring data (such as water level, flow velocity, and uplift pressure) are affected by natural factors (rainfall, tides) and human factors (sluice gate scheduling), resulting in complex and variable data distribution patterns. The ISODATA algorithm can automatically increase or decrease the number of clusters based on the actual distribution of the data. For example, during the high-water season, water flow conditions vary, potentially exhibiting high, medium, and low flow velocities; the ISODATA algorithm can automatically identify and classify multiple different velocity clusters. Conversely, during the low-water season, water flow conditions are relatively uniform, and the algorithm automatically reduces the number of clusters to avoid over-clustering. Compared to clustering algorithms with a fixed number of clusters, ISODATA better reflects the dynamic characteristics of water conservancy data.

[0117] In some implementations, such as Figure 7 As shown, the ISODATA algorithm is used to cluster data points in a one-dimensional sample space, with an initial number of clusters of 6 and an initial number of samples in each cluster. Where n is the number of data points in the current dimension, and the maximum number of iterations is 5. The optimal number of clusters is obtained based on the data distribution in the current dimension, thus achieving similar clustering levels among data points within the same cluster. .

[0118] Among them, water conservancy index data usually cover three basic states: "low, medium and high". Each state can be further subdivided into two sub-states (such as the stable / fluctuating state under high water level). Six initial clusters can cover most scenarios and avoid data mixing within clusters due to too few initial clusters.

[0119] Water conservancy data distribution is relatively stable, and 5 iterations are sufficient for cluster center convergence. After 3 or 4 iterations, the change in cluster center is usually <0.01. Too many iterations will increase the computation time and do not meet the requirements of real-time monitoring.

[0120] Based on the above technical solution, the ISODATA algorithm supports dynamic adjustment of the number of clusters, which can adapt to the multimodal distribution characteristics of water conservancy data, more accurately classify data categories, and avoid feature confusion caused by a single cluster class covering multiple states of data.

[0121] In one possible implementation, the method described in S3 above for comparing the data distribution across multiple monitoring periods and analyzing the degree of decline of each indicator can be specifically implemented through the following S31 to S33, which are explained in detail below:

[0122] S31. Normalize the outlier index of the single-dimensional data points in each monitoring period, and mark the single-dimensional data points whose normalization result is greater than the preset threshold as outliers.

[0123] Because some extreme cases exist that cause the data measured at a certain moment to differ significantly from the overall situation, such as the large instantaneous flow rate when the gate is first opened, which causes a sudden increase in the measured water flow velocity and an increase in the uplift pressure on the gate, the corresponding data point deviates from the distribution of the overall data. Therefore, outliers are screened by analyzing the distribution of data points within the cluster. That is, all points in the current dimension are traversed. The greater the degree of deviation of the current point from the center of its cluster and the worse the cluster's aggregation, the greater the degree of outlier the current point.

[0124] Normalization is performed to unify the quantification standard of dimensions. Normalization is usually performed to (0,1), and points with a normalization result greater than 0.5 are marked as outliers.

[0125] Outliers often correspond to instantaneous extreme operating conditions or minor equipment malfunctions (such as slight sensor drift or sudden pressure changes caused by local cracks in the gate body). Simply cleaning up these outliers will result in the loss of pre-fault warning information. By retaining and analyzing outliers, early warnings can be achieved. For example, a sudden appearance of a few outliers in the pressure during a certain cycle (a short-term pressure surge) could be an early signal of leakage at the bottom of the gate, providing an alarm several days to weeks earlier than traditional methods that wait for the fault to escalate. Furthermore, by combining the time and indices associated with outliers (such as outliers accompanied by abnormal flow velocity), the type of fault can be identified (e.g., abnormal pressure and flow velocity indicate wear in the gate body's flow channels).

[0126] S32. Based on the clustering index and number of outliers of single-dimensional data points within multiple clusters corresponding to each indicator in two consecutive monitoring periods, as well as the number of clusters of multiple clusters, determine the degree of fluctuation of each indicator in two consecutive monitoring periods.

[0127] like Figure 8 As shown, the distribution of indicators differs across monitoring periods. By comparing the clustering characteristics (clustering index, outliers, and number of clusters) of consecutive periods, the changes in indicators from stable to fluctuating can be quantified, allowing for early detection of early signals of sluice gate performance degradation.

[0128] Specifically, the formula for calculating the degree of indicator volatility is as follows:

[0129]

[0130] The index represents the fluctuation of a single-dimensional indicator in the T-th and T+1-th monitoring periods. It is used to characterize the degree of fluctuation of the indicator in a single dimension. The larger the index, the higher the degree of fluctuation of the indicator.

[0131] It represents the mean of the aggregation index of a single-dimensional indicator data in the (T+1)th monitoring period;

[0132] The mean of the aggregation index of the single-dimensional indicator data in the T-th monitoring period;

[0133] This represents the number of clusters in a single dimension during the (T+1)th monitoring period.

[0134] The number of single-dimensional clusters in the T-th monitoring period;

[0135] This represents the number of outliers in the (T+1)th monitoring period.

[0136] This represents the difference in the clustering index between the (T+1)th monitoring period and the Tth monitoring period. If the ratio is less than 1, it indicates that the clustering degree of the data in the (T+1)th period has decreased and the volatility of the indicator has tended to increase. If the ratio is greater than 1, it indicates that the clustering degree has increased and the volatility of the indicator has tended to decrease.

[0137] Based on the ISODATA algorithm, which addresses the characteristic that the number of clusters in different sample spaces is not fixed, the difference in the number of clusters in a single dimension is compared between two consecutive monitoring periods. The more clusters there are, the lower the degree of aggregation.

[0138] This represents the combined relationship between the number of clusters and the number of outliers.

[0139] The overall formula comprehensively quantifies the volatility of a single-dimensional indicator over two consecutive monitoring periods by multiplying the relative change in the mean of the clustering index with the combined relative changes in the number of clusters and outliers. A larger index indicates a more significant decrease in clustering, an increase in cluster dispersion, and a greater number of outliers from period T to period T+1, signifying higher volatility. Conversely, a smaller index indicates lower volatility.

[0140] S33. Determine the degree of decline of each indicator based on the aggregation index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods and the degree of indicator fluctuation of each indicator.

[0141] Specifically, the formula for calculating the degree of recession is:

[0142]

[0143] The recession index is the recession index of the k-th indicator, which is used to characterize the degree of recession of the k-th indicator. The larger the recession index, the higher the degree of recession.

[0144] The maximum aggregation index of the k-th indicator across multiple monitoring periods;

[0145] It represents the minimum clustering index of the k-th indicator across multiple monitoring periods;

[0146] The time interval between the monitoring period corresponding to the maximum value of the aggregation index (maximum aggregation index period) and the monitoring period corresponding to the minimum value of the aggregation index (minimum aggregation index period);

[0147] For the k-th indicator in multiple monitoring periods Mean;

[0148] This indicates the fluctuation range of the clustering index of the indicator over multiple periods. The larger this difference is, the greater the change in the indicator from stable to unstable. From the perspective of clustering, it reflects the potential for decline in the indicator's performance. If the indicator originally maintained high clustering (stability), but the clustering decreased significantly later (instability), the difference will be large, and the decline will be more significant.

[0149] This is to combine the magnitude of long-term clustered changes with the degree of fluctuation during the week. If the indicator not only has a large magnitude of clustered changes, but also fluctuates frequently during the week, the value of the numerator will be larger, thus reflecting the decline characteristics of the indicator from both the magnitude and frequency dimensions.

[0150] The overall formula quantifies the degree of decline of an indicator by dividing the numerator (the combined magnitude of clustered changes and the degree of cyclical fluctuations) by the denominator (the time interval between changes). The larger the decline index, the greater the degree of decline, indicating that the indicator either has a large magnitude of clustered changes and frequent cyclical fluctuations, or changes rapidly; conversely, the smaller the decline index, the lower the degree of decline.

[0151] Based on the above technical solution, outlier points are used to mine abnormal signals, and then the long-term trends of indicators are integrated to analyze the degree of indicator fluctuation, thereby enhancing the sensitivity of aging assessment and improving the accuracy of sluice gate aging analysis.

[0152] In one possible implementation, the method for analyzing the correlation between each indicator and other indicators in S3 above can be specifically implemented through the following S34, which will be explained in detail below:

[0153] S34. Determine the correlation degree of indicators based on the clustering index of single-dimensional data points in multiple clusters corresponding to each indicator in multiple monitoring periods, the degree of decay of each indicator, the number of indicators in multiple indicators, and the number of periods in multiple monitoring periods.

[0154] Specifically, the formula for calculating the correlation between indicators is as follows:

[0155]

[0156] The correlation degree of indicators is used to characterize the degree of influence of an indicator on other indicators.

[0157] The mean of the aggregation index of the single-dimensional indicator data in the T-th monitoring period;

[0158] The mean of the aggregation index of the k-th indicator in the T-th monitoring period;

[0159] This represents the average decline index of multiple indicators during the current monitoring period.

[0160] This is the decline index of the indicator in the current monitoring period;

[0161] The number of indicators monitored for the sluice gate;

[0162] The number of monitoring periods;

[0163] It represents the absolute value of the deviation between the mean of the clustering index and the mean of the clustering index of a single-dimensional indicator in that period, reflecting the degree of deviation of the clustering of the indicator in a period from the overall average clustering.

[0164] It represents the total deviation of the indicator from the overall clustering over all periods. The larger the total deviation, the greater the difference between the clustering change of the indicator and the overall clustering change.

[0165] It represents the absolute value of the deviation of the recession index from the mean of all recession indices in the current period relative to a certain benchmark (such as the historical mean of the recession index or the overall mean). If the recession degree of a certain indicator is much higher or lower than the overall mean, it indicates that it has a special performance in terms of recession and may have a more significant impact on other indicators.

[0166] The overall formula comprehensively quantifies the influence of an indicator on other indicators by multiplying the denominator (the reciprocal of the total deviation of the multi-indicator, multi-period clustering) and the numerator (the relative deviation ratio of the decline degree of a single indicator). The greater the correlation between indicators, the higher the influence of that indicator on other indicators in terms of the overall consistency of clustering and the relative anomaly of the decline degree; conversely, the smaller the correlation between indicators, the lower the influence.

[0167] Furthermore, the correlation degree of the indicators can be normalized to (0,1) and used as the weight of the indicators.

[0168] Based on the above technical solutions, traditional correlation coefficients (such as Pearson correlation coefficient) are based only on the linear relationship of current data and do not take into account the gradual change in the relationship between indicators caused by the aging of sluice gates. This solution incorporates historical data from multiple monitoring cycles and the degree of decline of indicators, so that the correlation coefficient can reflect the trend of coordinated changes between indicators during the aging process, which is more in line with the actual long-term operation of water conservancy facilities.

[0169] In one possible implementation, the method of S4 described above can be specifically implemented through the following S41 to S42, which are explained in detail below:

[0170] S41. Based on the indicator data of the current monitoring period, the aggregation index of single-dimensional data points in multiple clusters corresponding to each indicator, the degree of decline of each indicator and the correlation of indicators, determine the degree of regulation to be implemented for the sluice gate opening.

[0171] Specifically, the formula for calculating the degree of regulation is as follows:

[0172]

[0173] The control index is used to characterize the degree to which the sluice gate opening needs to be controlled. The larger the control index, the higher the degree to which it needs to be controlled.

[0174] The normalized result of the correlation degree of the k-th indicator is used as the weight of the k-th indicator;

[0175] The number of indicators monitored for the sluice gate;

[0176] This represents the decline index of the k-th indicator in the current monitoring period.

[0177] This represents the average value of the k-th indicator data in the current monitoring period.

[0178] The aggregation index of the k-th indicator in the current monitoring period is the average value. The higher the aggregation degree, the better the current sluice gate's ability to regulate water flow, and the less regulation is needed.

[0179] The overall formula combines the weight of each indicator with the combined effects of decay, data, and aggregation to integrate the aging status, current level, and distribution stability of multiple indicators into a single control index. This approach considers the differences in importance of different indicators while also taking into account the adaptability of sluice gate aging and the stability of water flow. It upgrades the opening control from a simple logic of only looking at the current water level to a multi-dimensional, full-lifecycle intelligent decision-making process, avoiding control failure due to neglecting aging.

[0180] Furthermore, the control index can be normalized to [-1,1]. The larger the control index, the greater the deviation between the current state of the sluice gate and the optimal operating state, and the higher the degree of control required.

[0181] S42. Determine the pre-regulation opening degree based on the sluice gate opening degree and the degree of regulation to be adjusted during the current monitoring period.

[0182] Specifically, the formula for calculating the pre-controlled opening degree is as follows:

[0183]

[0184] h represents the pre-regulation opening degree of the sluice gate in the next continuous monitoring cycle;

[0185] This refers to the sluice gate opening degree during the current monitoring period.

[0186] Q is the normalized result of the control index of the sluice gate opening.

[0187] The overall formula combines the base opening with the control demand by multiplying the current opening by (1 + control amplitude) to obtain the pre-control opening for the next cycle.

[0188] Based on the above technical solution, by incorporating the degree of decay, the control intensity can be matched with the actual performance of the sluice gate. At the same time, by combining the correlation of indicators to quantify the synergistic effect between indicators, the one-sided control caused by the dominance of a single indicator is avoided, thus realizing the precision, adaptability and aging adaptation of the sluice gate opening control.

[0189] In one possible implementation, the method of S51 described above can be specifically implemented through the following S511, which will be explained in detail below:

[0190] S511. Based on the indicator data, indicator correlation, pre-regulation opening degree, and the real-time sluice gate opening degree at the target acquisition time, determine the real-time sluice gate opening degree at the target acquisition time.

[0191] Specifically, the formula for calculating the real-time sluice gate opening is as follows:

[0192]

[0193] The real-time sluice gate opening at the w-th data collection time;

[0194] This refers to the index data of the k-th indicator at the w-th data collection time.

[0195] This is the normalized result of the correlation degree of the k-th indicator;

[0196] The number of indicators monitored for the sluice gate;

[0197] Pre-adjustment of the sluice gate opening;

[0198] The real-time sluice gate opening at the (w-1)th data collection time;

[0199] By weighting the correlation between indicators and summing the indicator data, the real-time water flow status of multiple indicators is integrated into a single comprehensive value of water flow change. Indicators with high importance (such as uplift pressure related to gate safety) will have a greater impact on this value, ensuring that anomalies in key indicators are responded to first. For example, when uplift pressure rises sharply, the weighted sum will change significantly, prompting adjustments to the gate opening to ensure gate safety.

[0200] This represents the normalized result of the water flow change at the w-th data collection time. It can eliminate the interference of different indicator dimensions on the calculation, so that the comprehensive value of water flow change can be directly used to calculate the opening adjustment range, ensuring that the dimensions of each part of the formula are consistent and the logic is coherent. The normalized result belongs to the interval [-1,1], and can be positive or negative. If the comprehensive change of the indicator requires a decrease in the opening, this value is negative; if the opening needs to be increased, this value is positive.

[0201] The overall formula integrates long-term forecast targets with instantaneous water flow changes and historical state constraints by predicting the opening degree × (1 + normalized adjustment range of water flow change × previous sluice gate opening degree). This upgrades the sluice gate opening degree from experience-based static scheduling to data-driven dynamic response, ensuring both long-term scheduling targets and precise responses to instantaneous water flow changes, ultimately improving the operational efficiency and safety of the water conservancy system.

[0202] Based on the above technical solution, under the basic framework of pre-regulation opening degree, dynamic correction is achieved by combining real-time indicator data, indicator correlation degree, and historical opening degree, so as to solve the problems of lag and insufficient scenario adaptation that may exist in pre-regulation opening degree.

[0203] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0204] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0205] In this embodiment of the invention, the intelligent water storage device with automatic adjustment function can be divided into functional units according to the above method example. For example, each function can be divided into different functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or software. It should be noted that the unit division in this embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0206] This invention also provides a schematic diagram of the hardware structure of an intelligent water storage and regulation device with automatic adjustment function, see below. Figure 9 The intelligent water storage device 900 with automatic adjustment function includes a processor 901, and optionally, a memory 902 connected to the processor 901.

[0207] In the first possible implementation, see Figure 9 The intelligent water storage and regulation device 900 with automatic adjustment function also includes a transceiver 903. The processor 901, memory 902, and transceiver 903 are connected via a bus. The transceiver 903 is used to communicate with other devices or communication networks. Optionally, the transceiver 903 may include a transmitter and a receiver. The device in the transceiver 903 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of the present invention. The device in the transceiver 903 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of the present invention.

[0208] Based on the first possible implementation method Figure 9 The structural diagram shown can be used to illustrate the structure of the intelligent water storage device with automatic adjustment function involved in the above embodiments.

[0209] in, Figure 9The diagram can also illustrate the system chip in an intelligent water storage device with automatic adjustment function. In this case, the actions performed by the aforementioned intelligent water storage device with automatic adjustment function can be implemented by this system chip. The specific actions performed can be found above and will not be repeated here.

[0210] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in this embodiment can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0211] The processor in this invention may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a standalone semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a System-on-a-Chip (SoC), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), or logic circuits that implement dedicated logic operations.

[0212] The memory in the embodiments of the present invention may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable read-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0213] This invention also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0214] This invention also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0215] This invention also provides a chip, which includes a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0216] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0217] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings and the disclosure, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In this invention, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several of the functions listed in this invention.

[0218] Although the invention has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made therein without departing from the spirit and scope of the invention. Accordingly, this specification and drawings are merely illustrative of the invention and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if such modifications and modifications of the invention fall within the scope of the invention and its equivalents, the invention is also intended to include such modifications and modifications.

Claims

1. A smart water control method with automatic adjustment function, characterized in that, include: Data on multiple indicators of the monitored sluice gate are acquired at multiple data collection points during the monitoring cycle. The indicator data is mapped to a multidimensional sample space to analyze the data distribution of the indicator data; each indicator corresponds to one dimension of the multidimensional sample space. By comparing the data distribution across multiple monitoring periods, we can analyze the degree of decline of each indicator and the correlation between each indicator and other indicators. Based on the indicator data of the current monitoring period, the data distribution, the degree of decline of each indicator, and the correlation of the indicators, the sluice gate opening degree of the current monitoring period is adjusted to obtain the pre-regulation opening degree for the next monitoring period. Based on the pre-controlled opening degree, the sluice gate opening degree for the next monitoring cycle is adjusted; The data distribution includes: the degree of clustering and outliers of the indicator data; the analysis of the data distribution of the indicator data includes: The indicator data is mapped to the multidimensional data points in the multidimensional sample space, mapped to the single-dimensional sample space corresponding to each indicator, and the single-dimensional data points in the single-dimensional sample space are clustered to obtain multiple clusters corresponding to each indicator. The clustering index of the single-dimensional data points within each cluster is determined based on the number of single-dimensional data points in each cluster and the distance between the single-dimensional data points and the cluster center; the clustering index of the single-dimensional data points characterizes the degree of clustering of the indicator data. Based on the distance between the single-dimensional data point and the cluster center within each cluster, and the degree of clustering of the single-dimensional data point, the outlier index of the single-dimensional data point within each cluster is determined; the outlier index of the single-dimensional data point characterizes the degree of outlier of the indicator data. The comparison of data distribution across multiple monitoring periods and the analysis of the degree of decline for each indicator include: The outlier index of the single-dimensional data points in each monitoring period is normalized, and single-dimensional data points whose normalization results are greater than a preset threshold are marked as outliers. The degree of fluctuation of each indicator in the two consecutive monitoring periods is determined based on the clustering index of single-dimensional data points in multiple clusters corresponding to each indicator in the two consecutive monitoring periods, the number of outliers, and the number of clusters in the multiple clusters. The degree of decline of each indicator is determined based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods and the degree of fluctuation of each indicator. The analysis of the correlation between each indicator and other indicators includes: The correlation degree of the indicators is determined based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods, the degree of decay of each indicator, the number of indicators of the multiple indicators, and the number of periods of the multiple monitoring periods. The step of adjusting the sluice gate opening degree for the current monitoring period based on the indicator data of the current monitoring period, the data distribution, the degree of decline of each indicator, and the correlation degree of the indicators, to obtain the pre-regulation opening degree for the next monitoring period, includes: Based on the indicator data of the current monitoring period, the aggregation index of single-dimensional data points within multiple clusters corresponding to each indicator, the degree of decay of each indicator, and the correlation of the indicators, the degree of regulation required for the sluice gate opening is determined. The pre-regulation opening degree is determined based on the sluice gate opening degree of the current monitoring period and the degree to be regulated.

2. The intelligent water control method according to claim 1, characterized in that, The adjustment of the sluice gate opening in the next monitoring cycle based on the pre-adjusted opening degree includes: Based on the indicator data acquired at each collection moment in the next monitoring cycle and the correlation degree of the indicators, the pre-regulation opening degree is adjusted to obtain the real-time sluice gate opening degree at each collection moment.

3. The intelligent water control method according to claim 1, characterized in that, The clustering of single-dimensional data points in the single-dimensional sample space yields multiple clusters corresponding to each indicator, including: By iterating through self-organizing data to determine the algorithm and pre-defined clustering parameters, the single-dimensional data points in the single-dimensional sample space are clustered to obtain the multiple clusters.

4. The intelligent water conservancy control method according to claim 1, characterized in that, The step of adjusting the pre-regulation opening degree based on the indicator data acquired at each collection time in the next monitoring cycle and the correlation degree of the indicators to obtain the real-time sluice gate opening degree at each collection time includes: The real-time sluice gate opening at the target acquisition time is determined based on the indicator data obtained at the target acquisition time, the correlation degree of the indicator, the pre-regulation opening degree, and the real-time sluice gate opening degree at the previous acquisition time.

5. The intelligent water control method according to any one of claims 1-4, characterized in that, The multiple indicators include at least: upstream actual water level, downstream actual water level, sluice gate uplift pressure, upstream water flow velocity, and downstream water flow velocity data.

6. An intelligent water storage and regulation device with automatic adjustment function, characterized in that, include: Data acquisition module, data processing module, and control module; The data acquisition module is used to acquire indicator data of multiple indicators of the monitored sluice gate at multiple acquisition times during the monitoring cycle. The data processing module is used to map the indicator data to a multidimensional sample space and analyze the data distribution of the indicator data; each indicator corresponds to one dimension of the multidimensional sample space. The data processing module is also used to compare the data distribution of multiple monitoring periods, analyze the degree of decline of each indicator, and analyze the correlation between each indicator and other indicators. The data processing module is also used to adjust the sluice gate opening degree of the current monitoring period based on the indicator data of the current monitoring period, the data distribution, the degree of decline of each indicator and the correlation degree of the indicators, so as to obtain the pre-regulation opening degree of the next monitoring period. The control module is used to control the sluice gate opening degree in the next monitoring cycle based on the pre-controlled opening degree. The data distribution includes: the degree of clustering and outliers of the indicator data; the analysis of the data distribution of the indicator data includes: The indicator data is mapped to the multidimensional data points in the multidimensional sample space, mapped to the single-dimensional sample space corresponding to each indicator, and the single-dimensional data points in the single-dimensional sample space are clustered to obtain multiple clusters corresponding to each indicator. The clustering index of the single-dimensional data points within each cluster is determined based on the number of single-dimensional data points in each cluster and the distance between the single-dimensional data points and the cluster center; the clustering index of the single-dimensional data points characterizes the degree of clustering of the indicator data. Based on the distance between the single-dimensional data point and the cluster center within each cluster, and the degree of clustering of the single-dimensional data point, the outlier index of the single-dimensional data point within each cluster is determined; the outlier index of the single-dimensional data point characterizes the degree of outlier of the indicator data. The comparison of data distribution across multiple monitoring periods and the analysis of the degree of decline for each indicator include: The outlier index of the single-dimensional data points in each monitoring period is normalized, and single-dimensional data points whose normalization results are greater than a preset threshold are marked as outliers. The degree of fluctuation of each indicator in the two consecutive monitoring periods is determined based on the clustering index of single-dimensional data points in multiple clusters corresponding to each indicator in the two consecutive monitoring periods, the number of outliers, and the number of clusters in the multiple clusters. The degree of decline of each indicator is determined based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods and the degree of fluctuation of each indicator. The analysis of the correlation between each indicator and other indicators includes: The correlation degree of the indicators is determined based on the clustering index of single-dimensional data points within multiple clusters corresponding to each indicator in multiple monitoring periods, the degree of decay of each indicator, the number of indicators of the multiple indicators, and the number of periods of the multiple monitoring periods. The step of adjusting the sluice gate opening degree for the current monitoring period based on the indicator data of the current monitoring period, the data distribution, the degree of decline of each indicator, and the correlation degree of the indicators, to obtain the pre-regulation opening degree for the next monitoring period, includes: Based on the indicator data of the current monitoring period, the aggregation index of single-dimensional data points within multiple clusters corresponding to each indicator, the degree of decay of each indicator, and the correlation of the indicators, the degree of regulation required for the sluice gate opening is determined. The pre-regulation opening degree is determined based on the sluice gate opening degree of the current monitoring period and the degree to be regulated.

Images