A dynamic calculation method of electrical water reduction coefficient based on multi-dimensional data fusion

Abstract

The application relates to the technical field of data processing, in particular to a dynamic calculation method of electricity-to-water conversion coefficients based on multi-dimensional data fusion, which comprises the following steps: obtaining electricity-to-water conversion coefficients of each period in history and monitoring data of each dimension, and then obtaining the influence degree of each dimension; obtaining the overall stable level of each dimension in the current period based on the monitoring data of each dimension in the current period; obtaining the gentle change level of one dimension in the current period according to the difference between each two adjacent monitoring data of the dimension; obtaining the regular fluctuation degree of one dimension in the current period by using the wave peak feature of the monitoring data of the dimension; and then obtaining the stable degree evaluation value of each dimension in the current period; and obtaining the optimized period and calculating the electricity-to-water conversion coefficient of the optimized period by using the fixed period length, the stable degree evaluation value of each dimension in the current period and the influence degree of each dimension. The application optimizes the period for calculating the electricity-to-water conversion coefficient, and improves the monitoring accuracy.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a method for dynamically calculating the electricity-to-water conversion coefficient based on multidimensional data fusion. Background Technology

[0002] The electricity-to-water conversion factor is a commonly used indicator in water supply, irrigation districts, water plants, pumping stations, and hydropower co-operation scenarios. It represents the water supply or transmission volume corresponding to a unit of electricity consumption, reflecting the relationship between electricity consumption and water delivery, and is an energy efficiency evaluation indicator. It can assess the overall operating efficiency of pump sets, motors, and water transmission networks, and promptly identify energy efficiency declines caused by equipment aging, impeller wear, pipe blockage, increased leakage, or abnormal operating conditions. Furthermore, by comparing and analyzing the electricity-to-water conversion factor under different time periods, different pump sets, and different operating conditions, it can also identify high-energy-consuming operating states, optimize equipment start-up and shutdown strategies and water transmission scheduling schemes, improve the energy-saving operation level of the water supply system, and reduce the unit water supply cost.

[0003] Current technologies typically calculate the electricity-to-water conversion factor based on electricity consumption and cumulative flow over a period of time. This usually involves using a fixed timeframe (e.g., one hour) ranging from five minutes to two hours to calculate the factor. However, there is no fixed, applicable length for this timeframe in practice. An excessively long timeframe may overlook details of energy efficiency variations in the water supply and pumping station systems; an excessively short timeframe may result in excessive noise from normal data fluctuations in the calculated electricity-to-water conversion factor. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a dynamic calculation method for the electro-water conversion coefficient based on multi-dimensional data fusion. The specific technical solution adopted is as follows:

[0005] One embodiment of this application provides a method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion. The method includes:

[0006] Obtain historical data on the electricity-to-water conversion factor and monitoring data for each dimension; take the mean of the monitoring data for one dimension in a given period as the monitoring value for that dimension in that period; and determine the degree of influence of each dimension based on the correlation between the historical monitoring values ​​for each dimension and the changes in the electricity-to-water conversion factor in each period.

[0007] Based on the monitoring data of each dimension in the current period, obtain the overall stability level of each dimension in the current period; based on the difference between every two adjacent monitoring data of a dimension in the current period, obtain the level of smooth change of that dimension in the current period; utilize the peak characteristics of the monitoring data of a dimension in the current period to obtain the degree of regular fluctuation of that dimension in the current period.

[0008] The overall stability level, gradual change level, and regular fluctuation degree of one dimension in the current period are integrated to obtain the stability assessment value of that dimension in the current period; the optimized period is obtained by using a fixed period length, the stability assessment values ​​of each dimension in the current period, and the influence degree of each dimension, and the electricity-to-water conversion factor of the optimized period is calculated.

[0009] Preferably, the degree of influence of each dimension is obtained based on the correlation between the monitoring values ​​of each dimension in different historical periods and the changes in the electricity-to-water conversion factor in each period, including:

[0010] The absolute value of the difference between the monitoring values ​​of every two adjacent time periods in history for a dimension is obtained as the monitoring difference for every two adjacent time periods in that dimension, and these differences are arranged in chronological order to form a monitoring difference sequence for that dimension. The difference between the electricity-to-water conversion factor of the subsequent time period and the preceding time period in history for every two adjacent time periods is obtained, and these differences are arranged in chronological order to form a difference sequence of electricity-to-water conversion factor. The monitoring differences in the monitoring difference sequence for that dimension that are greater than the mean of the monitoring difference sequence are recorded as changes in monitoring differences. The differences in the difference sequence of electricity-to-water conversion factor that are greater than or less than 0 are recorded as changes in differences. The number of changes in monitoring differences and changes in differences that are in the same position in the time sequence is obtained, and the number of changes in monitoring differences is divided by the number of changes in monitoring differences to obtain the degree of influence of that dimension.

[0011] Preferably, the overall stability level of each dimension in the current time period is obtained based on the monitoring data of each dimension in the current time period, including:

[0012] The overall stability level of that dimension in the current period is obtained by taking the reciprocal of the difference between the maximum and minimum values ​​of the monitoring data for a certain dimension in the current period and normalizing it.

[0013] Preferably, the level of gradual change in a dimension during the current time period is obtained based on the difference between every two adjacent monitoring data points for that dimension, including:

[0014] The absolute difference between every two adjacent monitoring data points in one dimension during the current time period is used to form a sequence of absolute difference values. The reciprocal of the mean of the absolute difference between each absolute difference value in the sequence and the mean of the sequence is obtained and normalized to obtain the level of gradual change in that dimension during the current time period.

[0015] Preferably, the degree of regular fluctuation in that dimension of the monitoring data during the current time period is obtained by utilizing the peak characteristics of the data during that time period, including:

[0016] Connect the monitoring data of one dimension in the current period into a curve and obtain the peak points; take the horizontal axis distance between two adjacent trough points of a peak point as the horizontal axis span corresponding to that peak point; multiply the absolute value of the difference between the horizontal axis spans corresponding to one peak point and the absolute value of the difference in amplitude of the two peak points to obtain the difference characteristic value of the two peak points; obtain the mean of the difference characteristic values ​​of one peak point and other peak points as the average difference characteristic value corresponding to that peak point; calculate the reciprocal of the mean of the average difference characteristic values ​​corresponding to each peak point and normalize it to obtain the degree of regular fluctuation of that dimension in the current period.

[0017] Preferably, the overall stability level, the level of gradual change, and the degree of regular fluctuation in one dimension of the current time period are fused to obtain the stability assessment value of that dimension for the current time period, including:

[0018] The stability assessment value of each dimension in the current period is obtained by averaging the overall stability level, the level of gradual change, and the degree of regular fluctuation in each dimension.

[0019] Preferably, the optimized time period is obtained by using a fixed time period length, the stability assessment values ​​of each dimension of the current time period, and the influence degree of each dimension, and the electricity-to-water conversion factor of the optimized time period is calculated, including:

[0020] The weighted average of the stability assessment values ​​of each dimension in the current period is obtained by using the degree of influence of each dimension as the weight of the stability assessment value of each dimension in the current period. The weighted average is then multiplied by a fixed period length to obtain the adjusted period length. The current period is then evenly divided into each optimized period using the adjusted period length. The electricity-to-water conversion factor for each optimized period is calculated based on the electricity consumption and water delivery of each optimized period.

[0021] The embodiments of this application have at least the following beneficial effects: This application obtains the electricity-to-water conversion coefficient and monitoring data of various dimensions for different historical time periods, and then analyzes and obtains the degree of influence of each dimension; then, based on the monitoring data of each dimension in the current time period, it analyzes and obtains the overall stability level, the level of gradual change, and the degree of regular fluctuation of each dimension in the current time period, and then integrates them to obtain the stability evaluation value of each dimension in the current time period, obtains the real instability risk that the electricity-to-water conversion coefficient may exist in the current time period, and then, combined with the degree of influence of each dimension, obtains the optimized time period and calculates the electricity-to-water conversion coefficient of the optimized time period, and obtains a more accurate calculation result of the electricity-to-water conversion coefficient; at the same time, in the process of constructing the stability evaluation model, it combines the analysis of multiple water supply modes of the electric water supply system under the electricity-to-water conversion scenario to obtain the overall stability level, the level of gradual change, and the degree of regular fluctuation, obtains a more accurate reflection of the actual existence of fluctuations in the electricity-to-water conversion coefficient, and improves the accuracy of the instability representation in the stability model evaluation result, which requires high-frequency calculation of the electricity-to-water conversion coefficient. Attached Figure Description

[0022] To more clearly illustrate the technical solutions and advantages in the embodiments of this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A first form diagram of the stability of a dynamic calculation method for the electro-water conversion coefficient based on multi-dimensional data fusion is provided for an embodiment of this application.

[0024] Figure 2 A second form diagram of the stability of a dynamic calculation method for the electro-water conversion coefficient based on multi-dimensional data fusion is provided for an embodiment of this application.

[0025] Figure 3 The third form diagram of the stability of a dynamic calculation method for the electro-water conversion coefficient based on multi-dimensional data fusion is provided in the embodiments of this application. Detailed Implementation

[0026] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dynamic calculation method for the electro-water conversion coefficient based on multi-dimensional data fusion proposed in this application. 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.

[0027] 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 application pertains.

[0028] The following description, in conjunction with the accompanying drawings, details a specific scheme for a dynamic calculation method of the electro-water conversion coefficient based on multi-dimensional data fusion provided in this application.

[0029] Example:

[0030] The main application scenario of this application is as follows: When calculating the electricity-to-water conversion factor for an electric water supply system, since the calculation requires the electricity consumption and water delivery over a period of time, a fixed time period, such as one hour, is usually used. However, for scenarios or periods with large changes in water consumption, this fixed time period is relatively long and cannot capture the detailed changes in the electricity-to-water conversion factor. When a shorter time period is used, for scenarios or periods with small changes in water consumption, there will be large changes in the calculated electricity-to-water conversion factor at common change nodes such as water pump switching, which may lead to misjudgment that there is equipment abnormality or energy consumption change in the electricity-to-water conversion factor. Therefore, a more accurate electricity-to-water conversion factor calculation time period is needed.

[0031] This application provides a method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion. The method includes the following steps:

[0032] Step S1: Obtain the electricity-to-water conversion factor and monitoring data for each dimension in historical periods; take the average value of the monitoring data for one dimension in a period as the monitoring value for that dimension in that period; obtain the degree of influence of each dimension based on the correlation between the monitoring values ​​of each dimension in historical periods and the changes in the electricity-to-water conversion factor in each period.

[0033] First, relevant data needs to be collected. Electricity meters are installed in the power system to monitor electricity consumption, and flow sensors are installed in the system pipelines to monitor water delivery. The historical electricity consumption and water delivery of the current electricity-water supply system are obtained. The historical data collection frequency is fixed once per hour, thus obtaining the electricity consumption and water delivery corresponding to each hour in history. Then, the electricity-to-water conversion factor for each time period is calculated based on the electricity consumption and water delivery for each time period (existing technology).

[0034] Meanwhile, multi-dimensional monitoring sensors are installed in the electric water supply system to collect monitoring data from multiple dimensions, including water source head, pipeline pressure, water pipe vibration amplitude, water pump bearing temperature, and water pump efficiency. The monitoring frequency of each multi-dimensional monitoring sensor is once per minute. The monitoring data for each dimension is stored separately in a single sequence.

[0035] Relying solely on electricity consumption and water delivery data cannot accurately characterize the specific form of change when the electricity-to-water conversion factor changes. For example, changes in the electricity-to-water conversion factor caused by pump switching or periodic changes in water usage do not require deliberate monitoring. However, abnormal changes in the electricity-to-water conversion factor caused by impeller wear, cavitation, leakage, abnormal head, etc., need to be monitored in a timely manner. Therefore, this study combines multi-dimensional monitoring data to analyze the corresponding form of change when the electricity-to-water conversion factor changes.

[0036] This process obtains multi-dimensional monitoring data for each historical period corresponding to the calculation of the electro-water conversion coefficient. Specifically, it obtains the monitoring data for multiple dimensions within each period corresponding to the calculation of the electro-water conversion coefficient. The change in the electro-water conversion coefficient relative to the previous calculation is used as the change in the current electro-water conversion coefficient. The mean of the monitoring data for one dimension within a period is used to represent the monitoring value of that dimension within that period. The change in the mean of the multi-dimensional monitoring data for the period corresponding to the electro-water conversion coefficient relative to the previous period is obtained. If there are more correlations between the data and the coefficient for that dimension in multiple calculations, it indicates that the electro-water conversion coefficient is more likely to change when the monitoring data for that dimension changes, and the influence of that dimension on the electro-water conversion coefficient is stronger. Because the frequency of change for each dimension's data is different, the impact of dimension data change on the electro-water conversion coefficient change cannot be directly assessed based on the number of times there is a common change. Instead, the number of times the monitoring data for each dimension changes is obtained. If the proportion of the number of times the electro-water conversion coefficient changes among the number of times the data for that dimension changes is larger, the correlation is stronger.

[0037] Because data in real-world scenarios fluctuates, and this fluctuation is not considered a true change in the data itself, directly judging data change based on different data values ​​can lead to many fluctuations being mistaken for changes. Therefore, it's necessary to first identify the time periods when actual data changes occur. Fluctuations tend to be close to the original data, but due to the fitting of averages to the changed values, the overall fluctuating values ​​are relatively smaller than the mean. Conversely, significant changes in data will be much larger than the mean obtained from fitting a large number of fluctuating values. Therefore, the mean can be used to separate fluctuating data from data with a certain degree of change.

[0038] Furthermore, taking the monitoring data of a single dimension as an example for analysis, the influence degree of each dimension is obtained based on the correlation between the monitoring values ​​of each dimension in different historical periods and the changes in the electricity-to-water conversion coefficient in different periods.

[0039] Specifically, the absolute value of the difference between the monitoring values ​​of every two adjacent time periods in history for a dimension is obtained as the monitoring difference for every two adjacent time periods in that dimension, and these differences are arranged in chronological order to form a monitoring difference sequence for that dimension. The difference between the electricity-to-water conversion factor and the subsequent time period in history for every two adjacent time periods is obtained, and these differences are arranged in chronological order to form a difference sequence using the electricity-to-water conversion factor. Monitoring differences in the monitoring difference sequence for that dimension that are greater than the mean of the monitoring difference sequence are recorded as changes in monitoring differences. Differences in the difference sequence using the electricity-to-water conversion factor that are greater than or less than 0 are recorded as changes in differences. The number of changes in monitoring differences and changes in differences that are in the same chronological position is obtained, and the ratio of this number to the number of changes in monitoring differences yields the degree of influence of that dimension.

[0040] The specific calculation model is as follows:

[0041] ,

[0042] in, The influence level of the w-th dimension represents the degree of influence of changes in the monitoring data of the w-th dimension on the change in the electricity-to-water conversion factor. The difference between the electricity-to-water conversion factor of each two adjacent time periods in history is obtained, and these differences are arranged chronologically to form a sequence of electricity-to-water conversion factor differences. If any difference in the sequence is greater than or less than 0, it indicates a change in the electricity-to-water conversion factor corresponding to that node, and this difference is recorded as the change difference. The number of change monitoring differences in the monitoring difference sequence of this dimension is obtained. And obtain the number of locations where these changes and monitoring differences correspond to changes in water content expressed as electricity. That is, the number of differences and variations in changes at the same time position, which is the proportion of the change in the electro-water conversion coefficient in the change value of the current dimension. The larger the value, the greater the proportion of changes in the electro-to-water conversion factor within the total number of changes in that dimension. Therefore, the stronger the correlation between changes in the data of that w-th dimension and changes in the electro-to-water conversion factor, and the stronger the influence of the monitored data in that dimension on changes in the electro-to-water conversion factor. Similarly, performing the same analysis on each dimension can yield the degree of influence for each dimension.

[0043] Step S2: Based on the monitoring data of each dimension in the current time period, obtain the overall stability level of each dimension in the current time period; based on the difference between every two adjacent monitoring data of a dimension in the current time period, obtain the smooth change level of that dimension in the current time period; use the peak characteristics of the monitoring data of a dimension in the current time period to obtain the regular fluctuation degree of that dimension in the current time period.

[0044] The monitoring data for each dimension in the most recent time period is obtained as the recent monitoring data. This means acquiring the monitoring data for each dimension in the current time period and assessing the stability of the data for each dimension based on its distribution characteristics. The stability is used to characterize the change in the electro-water conversion coefficient of the electric water supply system. When the stability of the multi-source monitoring data is high, there are fewer details of changes in the electro-water conversion coefficient that need to be monitored; when the stability of the multi-source monitoring data is low, there are more real fluctuations in the electro-water conversion coefficient that need to be monitored promptly.

[0045] Then, taking the monitoring data of one dimension as an example for analysis, firstly, based on the monitoring data of each dimension in the current period, the overall stability level of each dimension in the current period is obtained.

[0046] Specifically, the reciprocal of the difference between the maximum and minimum values ​​of the monitoring data for a certain dimension in the current period is taken and normalized to obtain the overall stability level of that dimension in the current period.

[0047] The specific calculation model is as follows:

[0048] ,

[0049] ,

[0050] in, This represents the overall stationary level of the w-th dimension during the current time period; and Let $\mathbf$ and $\mathbf$ represent the maximum and minimum values ​​of the monitoring data in the $w$-th dimension during the current time period, respectively. This represents the reciprocal of the difference between the maximum and minimum values ​​of the monitoring data in the w-th dimension during the current time period. This is the normalized value of the reciprocal; and The historical data of the w-th dimension respectively The minimum and maximum values ​​of the w-th dimension are used to normalize the overall stability level of each dimension to the same order of magnitude. The smaller the overall stability level, the smaller the maximum variation of the data in the w-th dimension, indicating that the data as a whole does not change significantly, and thus the more stable the data.

[0051] For monitoring data across various dimensions, there may be instances where the overall data shows significant changes but minimal fluctuations, exhibiting a relatively smooth overall change. This can also characterize the stability of the monitoring data for that dimension. For example, the head of a water source in a water supply system gradually decreases as water is pumped, resulting in a slow change with a strong overall trend and no sudden fluctuations. This still retains strong calculability using the electricity-to-water conversion factor, without needing to be broken down into smaller time periods. The smoothness of change in a dimension for the current time period is obtained by accumulating the differences between each adjacent monitoring data point within that dimension for the current time period. Therefore, the smoothness of change in that dimension for the current time period is obtained based on the differences between every two adjacent monitoring data points.

[0052] Specifically, the absolute difference between every two adjacent monitoring data points in a dimension during the current time period is used to form a sequence of absolute difference values. The reciprocal of the mean of the absolute difference between each absolute difference value in the sequence and the mean of the sequence is obtained and normalized to obtain the level of gradual change in that dimension during the current time period.

[0053] The specific calculation model is as follows:

[0054] ,

[0055] ,

[0056] in, This represents the level of gradual change in the w-th dimension during the current time period. Before normalization ; This represents the number of monitored data points in the w-th dimension during the current time period. Let be the absolute value of the difference between the i-th monitoring data and the adjacent preceding monitoring data. This represents the mean of the absolute values ​​of the differences in the sequence of absolute differences for the current time period within this dimension. This represents the cumulative difference between the change in the monitoring data of the w-th dimension and the mean of that change. The smaller the cumulative difference, the more gradual the change in the w-th dimension. It is expressed in an inverse proportion. . The normalized values ​​of the maximum and minimum levels of gradual change in this dimension are used to normalize the level of gradual change in each dimension to the same order of magnitude. and They represent the history of The maximum and minimum values.

[0057] For water supply systems using the electricity-to-water conversion factor, there are cases where monitoring data shows strong and highly volatile changes, but it can also be considered a form of high stability. For example, intermittent water supply in irrigation areas, multiple pumps rotating in a pumping station, and regular water use in factory equipment settings. The regular water use causes the monitoring data to show regular fluctuations. Although this type of fluctuation data is highly volatile, it is not meaningful to break it down into hourly segments for refined calculation using the electricity-to-water conversion factor. Calculating it in hourly segments would instead misjudge these fluctuations as abnormal fluctuations in the electricity-to-water conversion factor.

[0058] Therefore, the degree of regular fluctuation in that dimension can be obtained by utilizing the peak characteristics of the monitoring data in one dimension during the current period.

[0059] Specifically, the monitoring data of one dimension in the current period are connected into a curve and the peak points are obtained; the horizontal distance between two adjacent trough points of a peak point is taken as the horizontal span corresponding to that peak point; the absolute value of the difference between the horizontal spans corresponding to one peak point and the absolute value of the difference in amplitude are multiplied to obtain the difference characteristic value between the two peak points; the mean of the difference characteristic values ​​between a peak point and other peak points is obtained as the average difference characteristic value corresponding to that peak point; the reciprocal of the mean of the average difference characteristic values ​​corresponding to each peak point is calculated and normalized to obtain the degree of regular fluctuation of that dimension in the current period.

[0060] The specific calculation model is as follows:

[0061] ,

[0062] ,

[0063] in, This represents the reciprocal of the mean of the average difference characteristic values ​​corresponding to each peak point in the w-th dimension during the current time period. This indicates the degree of regular fluctuation in the w-th dimension during the current time period; The number of peak points corresponding to the w-th dimension in the current time period is determined by connecting the temporally adjacent monitoring data of the w-th dimension in the current time period to form a curve, and then obtaining all peak points in the data through the AMPD peak search algorithm. and These represent the horizontal spans corresponding to the i-th and j-th wave crests, respectively. This represents the absolute value of the difference in the horizontal span corresponding to these two wave crests. and These represent the amplitudes of the i-th and j-th wave crests, respectively. The difference eigenvalues ​​between the two peak points, Let be the mean of the difference eigenvalues ​​between the i-th wave crest and all other wave crests, which is also the average difference eigenvalue corresponding to the i-th wave crest. This amplitude and span characterize the wave crest waveform. The smaller the mean of this difference eigenvalue, the smaller the difference between the wave crest corresponding to the i-th wave crest and other wave crests. The mean of the average difference eigenvalues ​​corresponding to all peak points represents the higher the waveform similarity of all fluctuations, the stronger the regularity of fluctuations in the w-th dimension. This is expressed in an inverse proportional form. . Based on history maximum value and minimum value The calculated normalized value normalizes the level of variation in each dimension to the same order of magnitude.

[0064] Therefore, we can obtain the overall stability level, the level of gradual change, and the degree of regular fluctuation in various dimensions during the current period.

[0065] Step S3: In the current time period, the overall stability level, the level of gradual change, and the degree of regular fluctuation of one dimension are integrated to obtain the stability assessment value of that dimension in the current time period; the optimized time period is obtained by using the fixed time period length, the stability assessment values ​​of each dimension in the current time period, and the influence degree of each dimension, and the electricity-to-water conversion coefficient of the optimized time period is calculated.

[0066] The above steps obtain the overall stability level, smooth change level, and regular fluctuation degree of each dimension in the current period. For a single dimension, it is necessary to integrate the overall stability level, smooth change level, and regular fluctuation degree of that dimension. Specifically, the average of the overall stability level, smooth change level, and regular fluctuation degree of each dimension in the current period is used to obtain the stability assessment value of that dimension in the current period.

[0067] like Figure 1 , Figure 2 and Figure 3 The figure shows the performance of the stationarity in various forms, among which Figure 1 This reflects the stable performance of data from a water supply system with relatively small overall fluctuations. Figure 2 To ensure a stable data representation for water supply systems with relatively high levels of gradual change, Figure 3 This represents a stable data representation for power and water supply systems with relatively strong regular fluctuations.

[0068] Furthermore, by comprehensively considering the stability estimates across multiple dimensions, stronger stability across more dimensions corresponds to lower precision required for calculating the electricity-to-water conversion factor of the current electricity-supply water system, and a longer calculation period. This prevents noise from normal data fluctuations corresponding to system operating conditions from affecting the accuracy of the coefficient, making the electricity-to-water conversion factor calculation results more reliable. Conversely, weaker stability across more dimensions corresponds to higher precision required for calculating the electricity-to-water conversion factor of the current electricity-supply water system, and a shorter calculation period. This allows for timely detection of short-term anomalies in the electricity-to-water conversion factor and observation of energy efficiency issues reflected by the electricity-to-water conversion factor.

[0069] Next, the optimized time period is obtained by using the fixed time period length, the stability evaluation value of each dimension of the current time period, and the influence degree of each dimension, and the electricity-to-water conversion factor of the optimized time period is calculated.

[0070] Specifically, the weighted average of the stability assessment values ​​of each dimension in the current period is obtained by using the degree of influence of each dimension as the weight of the stability assessment value of each dimension in the current period. The weighted average value is then multiplied by a fixed period length to obtain the adjusted period length. The current period is then evenly divided into optimized periods using the adjusted period length. The electricity-to-water conversion factor for each optimized period is calculated based on the electricity consumption and water delivery volume of each optimized period.

[0071] The specific calculation model for the adjusted time period length is as follows:

[0072] ,

[0073] This refers to the adjusted time period length corresponding to the current time period; The original fixed calculation period length is based on the electricity-to-water conversion factor. This is the weighted cumulative sum and mean of recent historical multi-dimensional monitoring data after weighting by the degree of impact, where the stability assessment value of each dimension is included. The larger the size, the greater the impact. The larger the value, the longer the final calculation period; the stability assessment value The smaller the size, the greater the impact. The larger the value, the shorter the final computation time period.

[0074] Then, using the adjusted time period length calculated based on the monitoring data of various dimensions of the current time period, the current time period is evenly divided to obtain each optimized time period, and the electricity-to-water conversion factor is calculated for each time period:

[0075] ,

[0076] Where K is the electricity-to-water conversion factor for an optimized time period, W is the water delivery volume within that optimized time period, and E is the electricity consumption within that optimized time period. The resulting electricity-to-water conversion factor can improve the calculation accuracy of the electricity-to-water conversion factor as much as possible during periods of fluctuation, preventing the omission of details regarding energy efficiency changes in the electricity-supply water system and reducing noise coefficients caused by time period length fluctuations during normal operating conditions. Additionally, it should be noted that when dividing the current time period using the adjusted time period length, the division may not be completely uniform because the length of the current time period may not be divisible by the adjusted time period length. In this case, the last time period obtained from the division is placed within the previous time period, i.e., the last two time periods are joined together.

[0077] In summary, this application constructs the relationship between multidimensional monitoring data and the electricity-to-water conversion coefficient by combining multidimensional monitoring data of the electric water supply system. Furthermore, it evaluates data with high-frequency calculation significance for the electricity-to-water conversion coefficient based on the stability characteristics of multidimensional data under different water supply modes, thereby obtaining a relatively accurate time period for calculating the electricity-to-water conversion coefficient, obtaining more accurate calculation results for the electricity-to-water conversion coefficient, reducing noise interference, and preventing missed detection of coefficient fluctuations.

[0078] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0079] 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.

[0080] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion, characterized in that, The method includes: Obtain historical data on the electricity-to-water conversion factor and monitoring data for each dimension; take the mean of the monitoring data for one dimension in a given period as the monitoring value for that dimension in that period; and determine the degree of influence of each dimension based on the correlation between the historical monitoring values ​​for each dimension and the changes in the electricity-to-water conversion factor in each period. Based on the monitoring data of each dimension in the current period, obtain the overall stability level of each dimension in the current period; based on the difference between every two adjacent monitoring data of a dimension in the current period, obtain the level of smooth change of that dimension in the current period; utilize the peak characteristics of the monitoring data of a dimension in the current period to obtain the degree of regular fluctuation of that dimension in the current period. The overall stability level, gradual change level, and regular fluctuation degree of one dimension in the current period are integrated to obtain the stability assessment value of that dimension in the current period; the optimized period is obtained by using a fixed period length, the stability assessment values ​​of each dimension in the current period, and the influence degree of each dimension, and the electricity-to-water conversion factor of the optimized period is calculated.

2. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The method of obtaining the degree of influence of each dimension based on the correlation between the monitoring values ​​of each dimension in different historical periods and the changes in the electricity-to-water conversion factor in each period includes: The absolute value of the difference between the monitoring values ​​of every two adjacent time periods in history for a dimension is obtained as the monitoring difference for every two adjacent time periods in that dimension, and these differences are arranged in chronological order to form a monitoring difference sequence for that dimension. The difference between the electricity-to-water conversion factor of the subsequent time period and the preceding time period in history for every two adjacent time periods is obtained, and these differences are arranged in chronological order to form a difference sequence of electricity-to-water conversion factor. The monitoring differences in the monitoring difference sequence for that dimension that are greater than the mean of the monitoring difference sequence are recorded as changes in monitoring differences. The differences in the difference sequence of electricity-to-water conversion factor that are greater than or less than 0 are recorded as changes in differences. The number of changes in monitoring differences and changes in differences that are in the same position in the time sequence is obtained, and the number of changes in monitoring differences is divided by the number of changes in monitoring differences to obtain the degree of influence of that dimension.

3. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The process of obtaining the overall stability level of each dimension based on the monitoring data of the current time period includes: The overall stability level of that dimension in the current period is obtained by taking the reciprocal of the difference between the maximum and minimum values ​​of the monitoring data for a certain dimension in the current period and normalizing it.

4. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The step of obtaining the level of gradual change in a dimension during the current time period based on the difference between every two adjacent monitoring data points of that dimension includes: The absolute difference between every two adjacent monitoring data points in one dimension during the current time period is used to form a sequence of absolute difference values. The reciprocal of the mean of the absolute difference between each absolute difference value in the sequence and the mean of the sequence is obtained and normalized to obtain the level of gradual change in that dimension during the current time period.

5. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The method of obtaining the degree of regular fluctuation in a certain dimension of monitoring data during the current time period by utilizing the peak characteristics of that dimension includes: Connect the monitoring data of one dimension in the current period into a curve and obtain the peak points; take the horizontal axis distance between two adjacent trough points of a peak point as the horizontal axis span corresponding to that peak point; multiply the absolute value of the difference between the horizontal axis spans corresponding to one peak point and the absolute value of the difference in amplitude of the two peak points to obtain the difference characteristic value of the two peak points; obtain the mean of the difference characteristic values ​​of one peak point and other peak points as the average difference characteristic value corresponding to that peak point; calculate the reciprocal of the mean of the average difference characteristic values ​​corresponding to each peak point and normalize it to obtain the degree of regular fluctuation of that dimension in the current period.

6. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The process of fusing the overall stability level, the level of gradual change, and the degree of regular fluctuation in one dimension of the current time period to obtain the stability assessment value of that dimension includes: The stability assessment value of each dimension in the current period is obtained by averaging the overall stability level, the level of gradual change, and the degree of regular fluctuation in each dimension.

7. The method for dynamically calculating the electro-water conversion coefficient based on multi-dimensional data fusion according to claim 1, characterized in that, The process of obtaining an optimized time period using a fixed time period length, the stability assessment values ​​of each dimension of the current time period, and the degree of influence of each dimension, and calculating the electricity-to-water conversion factor for the optimized time period, includes: The weighted average of the stability assessment values ​​of each dimension in the current period is obtained by using the degree of influence of each dimension as the weight of the stability assessment value of each dimension in the current period. The weighted average is then multiplied by a fixed period length to obtain the adjusted period length. The current period is then evenly divided into each optimized period using the adjusted period length. The electricity-to-water conversion factor for each optimized period is calculated based on the electricity consumption and water delivery of each optimized period.

Images