Intelligent water dispenser control method based on user behavior prediction and related device

Abstract

The invention relates to the technical field of energy-saving control, and discloses an intelligent water dispenser control method based on user behavior prediction and a related device. The method comprises the following steps: acquiring historical data statistics; obtaining a topological connection relation of the water dispenser; obtaining the time period and weather information, and setting a water drinking amount prediction value for each water dispenser based on historical data; water dispenser predicted values of the water dispenser in the previous time period and the current time period are obtained, a water drinking amount actual value of the water dispenser in the previous time period is obtained, and a water drinking amount difference value between the water dispenser predicted value in the previous time period and the water drinking amount actual value is calculated; and based on the water drinking amount difference value corresponding to the water dispenser in the previous time period and the water drinking amount difference value corresponding to the water dispenser in topological connection with the water dispenser, calculating an adjustment value of the water heating amount of the water dispenser in the current time period, and replacing the predicted water drinking amount value of the water dispenser with the adjustment value. The heating and refrigerating capacity of the water dispenser can be dynamically adjusted, the energy consumption is optimized, and the water drinking experience of a user is improved.

Description

Technical Field

[0001] The present application relates to the technical field of energy-saving control, and in particular to a smart water dispenser control method and related devices based on user behavior prediction. Background Art

[0002] Water dispensers are common water resource provision equipment in modern society and are widely used in public places such as office buildings and shopping malls. Traditional water dispensers usually have the function of supplying hot water, cold water and warm water. Their working principle mainly relies on the internal heating and cooling system to ensure that drinking water at the appropriate temperature is provided when needed. However, with the complexity of usage scenarios, the operation mode of traditional water dispensers has gradually exposed problems such as high energy consumption, deteriorating water quality and insufficient user experience.

[0003] In large public places such as shopping malls, the layout and operation of water dispenser systems are more complex. Usually, water dispensers are distributed on multiple floors or in different functional areas to meet the dispersed drinking water needs of users. Although this layout is convenient for users, it also brings significant energy consumption problems. For example, water dispensers in shopping malls usually maintain a long-term heating or cooling state, and continue to consume electricity even when no one is using them, resulting in high overall energy consumption of the system. At the same time, water dispensers that are in a heating state for a long time will cause the "thousand-boil water" phenomenon, which not only wastes resources, but may also have adverse effects on water quality. In addition, traditional water dispensers cannot flexibly adjust the water supply mode according to real-time user needs. During peak hours, they may not be able to meet the needs of a large number of users for hot or cold water, and during non-peak hours, resources may be wasted. This mode directly affects the user's drinking experience, especially in scenarios with large flow fluctuations, the problem is particularly prominent. Summary of the invention

[0004] In order to dynamically adjust the heating and cooling capacity of the water dispenser, optimize energy consumption, avoid the "thousands of boiling water" phenomenon, and improve the user's drinking experience, the present application provides an intelligent water dispenser control method and related devices based on user behavior prediction.

[0005] In the first aspect, the present application provides a smart water dispenser control method based on user behavior prediction, which adopts the following technical solution: A method for controlling an intelligent water dispenser based on user behavior prediction comprises the following steps: S1. Obtain historical data statistics, where the historical data includes the temperature and weather at different times of the past days, and the corresponding drinking water volume of each water dispenser, including the amount of hot water, warm water and cold water; S2. Obtaining a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the terrain of the building and the distribution of water dispensers; S3. Obtain the time period and weather information to set a water consumption forecast value for each water dispenser based on historical data; wherein the water consumption forecast value is used to guide the water boiling amount of the water dispenser in the corresponding time period; S4. Obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the drinking water volume of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the drinking water volume; S5. Based on the difference in water volume corresponding to the water dispenser in the previous period and the difference in water volume corresponding to the water dispensers to which the water dispenser is topologically connected, calculate the adjustment value of the water boiling volume of the water dispenser in the current period, and use the adjustment value to replace the predicted water volume of the water dispenser to guide the actual water boiling volume in the current period.

[0006] Optionally, the statistical method of historical data comprises the following steps: S11. Divide a day into several time periods based on a first preset step length, divide the weather into different types, and pair the time periods and weather types in pairs to obtain an environmental data set, wherein the weather type includes temperature and weather, the temperature is divided into several temperature intervals according to a second preset step length, and the weather includes sunny days, cloudy days, rainy days, and snowy days; S12. Obtain the drinking water volume data corresponding to the drinking machine corresponding to the environmental data group in the database, and calculate the mean and standard deviation; S13. Determine whether the standard deviation is greater than a third preset threshold value. If so, reduce the first preset step size or the second preset step size, and return to S11 until the standard deviation is less than the third preset threshold value. S14. Combining the environmental data group and the corresponding drinking water volume data of the water dispenser into historical data.

[0007] Optionally, the water dispenser topological connection relationship is established based on the following steps: S21. Generate a route map inside the building based on the topographic map of the building; S22. Determine the location of the water dispenser on the internal route map based on the distribution of the water dispenser; S23. Determine the topological connection relationship between the water dispensers based on the position of the water dispensers on the internal route map; S24. Based on historical data statistics, determine the correlation between topologically connected water dispensers.

[0008] Optionally, S3 includes the following steps: S31. Obtain the current time period and weather information, and match the corresponding environmental data group; S32. Based on the environmental data group, the drinking water volume data of the corresponding water dispenser is matched from the historical data and used as the drinking water volume prediction value.

[0009] Optionally, the S5 comprises the following steps: S51. Calculate the ratio of the difference in the amount of water consumed by the water dispenser in the previous period to the predicted value of the amount of water consumed by the water dispenser, determine the deviation interval into which the ratio falls, and obtain the degree of deviation corresponding to the amount of water consumed by the water dispenser in the previous period, which is the first degree of deviation, wherein the deviation interval is a preset ratio interval, and different deviation intervals are used to represent different degrees of deviation; S52. Calculate the ratio of the difference in drinking water volume corresponding to the drinking machines topologically connected to the drinking machine in the previous period to the predicted drinking water volume of the drinking machine topologically connected to the drinking machine, determine the deviation interval into which the ratio falls, and obtain the deviation degree corresponding to the drinking water volume corresponding to the drinking machines topologically connected to the drinking machine in the previous period, which is the second deviation degree; S53. A first influence weight is obtained based on the first deviation degree, and a second influence weight of the water dispenser is obtained based on the difference and correlation between the first deviation degree and the second deviation degree, wherein the first influence weight is used to characterize the influence of the actual drinking water value at the previous moment on the drinking water value at the current moment, and the second influence weight is used to characterize the influence of the actual drinking water value of the topologically connected water dispensers at the previous moment on the drinking water value at the current moment; S54. Based on the product of the first influence weight and the predicted water volume of the water dispenser at the previous moment, plus the product of the second influence weight and the predicted water volume of the topologically connected water dispensers at the previous moment, the adjustment value of the water boiling volume of the water dispenser in the current period is used to guide the actual water boiling volume in the current period.

[0010] Optionally, the S53 includes the following steps: Determine whether the absolute value of the product of the difference between the first deviation degree and the second deviation degree and the correlation degree is greater than a third preset threshold value, and if so, when the first deviation degree is greater than the second deviation degree, set the second influence weight to a positive value, and when the first deviation degree is less than the second deviation degree, set the second influence weight to a negative value; If not, the second influence weight is zero or is set to a value close to zero and strongly correlated with the degree of association.

[0011] In the second aspect, the present application provides an intelligent water dispenser control device based on user behavior prediction, which adopts the following technical solution: An intelligent water dispenser control device based on user behavior prediction, comprising: The historical data statistics module is used to obtain historical data statistics, wherein the historical data includes the temperature and weather at different time periods of each day in the past, and the corresponding drinking water volume of each water dispenser, and the drinking water volume includes the amount of hot water, warm water and cold water; A topological relationship acquisition module, used to acquire a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the topography of the building and the distribution of the water dispensers; A prediction value calculation module is used to obtain the time period and weather information to set a water consumption prediction value for each water dispenser based on historical data; wherein the water consumption prediction value is used to guide the water boiling amount of the water dispenser in the corresponding time period; A water consumption difference calculation module is used to obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the water consumption of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the water consumption; The actual water boiling volume control module is used to calculate the adjustment value of the water boiling volume of the water dispenser in the current period based on the water drinking volume difference corresponding to the water dispenser in the previous period and the water drinking volume difference corresponding to the water dispenser to which the water dispenser is topologically connected, and use the adjustment value to replace the water drinking volume prediction value of the water dispenser to guide the actual water boiling volume in the current period.

[0012] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprising: one or more processors; Memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to: Execute the above-mentioned intelligent water dispenser control method based on user behavior prediction.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute the above method.

[0014] The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement: Such as the intelligent water dispenser control method based on user behavior prediction mentioned above.

[0015] In summary, the present application includes at least one of the following beneficial technical effects: 1. Through the classification and induction of historical data, as well as the dynamic matching based on the current period and environmental data, accurate prediction of each water dispenser under different environmental conditions is achieved. The standardized processing, deviation analysis and dynamic adjustment strategy of historical data effectively improve the accuracy and reliability of the prediction.

[0016] 2. This application uses the deviation analysis mechanism and topological correlation calculation to respond to the dynamic demand changes of water dispensers in real time. In particular, for the continuously distributed water dispensers on the main roads, the prediction values ​​can be optimized collaboratively; and for the water dispensers in hidden locations, differentiated deviation processing is used to ensure that their independent needs are met.

[0017] 3. By comprehensively analyzing the historical usage of a single water dispenser and its topological relationship with other water dispensers, this application achieves dynamic adjustment of the amount of water to be boiled, effectively avoiding oversupply or shortage of resources and improving resource utilization efficiency. At the same time, a closed-loop feedback mechanism is introduced to make corrections based on the difference between the prediction and the actual value of the previous period, continuously optimize the prediction model, and enhance the system's self-learning and adaptability in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a smart water dispenser control method based on user behavior prediction in a certain embodiment of the present application.

[0019] Figure 2 This is a module connection diagram of an intelligent water dispenser control device based on user behavior prediction in a certain embodiment of the present application. DETAILED DESCRIPTION

[0020] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0021] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0022] The present application embodiment discloses a smart water dispenser control method based on user behavior prediction, referring to Figure 1 , including the following steps S1-S5.

[0023] S1. Obtain historical data statistics, wherein the historical data includes the temperature and weather at different time periods of each day in the past, and the corresponding drinking water volume of each water dispenser, and the drinking water volume includes the amount of hot water, warm water and cold water.

[0024] The statistical method of historical data includes the following steps S11-S14.

[0025] S11. Divide a day into several time periods based on a first preset step length, divide the weather into different types, and pair the time periods and weather types in pairs to obtain an environmental data set, wherein the weather type includes temperature and weather, the temperature is divided into several temperature intervals according to a second preset step length, and the weather includes sunny days, cloudy days, rainy days, and snowy days; S12. Obtain the drinking water volume data corresponding to the drinking machine corresponding to the environmental data group in the database, and calculate the mean and standard deviation; S13. Determine whether the standard deviation is greater than a third preset threshold value. If so, reduce the first preset step size or the second preset step size, and return to S11 until the standard deviation is less than the third preset threshold value. S14. Combining the environmental data group and the corresponding drinking water volume data of the water dispenser into historical data.

[0026] First, historical data includes environmental factors such as temperature and weather, as well as water consumption data of each water dispenser. These data are not simply linearly related, but are affected by multiple factors. For example, changes in temperature will directly affect the demand for hot and cold water, while weather (such as sunny or rainy days) may indirectly change the demand for water consumption by affecting customer flow.

[0027] The pairwise pairing method is the key to analyzing the intrinsic relationship between data. By combining time periods and weather types, we can obtain an environmental data set that is highly consistent with the actual usage scenario. For example, in a sunny summer afternoon, the demand for cold water from a water dispenser may be much higher than in a cloudy winter evening.

[0028] Since there may be outliers with large fluctuations in historical data, these outliers will affect the accuracy of statistical results. To this end, S13 and S14 use the mean and standard deviation to determine whether the data distribution is stable, and use the threshold of the standard deviation as the adjustment division step. This process divides the data more finely by dynamically adjusting the first and second preset steps, thereby gradually reducing the standard deviation and improving the stability of the data.

[0029] Finally, the environmental data group and the drinking water data are combined into historical data. On the one hand, this can reduce redundant data and improve data utilization efficiency; on the other hand, it is also convenient for subsequent models to be directly called, simplifying the implementation of the algorithm.

[0030] S2. Obtaining a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the topography of the building and the distribution of the water dispensers.

[0031] The topological connection relationship of the water dispenser is established based on the following steps S21-S24.

[0032] S21. Generate a building interior route map based on the building's topographic map.

[0033] S22. Determine the location of the water dispenser on the internal road map based on the distribution of the water dispensers.

[0034] S23. Determine the topological connection relationship between the water dispensers based on the positions of the water dispensers on the internal route map.

[0035] S24. Based on historical data statistics, determine the correlation between topologically connected water dispensers.

[0036] First, the topological connection relationship between water dispensers is constructed based on the internal topographic map of the building. The internal topographic map of the building can be understood as a graphic model that describes the layout and path distribution of the building, which can clearly present the distribution characteristics of water dispensers in the physical space. For example, in a large commercial building, water dispensers on different floors may be distributed at the end of the corridor, next to the meeting room, or on both sides of the main road. By generating an internal route map based on the building topographic map, the actual spatial relationship between water dispensers can be accurately described.

[0037] In S22 and S23, based on the actual distribution location of the water dispenser, the water dispenser is mapped to the internal road map. This step realizes the connection between the physical space information and the water dispenser data, so that each water dispenser has a specific coordinate position in the topological relationship network. It should be noted that there are obvious differences in the distribution of water dispensers in hidden locations and on main roads. For example, hidden water dispensers may exist in isolation, while water dispensers on main roads may be distributed continuously. Therefore, this mapping relationship is not only a simple mark of the location, but also needs to be combined with the actual scene to analyze the connection possibility and closeness between different water dispenser points.

[0038] In S24, by calculating the geographical distance, path accessibility, and user behavior data between each water dispenser, the correlation between the water dispensers can be quantified. For example, the continuous distribution of water dispensers on the main road may have a higher correlation because the changes in passenger flow tend to be synchronized; while the hidden water dispensers may have a lower correlation because the user behavior is relatively independent. In this process, the topological connection relationship actually constructs a network that reflects the spatial distribution of water dispensers and the characteristics of user behavior.

[0039] It should be noted that in S2, the correlation between topologically connected water dispensers is determined based on historical data statistics. This step reveals the dynamic relationship between water dispensers by analyzing historical passenger flow and water consumption data. For example, if the water consumption trends of two water dispensers have a high correlation, it means that they may be affected by the same passenger flow group and have a high correlation. On the contrary, if the water consumption changes of the two water dispensers are independent or anti-correlated, it means that their correlation is low. The correlation here is not static, it may be dynamically adjusted with changes in time and environmental conditions, so as to more realistically reflect the relationship between water dispensers.

[0040] S3. Obtain the time period and weather information to set a predicted drinking water volume for each water dispenser based on historical data; wherein the predicted drinking water volume is used to guide the amount of water boiled by the water dispenser in the corresponding time period.

[0041] Specifically, S3 includes the following steps S31-S32.

[0042] S31. Obtain the current time period and weather information, and match and obtain the corresponding environmental data group.

[0043] S32. Based on the environmental data group, the drinking water volume data of the corresponding water dispenser is matched from the historical data and used as the drinking water volume prediction value.

[0044] This step uses environmental information (time period and weather) to match historical data and extracts drinking water data that matches the current environment as a prediction value. The drinking water prediction value is directly used to guide the amount of water boiled by the water dispenser. Since time period and weather are one of the main factors affecting drinking water, this method can effectively capture the phased changes in user demand. For example, during the noon period on a hot and sunny day, users usually have a higher demand for cold water; while on cold winter nights, the demand for hot water may dominate.

[0045] First, the system obtains the current time period and weather information in real time, and matches this information with the environmental data group in the historical data. The environmental data group contains a combination of time, temperature and weather information, such as "sunny + 30°C + 3 to 4 pm". By matching the current environmental information with the historical environmental data group, the historical data closest to the current environment can be quickly located.

[0046] Secondly, the matched environmental data group is used to extract the corresponding water intake data from the historical data and use it as the water intake prediction value. The prediction value here is not simply to select a specific historical value, but is based on the statistical results of historical data (such as the mean). Smoothing the fluctuations in historical data by the mean can effectively reduce the randomness of the prediction value and improve its stability. For example, assuming that the historical water intake data for a certain period of time are 10L, 12L and 9L respectively, the mean is 10.33L.

[0047] S4. Obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the drinking water volume of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the drinking water volume.

[0048] This step provides a basis for adjusting the amount of water boiled by the water dispenser in the current period by calculating the difference between the predicted value and the actual value of the previous period. This process is based on a feedback mechanism, and its purpose is to dynamically monitor the difference between the predicted and actual values, thereby improving the accuracy and adaptability of subsequent predicted values. Specifically, S4 first obtains the predicted value and actual value of the previous period, where the predicted value is generated by S3 and the actual value is collected in real time through sensors or metering equipment. The difference between the two reflects the degree of deviation of the system's prediction. By calculating the difference, the system can quickly determine whether the current prediction is accurate and whether the prediction for the next period needs to be adjusted.

[0049] It should be noted that the positive or negative value of the difference here has meaning. If the difference is positive, it means that the predicted value of the previous period is lower than the actual value, and the user demand exceeds the system estimate; conversely, if the difference is negative, it means that the predicted value is higher than the actual value, and there may be a risk of wasting resources. For example, assuming that the predicted value of the previous period is 12L and the actual value is 15L, the difference is 3L. This result indicates that the user demand of the previous period exceeded expectations, and the system needs to appropriately increase the predicted value in the subsequent period to meet the actual needs of users.

[0050] The introduction of this feedback mechanism enables the system to dynamically capture the changing trends of user demand. For example, in some special scenarios (such as sudden activities or seasonal changes), water intake may fluctuate significantly. By calculating the difference in real time and analyzing the changing pattern of the deviation, the system can quickly adapt to such demand fluctuations. More importantly, the calculation of the difference provides data for dynamic weight allocation such as in S5.

[0051] Specifically, the difference in water intake ΔV=AP, where A is the actual value of the previous period, P is the predicted value of the previous period, and ΔV represents the difference in water intake.

[0052] ΔV>0: The predicted value is lower than the actual value, indicating that user demand exceeds the prediction.

[0053] ΔV<0: The predicted value is higher than the actual value, indicating that the system prediction is too high.

[0054] S5. Based on the difference in water volume corresponding to the water dispenser in the previous period and the difference in water volume corresponding to the water dispensers to which the water dispenser is topologically connected, calculate the adjustment value of the water boiling volume of the water dispenser in the current period, and use the adjustment value to replace the predicted water volume of the water dispenser to guide the actual water boiling volume in the current period.

[0055] S5 includes the following steps S51-S54.

[0056] S51. Calculate the ratio of the difference in the amount of water consumed by the water dispenser in the previous period to the predicted value of the amount of water consumed by the water dispenser, determine the deviation interval into which the ratio falls, and obtain the degree of deviation corresponding to the amount of water consumed by the water dispenser in the previous period, which is the first degree of deviation, wherein the deviation interval is a preset proportional interval, and different deviation intervals are used to represent different degrees of deviation.

[0057] In S51, the difference in water consumption is defined as the difference between the actual value and the predicted value in the previous period. This difference directly reflects the degree of deviation of the system's estimate of actual demand. However, using the difference alone cannot intuitively reflect the relative impact of the deviation. In order to more clearly measure the severity of the deviation, S51 introduces the ratio of the difference to the predicted value as a key indicator. This ratio represents the proportion of the difference to the predicted value, which can unify the dimensions and make the degree of deviation of different water dispensers comparable. The formula is expressed as: .

[0058] in, is the actual amount of water consumed in the previous period, is the forecast value for the previous period, is the deviation ratio. The deviation ratio eliminates the positive and negative signs of the difference through absolute value calculation and only focuses on the magnitude of the deviation, which is suitable for analysis in different scenarios.

[0059] After calculating the deviation ratio, S51 classifies the ratio into different deviation degrees according to a preset ratio interval. For example, the ratio interval is divided into the following categories: : Low degree of deviation, indicating that the system prediction value is basically consistent with the actual value.

[0060] : A medium degree of deviation, indicating that there is a certain error in the system prediction.

[0061] : A high degree of deviation indicates that the system prediction value deviates significantly from the actual demand.

[0062] In this way, the system can quantitatively evaluate the accuracy of the forecast for the previous period and provide a basis for subsequent steps. If the deviation of a certain water dispenser is high, it may mean that the demand pattern at that point has changed abnormally, such as a temporary event or a sudden increase in customer flow. In this case, the system needs to focus on adjusting the forecast value of this water dispenser in the next period to better match the actual demand.

[0063] S52. Calculate the ratio of the difference in water consumption corresponding to the water dispensers topologically connected to the water dispenser in the previous period to the predicted value of water consumption of the water dispensers topologically connected to the water dispenser, determine the deviation interval in which the ratio falls, and obtain the deviation degree corresponding to the water consumption corresponding to the water dispensers topologically connected to the water dispenser in the previous period, which is the second deviation degree.

[0064] First, the basis of S52 is the topological connection relationship. Topologically connected drinking fountains usually have passenger flow interactions or resource sharing relationships. For example, drinking fountains distributed continuously on the main road may show similar trends in water consumption due to passenger flow fluctuations. In this context, S52 introduces a deviation ratio to measure the degree of prediction deviation of topologically connected drinking fountains. Similar to S51, the deviation ratio here is defined as: .

[0065] in, is the actual value of the topologically connected drinking fountain j in the previous period, For its predicted value, Its deviation ratio.

[0066] If the deviation degree of the topologically connected water dispensers is high, it may have a greater dynamic adjustment impact on the current water dispenser; if the deviation degree is low, it means that its demand changes are relatively stable and the impact on the current water dispenser is relatively limited.

[0067] Further analysis shows that the design of S52 reflects the consideration of the interactivity of topologically connected water dispensers. In actual scenarios, water dispensers on the main roads may have a strong correlation with each other, and their changes in water consumption often show a synergistic effect. Water dispensers in hidden locations may be more affected by single-point user behavior and have lower interactivity with other water dispensers. By quantifying the degree of deviation of topologically connected water dispensers, S52 can capture these complex relationships and provide a more refined basis for adjustment in subsequent steps. For example, if a main road water dispenser has a low degree of deviation, while its adjacent water dispenser has a high degree of deviation, this may indicate that the adjacent water dispenser is attracting more customers, thereby affecting the demand forecast of the current water dispenser.

[0068] S53. Obtain a first influence weight based on the first deviation degree, and obtain a second influence weight of the water dispenser based on the difference and correlation between the first deviation degree and the second deviation degree, wherein the first influence weight is used to characterize the influence of the actual drinking water value at the previous moment on the drinking water value at this moment, and the second influence weight is used to characterize the influence of the actual drinking water value of the topologically connected water dispensers at the previous moment on the drinking water value at this moment.

[0069] In this step, the concept of the first influence weight is proposed to quantify the impact of the actual drinking value of the water dispenser at the previous moment on the current predicted value. This weight is directly calculated based on the deviation degree of the current water dispenser, and the formula is: .in, is the deviation degree of the water dispenser, is the first impact weight, and f is the function used to map the degree of deviation to a weight. The design of this process reflects the dominance of single-point historical data on the current prediction. When the degree of deviation is high, it means that there is a significant difference between the actual water intake in the previous period and the predicted value. At this time, the predicted value of the current period needs to be adjusted to a greater extent, so the first impact weight will be higher; conversely, when the degree of deviation is low, it means that the prediction of the previous period is more accurate, and the prediction model can be used in the current period without major adjustments.

[0070] In this step, a second influence weight is also introduced to characterize the influence of the deviation degree of topologically connected water dispensers on the current water dispenser. The formula is: .in, and are the deviation degrees of the current water dispenser and the topologically connected water dispensers, is the correlation between the two, and g is a function that maps the deviation difference and correlation to weights.

[0071] The behavioral characteristics of topologically connected water dispensers (such as changes in passenger flow or demand shifts) have an important impact on the predicted value of the current water dispenser. When the degree of deviation is large, it means that there is a significant inconsistency between the demand pattern of the current water dispenser and the topologically connected water dispenser. This may be caused by an emergency (such as an event). At this time, it is necessary to focus on the impact of their correlation on the predicted value. If the degree of deviation between the two is close, it means that their demand change trends are consistent, and the correlation will play a more significant positive role.

[0072] When calculating the second impact weight, the combination of deviation degree difference and correlation is particularly critical. For example, if the deviation degree of water dispenser i and the adjacent water dispenser j on a main road is very different, but the correlation between the two is high, it may indicate that the passenger flow on the main road is shifting to j, which will have a significant impact on the predicted value of i. At this time, the system can adjust the predicted value according to the weight to make resource allocation more reasonable.

[0073] Specifically, S53 includes the following steps: Determine whether the absolute value of the product of the difference between the first deviation degree and the second deviation degree and the correlation degree is greater than a third preset threshold value. If so, when the first deviation degree is greater than the second deviation degree, the second influence weight is set to a positive value, and when the first deviation degree is less than the second deviation degree, the second influence weight is set to a negative value; if not, the second influence weight is zero or is set to a value close to zero and strongly correlated with the correlation degree.

[0074] The deviation difference reflects the different degrees of demand change between the current water dispenser and the topologically associated water dispenser. By multiplying the deviation difference and the correlation, the system can quantify the potential impact of this difference on the relationship between the two. The formula is: .

[0075] The larger the deviation difference and the higher the correlation, the more significant the impact on forecast adjustment. For example, if the deviation degree of two strongly correlated water dispensers is significantly different, it may have a strong negative effect on resource allocation and forecast adjustment.

[0076] The system sets a third preset threshold to determine whether the product of the deviation difference and the correlation is large enough: like , indicating that the difference in deviations is significant and has an important impact on the adjustment.

[0077] like , indicating that the deviation difference is not significant or the correlation is low, and the effect on adjustment is limited.

[0078] like , that is, the deviation degree of the current water dispenser is greater than that of the topologically associated water dispenser, and the second influence weight is set to a positive value. This indicates that the demand change of the current water dispenser may have a positive effect on the associated water dispenser, such as diverting users or sharing resources.

[0079] like , that is, the deviation degree of the current water dispenser is less than that of the topologically associated water dispenser, and the second influence weight is set to a negative value. This indicates that the demand change of the associated water dispenser may have a negative impact on the current water dispenser, such as attracting users and causing the current water dispenser to be used less.

[0080] In the case of insignificant deviations, the system sets the second impact weight to zero or a value close to zero. The value close to zero is related to the degree of association to ensure that a certain adjustment sensitivity is retained in the case of low significance. This value close to zero is set based on experience and can be 0 to 0.3 multiplied by the degree of association.

[0081] S54. Based on the product of the first influence weight and the predicted water volume of the water dispenser at the previous moment, plus the product of the second influence weight and the predicted water volume of the topologically connected water dispensers at the previous moment, the adjustment value of the water boiling volume of the water dispenser in the current period is used to guide the actual water boiling volume in the current period.

[0082] The adjustment value is calculated as: .

[0083] in, : The first influence weight represents the direct impact of the historical deviation of the current water dispenser on its predicted value.

[0084] : The second influence weight of each topologically associated water dispenser jj reflects its dynamic influence on the current water dispenser.

[0085] : The correlation between the current water dispenser and the topologically connected water dispenser jj.

[0086] and : They are the predicted values ​​of the current water dispenser and the topologically associated water dispenser respectively.

[0087] : The current water boiling volume adjustment value of the water dispenser.

[0088] For example, suppose the current forecast value of water dispenser A in the previous period is , the first impact weight is The predicted values ​​and associated data of the two adjacent water dispensers B and C are as follows: , the degree of deviation , correlation .

[0089] , the degree of deviation , correlation .

[0090] , the third preset threshold is 0.2.

[0091] For B, Since 0.06<0.2, .

[0092] The adjustment values ​​are: , after substituting the data: The final adjustment value is 7L, which means that the current water dispenser does not need to be affected by the topologically associated water dispensers and only relies on its own historical data for adjustment.

[0093] In one embodiment, a smart water dispenser control device based on user behavior prediction is provided, and the smart water dispenser control device based on user behavior prediction corresponds one-to-one to the smart water dispenser control method based on user behavior prediction in the above embodiment. Figure 2As shown, the functional modules of the intelligent water dispenser control device based on user behavior prediction are described in detail as follows: The historical data statistics module is used to obtain historical data statistics, wherein the historical data includes the temperature and weather at different time periods of each day in the past, and the corresponding drinking water volume of each water dispenser, and the drinking water volume includes the amount of hot water, warm water and cold water; A topological relationship acquisition module, used to acquire a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the topography of the building and the distribution of the water dispensers; A prediction value calculation module is used to obtain the time period and weather information to set a water consumption prediction value for each water dispenser based on historical data; wherein the water consumption prediction value is used to guide the water boiling amount of the water dispenser in the corresponding time period; A water consumption difference calculation module is used to obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the water consumption of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the water consumption; The actual water boiling volume control module is used to calculate the adjustment value of the water boiling volume of the water dispenser in the current period based on the water drinking volume difference corresponding to the water dispenser in the previous period and the water drinking volume difference corresponding to the water dispenser to which the water dispenser is topologically connected, and use the adjustment value to replace the water drinking volume prediction value of the water dispenser to guide the actual water boiling volume in the current period.

[0094] For the specific definition of the intelligent water dispenser control device based on user behavior prediction, please refer to the definition of the intelligent water dispenser control method based on user behavior prediction in the above text, which will not be repeated here. Each module in the above-mentioned intelligent water dispenser control device based on user behavior prediction can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0095] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the intelligent water dispenser control method based on user behavior prediction in the above embodiment is implemented. Alternatively, when the processor executes the computer program, the functions of each module / unit of the intelligent water dispenser control device based on user behavior prediction in the above embodiment are implemented. To avoid repetition, it will not be described here.

[0096] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the intelligent water dispenser control method based on user behavior prediction in the above embodiment is implemented. Alternatively, when the computer program is executed by a processor, the functions of each module / unit in the intelligent water dispenser control device based on user behavior prediction in the above device embodiment are implemented. To avoid repetition, it will not be repeated here.

[0097] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments of the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0098] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0099] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A smart water dispenser control method based on user behavior prediction, characterized in that: The following steps are involved: S1. Obtain historical data statistics, where the historical data includes the temperature and weather at different times of the past days, and the corresponding drinking water volume of each water dispenser, including the amount of hot water, warm water and cold water; S2. Obtaining a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the terrain of the building and the distribution of water dispensers; S3. Obtain the time period and weather information to set a water consumption forecast value for each water dispenser based on historical data; wherein the water consumption forecast value is used to guide the water boiling amount of the water dispenser in the corresponding time period; S4. Obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the drinking water volume of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the drinking water volume; S5. Based on the difference in water volume corresponding to the water dispenser in the previous period and the difference in water volume corresponding to the water dispensers to which the water dispenser is topologically connected, calculate the adjustment value of the water boiling volume of the water dispenser in the current period, and use the adjustment value to replace the predicted water volume of the water dispenser to guide the actual water boiling volume in the current period.

2. The intelligent water dispenser control method based on user behavior prediction according to claim 1 is characterized in that: The statistical method of historical data comprises the following steps: S11. Divide a day into several time periods based on a first preset step length, divide the weather into different types, and pair the time periods and weather types in pairs to obtain an environmental data set, wherein the weather type includes temperature and weather, the temperature is divided into several temperature intervals according to a second preset step length, and the weather includes sunny days, cloudy days, rainy days, and snowy days; S12. Obtain the drinking water volume data corresponding to the drinking machine corresponding to the environmental data group in the database, and calculate the mean and standard deviation; S13. Determine whether the standard deviation is greater than a third preset threshold value. If so, reduce the first preset step size or the second preset step size, and return to S11 until the standard deviation is less than the third preset threshold value. S14. Combining the environmental data group and the corresponding drinking water volume data of the water dispenser into historical data.

3. The intelligent water dispenser control method based on user behavior prediction according to claim 2 is characterized in that: The water dispenser topology connection relationship is established based on the following steps: S21. Generate a route map inside the building based on the topographic map of the building; S22. Determine the location of the water dispenser on the internal route map based on the distribution of the water dispenser; S23. Determine the topological connection relationship between the water dispensers based on the position of the water dispensers on the internal route map; S24. Based on historical data statistics, determine the correlation between topologically connected water dispensers.

4. The intelligent water dispenser control method based on user behavior prediction according to claim 3 is characterized in that: The S3 comprises the following steps: S31. Obtain the current time period and weather information, and match the corresponding environmental data group; S32. Based on the environmental data group, the drinking water volume data of the corresponding water dispenser is matched from the historical data and used as the drinking water volume prediction value.

5. The intelligent water dispenser control method based on user behavior prediction according to claim 4 is characterized in that: The S5 comprises the following steps: S51. Calculate the ratio of the difference in the amount of water consumed by the water dispenser in the previous period to the predicted value of the amount of water consumed by the water dispenser, determine the deviation interval into which the ratio falls, and obtain the degree of deviation corresponding to the amount of water consumed by the water dispenser in the previous period, which is the first degree of deviation, wherein the deviation interval is a preset ratio interval, and different deviation intervals are used to represent different degrees of deviation; S52. Calculate the ratio of the difference in drinking water volume corresponding to the drinking machines topologically connected to the drinking machine in the previous period to the predicted drinking water volume of the drinking machine topologically connected to the drinking machine, determine the deviation interval into which the ratio falls, and obtain the deviation degree corresponding to the drinking water volume corresponding to the drinking machines topologically connected to the drinking machine in the previous period, which is the second deviation degree; S53. A first influence weight is obtained based on the first deviation degree, and a second influence weight of the water dispenser is obtained based on the difference and correlation between the first deviation degree and the second deviation degree, wherein the first influence weight is used to characterize the influence of the actual drinking water value at the previous moment on the drinking water value at the current moment, and the second influence weight is used to characterize the influence of the actual drinking water value of the topologically connected water dispensers at the previous moment on the drinking water value at the current moment; S54. Based on the product of the first influence weight and the predicted water volume of the water dispenser at the previous moment, plus the product of the second influence weight and the predicted water volume of the topologically connected water dispensers at the previous moment, the adjustment value of the water boiling volume of the water dispenser in the current period is used to guide the actual water boiling volume in the current period.

6. The intelligent water dispenser control method based on user behavior prediction according to claim 5 is characterized in that: The S53 comprises the following steps: Determine whether the absolute value of the product of the difference between the first deviation degree and the second deviation degree and the correlation degree is greater than a third preset threshold value, and if so, when the first deviation degree is greater than the second deviation degree, set the second influence weight to a positive value, and when the first deviation degree is less than the second deviation degree, set the second influence weight to a negative value; If not, the second influence weight is zero or is set to a value close to zero and strongly correlated with the degree of association.

7. An intelligent water dispenser control device based on user behavior prediction, characterized in that: include: The historical data statistics module is used to obtain historical data statistics, wherein the historical data includes the temperature and weather at different time periods of each day in the past, and the corresponding drinking water volume of each water dispenser, and the drinking water volume includes the amount of hot water, warm water and cold water; A topological relationship acquisition module, used to acquire a topological connection relationship of water dispensers, wherein the topological connection relationship of water dispensers is established based on the topography of the building and the distribution of the water dispensers; A prediction value calculation module is used to obtain the time period and weather information to set a water consumption prediction value for each water dispenser based on historical data; wherein the water consumption prediction value is used to guide the water boiling amount of the water dispenser in the corresponding time period; A water consumption difference calculation module is used to obtain the predicted value of the water dispenser in the previous period and the current period, obtain the actual value of the water consumption of the water dispenser in the previous period, and calculate the difference between the predicted value of the water dispenser in the previous period and the actual value of the water consumption; The actual water boiling volume control module is used to calculate the adjustment value of the water boiling volume of the water dispenser in the current period based on the water drinking volume difference corresponding to the water dispenser in the previous period and the water drinking volume difference corresponding to the water dispenser to which the water dispenser is topologically connected, and use the adjustment value to replace the water drinking volume prediction value of the water dispenser to guide the actual water boiling volume in the current period.

8. A computer device, characterized in that: It includes: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to: execute the smart water dispenser control method based on user behavior prediction according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement: the intelligent water dispenser control method based on user behavior prediction as described in any one of claims 1 to 6.

Images