Water plant water quantity management method and device based on machine learning

Abstract

The invention belongs to the technical field of water plant management, and discloses a water plant water quantity management method and device based on machine learning, high-precision water plant water demand adapting to different environments is obtained by combining meteorological data and date data, then reasonable distribution of water plant resources can be guaranteed by using a machine learning model, and the water plant water quantity management efficiency is improved. According to the obtained target water supply amount and historical water supply data of each water plant, intelligently generating target scheduling data of the water pumping station; secondly, various water pump state parameters can be obtained according to the target scheduling data and the historical operation parameters of the water pump station, the water pump station is controlled through all the water pump state parameters to achieve intelligent scheduling of the water volume of the water plant, the overall energy consumption can be reduced, the scheduling frequency is effectively reduced on the basis of machine learning, and the scheduling efficiency is improved. And meanwhile, the raw water scheduling efficiency and the control precision are improved.

Description

Technical Field

[0001] The present application belongs to the field of water plant management technology, and in particular, relates to a water plant water volume management method and device based on machine learning. Background Art

[0002] Modern urban water resources management is a key factor in the sustainable development of social economy. It involves many aspects such as the rational development, efficient utilization, effective protection and scientific management of urban water resources. With the acceleration of urbanization, population growth and economic development needs, urban water resources are facing unprecedented pressure. Therefore, how to dispatch raw water for water resources is particularly important.

[0003] Usually, the raw water dispatching method is that the water pump station (or raw water pump room) dispatches the raw water to the water plant through the water pump, and the water plant supplies the clean water to users through the urban pipeline network after the treated clean water meets the liquid level requirements. However, this dispatching method is mainly formulated based on the experience of the operators. It not only relies heavily on the dispatching experience, but also has problems such as high dispatching frequency or inaccurate liquid level control, which in turn affects the overall energy consumption and efficiency of raw water dispatching. Summary of the invention

[0004] This application aims to solve the above-mentioned existing scheduling methods, which are mainly formulated based on the experience of operators. Not only is it highly dependent on scheduling experience, but it also has problems such as high scheduling frequency or inaccurate liquid level control, which in turn affects the overall energy consumption and efficiency of raw water scheduling. Technical defects such as this, a water plant water volume management method and device based on machine learning are proposed, and its technical solution is as follows: In a first aspect, an embodiment of the present application provides a water volume management method for a water plant based on machine learning, comprising: Determine the target meteorological data and date data within the specified period, and obtain the target water demand required for urban water use based on the target meteorological data and date data; At least one water plant is numbered according to a preset numbering method, and the target water demand and the numbering result are input into a machine learning model to obtain the target water supply of each water plant, and the target scheduling data of the water pump station is generated according to the target water supply of all water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; According to the target dispatching data and the historical operating parameters of the water pumping station, at least two water pump status parameters corresponding to each water plant are obtained, and at least two total operating powers of the water pumping station are generated based on all the water pump status parameters; All water pump status parameters and target scheduling data corresponding to the minimum total operating power are sent to the water pump station, so that the water pump station outputs water to all water plants according to all corresponding water pump status parameters and target scheduling data.

[0005] In an optional solution of the first aspect, the target meteorological data includes a target rainfall and a target temperature; Determine the target weather data and date data within the specified period, including: Evenly divide the specified time period into at least two interval periods, and obtain at least two historical rainfall amounts corresponding to all interval periods; When there is no corresponding historical rainfall in any interval period, the supplementary rainfall is obtained according to all the historical rainfalls corresponding to all interval periods adjacent to the interval period; According to the arrangement of all interval periods, the supplementary rainfall is updated in all historical rainfalls, and all the updated historical rainfalls are used as target rainfalls; Obtain the historical temperature interval corresponding to each interval, select the lowest temperature and the highest temperature from each historical temperature interval as a temperature set, and use all temperature sets as target temperatures; When it is detected that the specified time period contains dates corresponding to holidays, the preset date feature vector is used as date data.

[0006] In another optional solution of the first aspect, obtaining the target water demand for urban water use according to the target meteorological data and date data includes: The target rainfall and the target temperature are standardized respectively, and the processed target rainfall and the processed target temperature are combined to obtain a meteorological feature vector; The meteorological feature vector and date data are input into the regularized regression model to obtain the target water demand required for urban water use; wherein the regularized regression model is trained by at least two groups of feature vector samples and the historical water demand corresponding to each group of feature vector samples, and each group of feature vector samples includes a meteorological feature vector sample and a date feature vector sample.

[0007] In yet another optional solution of the first aspect, the historical water supply data includes historical scheduling data and historical water supply volume; According to the target water supply of all water plants and the historical water supply data corresponding to each water plant, the target dispatching data of the water pumping station is generated, including: Obtain the historical dispatching data and historical water supply of each water plant in a specified period, and train the Bayesian optimization model based on all the historical dispatching data and all the historical water supply; The target water supply of each water plant is input into the trained Bayesian optimization model to obtain the corresponding pump scheduling data, and the target scheduling data of the pump station is generated based on all the pump scheduling data.

[0008] In another optional solution of the first aspect, the water pump scheduling data includes at least one water pump number and a scheduling period corresponding to each water pump number; Generate target dispatch data of the pump station based on all pump dispatch data, including: Integrate all water pump dispatch data to obtain at least one dispatch period corresponding to each water pump number; When it is detected that all the dispatching periods corresponding to any water pump number have overlapping periods, the period ratio between the overlapping period and all the corresponding dispatching periods is calculated; Arrange all time period proportions according to a preset proportion order, and update all corresponding scheduling time periods according to the arrangement result, so that there is no overlapping time period between all the scheduling time periods after processing; Generate target dispatching data of the pumping station according to all dispatching periods corresponding to all pump numbers.

[0009] In another optional solution of the first aspect, the water pump scheduling data further includes the water output corresponding to each water pump number; According to the target dispatching data and the historical operating parameters of the water pumping station, at least two water pump status parameters corresponding to each water plant are obtained, including: All pump numbers corresponding to each water plant are screened out from the target dispatching data, and based on all pump numbers corresponding to each water plant, the historical operating parameters of the water pump station are divided and processed to obtain the corresponding historical operating parameters; The historical dispatching data is divided and processed based on all the pump numbers corresponding to each water plant to obtain the corresponding historical water output, and the decision tree learning model is trained based on the historical operating parameters and historical water output corresponding to each water plant; The water output corresponding to all the water pump numbers of each water plant is input into the trained decision tree learning model to obtain corresponding at least two water pump state parameters.

[0010] In another optional solution of the first aspect, generating at least two total operating powers of the water pump station based on all water pump state parameters includes: According to all the water pump state parameters corresponding to each water plant, at least two state parameter sets of the water pump station are combined; wherein each state parameter set includes any one water pump state parameter corresponding to each water plant; According to the water pump state parameters corresponding to each water plant in each state parameter set, all water pump numbers and the scheduling period corresponding to each water pump number, the operating time corresponding to each water pump number is obtained; The operating time corresponding to each water pump number and the preset pump station power are multiplied, and all product results are summed to obtain the total operating power of the corresponding state parameter set.

[0011] In a second aspect, an embodiment of the present application provides a water plant water volume management device based on machine learning, comprising: The water demand forecasting module is used to determine the target meteorological data and date data within a specified period of time, and obtain the target water demand required for urban water use based on the target meteorological data and date data; A water pump scheduling prediction module is used to number at least one water plant according to a preset numbering method, input the target water demand and the numbering result into the machine learning model, obtain the target water supply of each water plant, and generate the target scheduling data of the water pump station according to the target water supply of all water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; A water pump state prediction module is used to obtain at least two water pump state parameters corresponding to each water plant according to the target scheduling data and the historical operating parameters of the water pump station, and generate at least two total operating powers of the water pump station based on all the water pump state parameters; The water volume management module is used to send all water pump status parameters and target scheduling data corresponding to the minimum total operating power to the water pump station, so that the water pump station outputs water to all water plants according to all corresponding water pump status parameters and target scheduling data.

[0012] In a third aspect, an embodiment of the present application further provides a water plant water volume management device based on machine learning, including a processor and a memory; The processor is connected to the memory; A memory for storing executable program codes; The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the water volume management method of the water plant based on machine learning provided by the first aspect of the embodiment of the present application or any implementation method of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the water volume management method of a water plant based on machine learning provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect can be implemented.

[0014] Beneficial effects of this application: In the water volume management process of the water plant for raw water scheduling, the target meteorological data and date data within the specified time period are determined, and the target water demand required for urban water use is obtained based on the target meteorological data and date data; at least one water plant is numbered according to a preset numbering method, and the target water demand and the numbering result are input into the machine learning model to obtain the target water supply of each water plant, and the target scheduling data of the pump station is generated based on the target water supply of all water plants and the historical water supply data corresponding to each water plant; at least two pump status parameters corresponding to each water plant are obtained based on the target scheduling data and the historical operating parameters of the pump station, and at least two total operating powers of the pump station are generated based on all the pump status parameters; all the pump status parameters corresponding to the minimum total operating power and the target scheduling data are sent to the pump station, so that the pump station outputs water to all water plants according to all the corresponding pump status parameters and the target scheduling data. By combining meteorological data and date data, we can obtain high-precision water demand for water plants that adapt to different environments. Then, we can use machine learning models to ensure the rational distribution of water plant resources, and intelligently generate target scheduling data for pump stations based on the target water supply and historical water supply data of each water plant. Secondly, we can also obtain a variety of pump status parameters based on the target scheduling data and the historical operating parameters of the pump station, so as to control the pump station through all pump status parameters to achieve intelligent scheduling of water volume in the water plant, reduce overall energy consumption, and effectively reduce the scheduling frequency based on machine learning, while improving the efficiency and control accuracy of raw water scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 An overall flow chart of a water plant water volume management method based on machine learning provided in an embodiment of the present application; Figure 2 A schematic diagram of a water resource management architecture provided in an embodiment of the present application; Figure 3A schematic diagram of the structure of a water plant water volume management device based on machine learning provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of another water plant water quantity management device based on machine learning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0018] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, although the embodiment may not be clearly recorded in the following text.

[0019] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.

[0020] See also Figure 1 , Figure 1 An overall flow chart of a water plant water volume management method based on machine learning provided in an embodiment of the present application is shown.

[0021] like Figure 1 As shown, the water volume management method of a water plant based on machine learning may at least include the following steps: Step 102: Determine target meteorological data and date data within a specified period, and obtain target water demand for urban water use based on the target meteorological data and date data.

[0022] In the embodiment of the present application, the water volume management method of a water plant based on machine learning can be applied to, but is not limited to, a control terminal. The control terminal can establish connections with a water pump station and a water plant respectively, so as to intelligently formulate a scheduling plan based on the historical data of the water pump station and the historical data of the water plant, and then automatically control the water pump station to output water to the water plant to meet the clean water level requirements. Among them, the water pump station may include one or more water pumps, and the type of each water pump may be, but is not limited to, a power frequency pump or a variable frequency pump, and each water pump may establish a connection with one or more water plants to extract raw water from the water source to the water plant according to the control instructions issued by the control terminal; here, the control instructions issued by the control terminal can be generated based on the historical data of the water pump station, and the historical data of the water pump station can at least include the historical water demand dispatched by the water pump station to all water plants (that is, the historical water volume dispatched by the water pump station to each water plant), the historical scheduling period corresponding to when each water pump dispatches raw water to the corresponding water plant (that is, the historical scheduling period when the water pump dispatches raw water to the water plant The generated control instructions include but are not limited to the target scheduling period (that is, the target period of the current water pump dispatching to the water plant), the target water output (that is, the target water volume of the current water pump dispatching to the water plant), and the water pump status parameters (that is, the target switching state of the current water pump dispatching to the water plant) corresponding to each water pump dispatching raw water to the corresponding water plant.

[0023] There can be one or more water plants, and each water plant can process the raw water dispatched by at least one water pump until the treated clean water meets the liquid level requirements and then supplies the clean water to users through the urban pipeline network. The historical data of the water plant can be the historical water supply of all water plants supplied to users through the urban pipeline network (it can also be understood as the urban water demand, that is, the sum of the historical water supply of all water plants).

[0024] Also see here Figure 2 A schematic diagram of a water resource management architecture provided by an embodiment of the present application is shown, such as Figure 2 As shown, the water pump station may include three water pumps (respectively represented as pump numbered 1, pump numbered 2 and pump numbered 3), wherein pump numbered 1 can dispatch raw water from the water source to the water plant numbered 1, so that the water plant numbered 1 can supply the treated clean water to users through the urban pipeline network; pump numbered 2 and pump numbered 3 can dispatch raw water from the water source to the water plant numbered 2, so that the water plant numbered 2 can supply the treated clean water to users through the urban pipeline network.

[0025] In addition, the control terminal can also, but is not limited to, obtain the historical meteorological data and historical date data corresponding to when the water pump station dispatches raw water to all water plants. The historical meteorological data can include, but is not limited to, historical rainfall and historical temperature ranges. The historical date data can include, but is not limited to, holiday period data. The control terminal can also, but is not limited to, combine the historical data of the water pump station mentioned above and the historical data of the water plant to jointly and intelligently formulate a scheduling plan, and is not limited to this.

[0026] It can be understood that the control terminal can obtain high-precision water demand of the water plant adapted to different environments by combining meteorological data and date data, and then use the machine learning model to ensure the rational distribution of water plant resources, and intelligently generate the target scheduling data of the pump station based on the target water supply and historical water supply data of each water plant; secondly, a variety of pump status parameters can be obtained according to the target scheduling data and the historical operating parameters of the pump station, so as to control the pump station through all pump status parameters to realize intelligent scheduling of water volume in the water plant, reduce overall energy consumption, and effectively reduce the scheduling frequency based on machine learning, while improving the efficiency and control accuracy of raw water scheduling.

[0027] Specifically, in the water plant water volume management process of raw water scheduling, the control terminal can, but is not limited to, obtain historical meteorological data and historical date data within a specified period, and analyze and process the historical meteorological data and historical date data respectively to obtain corresponding target meteorological data and date data. Here, taking the specified period as the past thirty days as an example, the target meteorological data can, but is not limited to, include multiple rainfall, maximum temperature, and minimum temperature within the past thirty days, and the date data can, but is not limited to, the number of holidays within the past thirty days.

[0028] As an option in the embodiment of the present application, the target meteorological data includes target rainfall and target temperature; Determine the target weather data and date data within the specified period, including: Evenly divide the specified time period into at least two interval periods, and obtain at least two historical rainfall amounts corresponding to all interval periods; When there is no corresponding historical rainfall in any interval period, the supplementary rainfall is obtained according to all the historical rainfalls corresponding to all interval periods adjacent to the interval period; According to the arrangement of all interval periods, the supplementary rainfall is updated in all historical rainfalls, and all the updated historical rainfalls are used as target rainfalls; Obtain the historical temperature interval corresponding to each interval, select the lowest temperature and the highest temperature from each historical temperature interval as a temperature set, and use all temperature sets as target temperatures; When it is detected that the specified time period contains dates corresponding to holidays, the preset date feature vector is used as date data.

[0029] Specifically, taking the target meteorological data including target rainfall and target temperature as an example, when determining the target meteorological data and date data within the specified time period, the control terminal can divide the specified time period into multiple interval time periods. For example, when the specified time period is the past thirty days, it can be divided into thirty interval time periods according to the number of days, and the historical rainfall in the past thirty days can be queried through the weather software.

[0030] Then, after obtaining the historical rainfall in the specified time period, the control terminal can also determine whether each interval period has corresponding historical rainfall to ensure the integrity of the target meteorological data. For example, taking the interval period divided into thirty periods according to the number of days, when it is detected that the number of days corresponding to the historical rainfall is less than the total number of interval periods, that is, there is any at least one interval period without corresponding historical rainfall, and then the historical rainfall corresponding to all corresponding interval periods can be determined according to the number of days corresponding to the historical rainfall, and one or more interval periods without historical rainfall can be screened out from all interval periods, and the corresponding supplementary rainfall is determined by combining all the historical rainfall corresponding to all interval periods adjacent to each interval period.

[0031] Here, when the number of intervals adjacent to any interval in which there is no historical rainfall is two, the average of all historical rainfalls corresponding to the two adjacent intervals can be used as the supplementary rainfall for the interval; when the number of intervals adjacent to any interval in which there is no historical rainfall is one, all historical rainfalls corresponding to the adjacent intervals can be directly used as the supplementary rainfall for the interval; when the number of intervals adjacent to any interval in which there is no historical rainfall is zero, the supplementary rainfall corresponding to the interval adjacent to the interval can be first determined by the above method, and then the supplementary rainfall for the interval can be determined based on all supplementary rainfalls corresponding to the adjacent intervals, and this is not limited to this.

[0032] Then, after obtaining the supplementary rainfall corresponding to all the interval time periods without historical rainfall, one or more supplementary rainfall can be updated in all the historical rainfall according to the arrangement order and position relationship of each interval time period, so that each interval time period corresponds to historical rainfall, and all the updated historical rainfall can be used as the target rainfall in the target meteorological data.

[0033] It is understandable that after the control terminal divides the specified period into multiple interval periods, for example, taking the specified period as the past thirty days, and dividing it into thirty interval periods according to the number of days, the weather software can also be used to obtain the historical temperature intervals of each day in the past thirty days, and the highest temperature and the lowest temperature are selected from each historical temperature interval as a temperature set, so as to use the thirty temperature sets as the target temperature in the target meteorological data. Of course, the target temperature in the embodiment of the present application can also be the highest temperature and the lowest temperature in the past thirty days, and is not limited to this.

[0034] It can also be understood that after the control terminal divides the specified period into multiple interval periods, for example, taking the specified period as the past thirty days, and dividing it into thirty interval periods according to the number of days, the calendar software can also be used to query the number of holidays in the past thirty days, and the preset date feature vector corresponding to the holiday number can be used as date data. Here, different holiday days correspond to different preset date feature vectors, and the dimension of the preset date feature vector can be consistent with the dimension of the feature vector corresponding to the target meteorological data mentioned above, but is not limited thereto.

[0035] Further, after determining the target meteorological data and date data within the specified time period, the control terminal can, but is not limited to, perform standardization processing on the target rainfall and target temperature in the target meteorological data in a matrix format, that is, to ensure the consistency of the data, and obtain the corresponding meteorological feature vector by merging the processed target rainfall and the processed target temperature. Here, the embodiment of the present application can also pre-process the target rainfall and target temperature in the target meteorological data to further ensure the integrity and validity of the data, but it is not repeated here.

[0036] Further, after obtaining the meteorological feature vector, the control terminal can input the meteorological feature vector and date data (i.e., the preset date feature vector) into the regularized regression model, and predict the high-precision water demand of the water plant adapted to different environments through the machine learning algorithm, that is, the target water demand required for urban water use. It can be understood that the regularized regression model can be understood as the minimum absolute shrinkage and selection operator regression model (i.e., the Lasso regression model) that introduces the regularization term, which can effectively ensure data processing efficiency and avoid overfitting, and effectively predict the water demand required for urban water use by combining the meteorological changes with periodic laws and the holiday effect factors. It can be obtained by training multiple groups of feature vector samples and the historical water demand corresponding to each group of feature vector samples (i.e., the historical water volume dispatched by the pump station to all water plants within a specified period of time). Each group of feature vector samples here may include meteorological feature vector samples and date feature vector samples obtained by manual analysis, and is not limited to this.

[0037] Step 104: number at least one water plant according to a preset numbering method, input the target water demand and the numbering result into the machine learning model to obtain the target water supply of each water plant, and generate the target scheduling data of the water pump station based on the target water supply of all water plants and the historical water supply data corresponding to each water plant.

[0038] Specifically, after determining the target water demand for urban water use, the control terminal can, but is not limited to, number all water plants corresponding to the water pumping stations according to a preset numbering method to effectively distinguish each water plant, and can input the numbering results of all water plants and the target water demand mentioned above into the machine learning model. The target water supply of each water plant is predicted through the machine learning algorithm, that is, the amount of water supplied to users by each water plant through the urban pipeline network, so as to optimize the resource distribution of each water plant while ensuring water supply safety.

[0039] Here, the machine learning model can be but is not limited to a multilayer perceptron model (MLP) or other neural network models, including an input layer, multiple hidden layers (each hidden layer may include multiple neurons) and an output layer, etc., structures well known in the art and preset functions (such as activation functions and loss functions, etc.), and can be trained by multiple historical water demands dispatched by the water pumping station to all water plants within a specified period of time, all water plant numbers corresponding to each historical water demand, and the historical water supply corresponding to each water plant in each water plant number (that is, the amount of water supplied to users by each water plant through the urban pipeline network within a specified period of time).

[0040] Further, after predicting the target water supply of each water plant, the control terminal may, but is not limited to, obtain the historical water supply data corresponding to each water plant, and predict the target dispatching data of the water pump station based on the historical water supply data and the target water supply of each water plant, such as predicting the dispatching time period and water output corresponding to each water pump when the water pump station controls each water pump to dispatch raw water to the water plant. Here, the historical water supply data corresponding to each water plant may at least include the historical dispatching data corresponding to when the water pump station controls one or more water pumps to dispatch raw water to the corresponding water plant within a specified time period, and the historical water supply supplied to users by the corresponding water plant through the urban pipe network within a specified time period. The historical dispatching data may at least include at least one water pump number, the historical dispatching time period and the historical water output corresponding to each water pump number.

[0041] As another option of the embodiment of the present application, the historical water supply data includes historical scheduling data and historical water supply volume; According to the target water supply of all water plants and the historical water supply data corresponding to each water plant, the target dispatching data of the water pumping station is generated, including: Obtain the historical dispatching data and historical water supply of each water plant in a specified period, and train the Bayesian optimization model based on all the historical dispatching data and all the historical water supply; The target water supply of each water plant is input into the trained Bayesian optimization model to obtain the corresponding pump scheduling data, and the target scheduling data of the pump station is generated based on all the pump scheduling data.

[0042] Specifically, taking the historical water supply data including historical dispatching data and historical water supply as an example, the control terminal can obtain the data when the water pump station controls each water pump to dispatch raw water to the water plant during a specified period of time (that is, historical dispatching data) from the platform terminal where the water pump station is located, and obtain the water volume supplied to users by the corresponding water plant through the urban pipeline network during a specified period of time (that is, historical water supply) from the platform terminal where each water plant is located, and use the historical dispatching data and historical water supply as training sets to train the Bayesian optimization model. Here, the Bayesian optimization model can be understood as a probability model constructed using the Bayesian optimization algorithm, which optimizes and adjusts the data when the water pump station controls the water pump to dispatch raw water to the corresponding water plant by combining the water volume supplied to users by each water plant through the urban pipeline network.

[0043] It can be understood that in the training process of the Bayesian optimization model, the Gaussian process model can be selected as the prior probability model, but is not limited to it, so as to use the mean function and covariance function to describe the probability distribution, and can effectively capture the uncertainty of the data; secondly, the Bayesian theorem can be used to perform a posterior distribution on the historical scheduling data, historical water supply and prior probability model mentioned above, and select a preset acquisition function for iterative optimization, such as but not limited to when the maximum number of iterations is reached or the convergence criteria are met (of course, it can also be judged based on whether the clean water level of the water plant is stable, but is not limited to this), to determine that the Bayesian optimization model training is completed.

[0044] Then, after the Bayesian optimization model training is completed, the control terminal can also input the target water supply of each water plant (including the corresponding water plant number) into the trained Bayesian optimization model to predict the corresponding water pump scheduling data (that is, the data of the water pump station controlling the water pump when scheduling raw water to the corresponding water plant, which may include but is not limited to at least one water pump number, the scheduling period corresponding to each water pump number, and the water output) through the machine learning algorithm, and generate the target scheduling data of the water pump station based on all the water pump scheduling data to ensure that each water pump of the water pump station can normally schedule raw water to the water plant.

[0045] As another option of the embodiment of the present application, the water pump scheduling data includes at least one water pump number and a scheduling period corresponding to each water pump number; Generate target dispatch data of the pump station based on all pump dispatch data, including: Integrate all water pump dispatch data to obtain at least one dispatch period corresponding to each water pump number; When it is detected that all the dispatching periods corresponding to any water pump number have overlapping periods, the period ratio between the overlapping period and all the corresponding dispatching periods is calculated; Arrange all time period proportions according to a preset proportion order, and update all corresponding scheduling time periods according to the arrangement result, so that there is no overlapping time period between all the scheduling time periods after processing; Generate target dispatching data of the pumping station according to all dispatching periods corresponding to all pump numbers.

[0046] Specifically, when generating the target dispatching data of the pump station according to the pump dispatching data corresponding to each water plant, the control terminal can classify all dispatching time periods according to the pump numbers in all the pump dispatching data to obtain all dispatching time periods of all water plants corresponding to each pump number, and can determine whether there are overlapping time periods between all dispatching time periods corresponding to each pump number, that is, determine whether there is abnormal operation of the pump. It can be understood that when it is detected that there are overlapping time periods in all dispatching time periods corresponding to any pump number, it indicates that it is necessary to stagger all dispatching time periods with repeated time periods, and then calculate the time period ratio between the overlapping time period and all corresponding dispatching time periods. Here, all dispatching time periods can be expressed as 9:00-13:00 and 11:00-14:00 as an example. Since there is an overlapping time period from 11:00 to 13:00, it can be calculated that the proportions of the two time periods are 0.50 and 0.67 respectively.

[0047] Then, after obtaining the proportions of all time periods, the control terminal can also arrange all time period proportions according to the preset proportion order, for example, all time period proportions can be arranged in order from large to small, and the scheduling period corresponding to the maximum time period proportion can be kept unchanged, so as to update the scheduling period corresponding to other time period proportions according to the time sequence between the scheduling period corresponding to other time period proportions and the scheduling period corresponding to the maximum time period proportion, and ensure that there is no overlapping time period between all the processed scheduling periods. For example, taking the above-mentioned scheduling periods as 9:00-13:00 and 11:00-14:00 as examples, after calculating the proportions of the two time periods of 0.50 and 0.67 respectively, since the scheduling period represented by 11:00-14:00 has the largest calculated proportion, the scheduling period represented by 9:00-13:00 is before the scheduling period represented by 11:00-14:00, and thus the scheduling period represented by 11:00-14:00 can be kept unchanged, and the scheduling period represented by 9:00-13:00 can be updated to 7:00-11:00.

[0048] Then, after determining that there is no overlapping period in all scheduling periods corresponding to each water pump number, the control terminal can integrate all scheduling periods corresponding to all water pump numbers and the water output corresponding to each scheduling period into the target scheduling data of the water pump station.

[0049] Step 106: Obtain at least two water pump status parameters corresponding to each water plant according to the target scheduling data and the historical operating parameters of the water pumping station, and generate at least two total operating powers of the water pumping station based on all the water pump status parameters.

[0050] Specifically, after determining the target dispatching data of the water pump station, the control terminal may, but is not limited to, filter out all the water pump numbers corresponding to each water plant in the target dispatching data, and divide the historical operation parameters corresponding to each water plant in the historical operation parameters of the water pump station according to all the water pump numbers corresponding to each water plant. Here, the historical operation parameters of the water pump station can be understood as the historical switch state of each water pump of the water pump station dispatched to the water plant within a specified period of time, and the historical operation parameters corresponding to each water plant divided out can be understood as the historical switch state of one or more water pumps corresponding to each water plant dispatched to the corresponding water plant within a specified period of time, and the historical switch state can be, but is not limited to, being in an open operation state from 0:00 (i.e., 24:00) to 12:00, and being in a closed shutdown state from 12:00 to 24:00 (i.e., 0:00).

[0051] It is understandable that after screening out all the pump numbers corresponding to each water plant, the control terminal can also divide the historical water output corresponding to each water plant in the above-mentioned historical dispatching data according to all the pump numbers corresponding to each water plant, that is, the amount of water dispatched to the corresponding water plant by one or more pumps corresponding to each water plant within a specified period of time, and can train the decision tree learning model in combination with the historical operating parameters and historical water output corresponding to each water plant. Here, the decision tree learning model can be, but is not limited to, an integrated learning model constructed for a random forest algorithm, so as to predict at least two operating parameters that meet the energy consumption requirements according to the water output corresponding to each water plant, and in the embodiment of the present application, the decision tree learning model can also be trained in combination with a preset energy consumption optimization function to achieve the purpose of saving energy consumption, but is not limited to this.

[0052] Furthermore, after the decision tree learning model is trained, the control terminal can input the water output corresponding to all the pump numbers of each water plant into the trained decision tree learning model to predict at least two corresponding pump state parameters through the machine learning algorithm. Each pump state parameter can be understood as the switching state of one or more pumps corresponding to each water plant when dispatching raw water to the corresponding water plant, thereby achieving a perfect balance between energy saving and efficient operation.

[0053] Furthermore, after determining all the water pump status parameters corresponding to each water plant, the control terminal can also randomly combine all the water pump status parameters corresponding to each water plant to obtain at least two status parameter sets of the water pump station, each of which may include any water pump status parameter corresponding to each water plant, and it is necessary to ensure that each water pump in each status parameter set is in a normal switching state at any time (that is, each water pump cannot be in the on-running state and the off-shutdown state at the same time at any time).

[0054] Furthermore, after obtaining multiple status parameter sets of the water pump station, the control terminal can also count the operating time of each pump number when it is in the on-state based on the pump status parameters corresponding to each water plant in each status parameter set, all pump numbers, and all scheduling time periods corresponding to each pump number. For example, but not limited to, when the pump status parameters corresponding to the pump number 1 are in the on-state from 0:00 to 12:00, and in the off-state from 12:00 to 24:00, and all corresponding scheduling time periods are expressed as 9:00-13:00, it can be determined that the operating time of the pump number 1 is 4 hours, and is not limited to this.

[0055] Furthermore, after determining the operating time corresponding to each pump number, the control terminal can also perform product processing on the operating time corresponding to each pump number and the preset pump station power, and sum all the product results to obtain the total operating power of the corresponding state parameter set. Here, you can also refer to the power calculation formula shown below: In the above formula, It can correspond to the total operating power of the water pump station obtained according to the water pump state parameters corresponding to each water plant in each state parameter set; K can correspond to the number of water pumps in the water pump station; It can correspond to the rated power corresponding to the kth water pump (that is, the preset pump station power); It can correspond to the running time of the kth water pump.

[0056] Step 108: Send all water pump status parameters and target scheduling data corresponding to the minimum total operating power to the water pump station, so that the water pump station outputs water to all water plants according to all corresponding water pump status parameters and target scheduling data.

[0057] Specifically, after obtaining the total operating power corresponding to each set of state parameters, the control terminal can, but is not limited to, use the minimum total operating power as the target total operating power, that is, to maximize the reduction of the energy consumption of the water pump station during operation, and avoid the problem of high frequency of water pump scheduling caused by human factors, and send all water pump state parameters and target scheduling data corresponding to the minimum total operating power to the water pump station, so that the water pump station controls each water pump to dispatch raw water to the corresponding water plant according to the scheduling period and switch status based on all the water pump state parameters and target scheduling data.

[0058] See next Figure 3 , Figure 3 A structural schematic diagram of a water plant water volume management device based on machine learning provided in an embodiment of the present application is shown.

[0059] like Figure 3 As shown, the water plant water management device based on machine learning may at least include a water demand prediction module 301, a water pump scheduling prediction module 302, a water pump state prediction module 303 and a water management module 304, wherein: The water demand prediction module 301 is used to determine the target meteorological data and date data within a specified period of time, and obtain the target water demand required for urban water use according to the target meteorological data and date data; The water pump scheduling prediction module 302 is used to number at least one water plant according to a preset numbering method, input the target water demand and the numbering result into the machine learning model, obtain the target water supply of each water plant, and generate the target scheduling data of the water pump station according to the target water supply of all water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; The water pump state prediction module 303 is used to obtain at least two water pump state parameters corresponding to each water plant according to the target scheduling data and the historical operating parameters of the water pump station, and generate at least two total operating powers of the water pump station based on all the water pump state parameters; The water volume management module 304 is used to send all water pump status parameters and target scheduling data corresponding to the minimum total operating power to the water pump station, so that the water pump station outputs water to all water plants according to all corresponding water pump status parameters and target scheduling data.

[0060] In some possible embodiments, the target meteorological data includes target rainfall and target temperature; Determine the target weather data and date data within the specified period, including: Evenly divide the specified time period into at least two interval periods, and obtain at least two historical rainfall amounts corresponding to all interval periods; When there is no corresponding historical rainfall in any interval period, the supplementary rainfall is obtained according to all the historical rainfalls corresponding to all interval periods adjacent to the interval period; According to the arrangement of all interval periods, the supplementary rainfall is updated in all historical rainfalls, and all the updated historical rainfalls are used as target rainfalls; Obtain the historical temperature interval corresponding to each interval, select the lowest temperature and the highest temperature from each historical temperature interval as a temperature set, and use all temperature sets as target temperatures; When it is detected that the specified time period contains dates corresponding to holidays, the preset date feature vector is used as date data.

[0061] In some possible embodiments, the target water demand for urban water use is obtained according to the target meteorological data and date data, including: The target rainfall and the target temperature are standardized respectively, and the processed target rainfall and the processed target temperature are combined to obtain a meteorological feature vector; The meteorological feature vector and date data are input into the regularized regression model to obtain the target water demand required for urban water use; wherein the regularized regression model is trained by at least two groups of feature vector samples and the historical water demand corresponding to each group of feature vector samples, and each group of feature vector samples includes a meteorological feature vector sample and a date feature vector sample.

[0062] In some possible embodiments, the historical water supply data includes historical scheduling data and historical water supply volume; According to the target water supply of all water plants and the historical water supply data corresponding to each water plant, the target dispatching data of the water pumping station is generated, including: Obtain the historical dispatching data and historical water supply of each water plant in a specified period, and train the Bayesian optimization model based on all the historical dispatching data and all the historical water supply; The target water supply of each water plant is input into the trained Bayesian optimization model to obtain the corresponding pump scheduling data, and the target scheduling data of the pump station is generated based on all the pump scheduling data.

[0063] In some possible embodiments, the water pump scheduling data includes at least one water pump number and a scheduling period corresponding to each water pump number; Generate target dispatch data of the pump station based on all pump dispatch data, including: Integrate all water pump dispatch data to obtain at least one dispatch period corresponding to each water pump number; When it is detected that all the dispatching periods corresponding to any water pump number have overlapping periods, the period ratio between the overlapping period and all the corresponding dispatching periods is calculated; Arrange all time period proportions according to a preset proportion order, and update all corresponding scheduling time periods according to the arrangement result, so that there is no overlapping time period between all the scheduling time periods after processing; Generate target dispatching data of the pumping station according to all dispatching periods corresponding to all pump numbers.

[0064] In some possible embodiments, the water pump scheduling data further includes the water output corresponding to each water pump number; According to the target dispatching data and the historical operating parameters of the water pumping station, at least two water pump status parameters corresponding to each water plant are obtained, including: All pump numbers corresponding to each water plant are screened out from the target dispatching data, and based on all pump numbers corresponding to each water plant, the historical operating parameters of the water pump station are divided and processed to obtain the corresponding historical operating parameters; The historical dispatching data is divided and processed based on all the pump numbers corresponding to each water plant to obtain the corresponding historical water output, and the decision tree learning model is trained based on the historical operating parameters and historical water output corresponding to each water plant; The water output corresponding to all the water pump numbers of each water plant is input into the trained decision tree learning model to obtain corresponding at least two water pump state parameters.

[0065] In some possible embodiments, at least two total operating powers of the water pump station are generated based on the water pump status parameters corresponding to all water plants, including: According to all the water pump state parameters corresponding to each water plant, at least two state parameter sets of the water pump station are combined; wherein each state parameter set includes any one water pump state parameter corresponding to each water plant; According to the water pump state parameters corresponding to each water plant in each state parameter set, all water pump numbers and the scheduling period corresponding to each water pump number, the operating time corresponding to each water pump number is obtained; The operating time corresponding to each water pump number and the preset pump station power are multiplied, and all product results are summed to obtain the total operating power of the corresponding state parameter set.

[0066] See next Figure 4 , Figure 4 A structural schematic diagram of another water plant water volume management device based on machine learning provided in an embodiment of the present application is shown.

[0067] like Figure 4As shown, the water plant water quantity management device 400 based on machine learning may include at least one processor 401, at least one network interface 404, a user interface 403, a memory 405 and at least one communication bus 402.

[0068] The communication bus 402 may be used to realize the connection and communication among the above-mentioned components.

[0069] The user interface 403 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0070] The network interface 404 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0071] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts of the water plant water management device 400 based on machine learning, and executes various functions and processes data of the water plant water management device 400 based on machine learning by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 401 can integrate one or a combination of CPU, GPU, modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application program; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented by a single chip.

[0072] The memory 405 may include a RAM or a ROM. Optionally, the memory 405 includes a non-transitory computer-readable medium. The memory 405 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may optionally be at least one storage device located away from the aforementioned processor 401. As Figure 4 As shown, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a water plant water volume management application based on machine learning.

[0073] Specifically, the processor 401 can be used to call the water plant water volume management application based on machine learning stored in the memory 405, and specifically perform the following operations: Determine the target meteorological data and date data within the specified period, and obtain the target water demand required for urban water use based on the target meteorological data and date data; At least one water plant is numbered according to a preset numbering method, and the target water demand and the numbering result are input into a machine learning model to obtain the target water supply of each water plant, and the target scheduling data of the water pump station is generated according to the target water supply of all water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; According to the target dispatching data and the historical operating parameters of the water pumping station, at least two water pump status parameters corresponding to each water plant are obtained, and at least two total operating powers of the water pumping station are generated based on all the water pump status parameters; All water pump status parameters and target scheduling data corresponding to the minimum total operating power are sent to the water pump station, so that the water pump station outputs water to all water plants according to all corresponding water pump status parameters and target scheduling data.

[0074] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0075] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0076] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0077] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0078] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0079] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

Claims

1. A water plant water volume management method based on machine learning, characterized in that: include: Determine target meteorological data and date data within a specified period, and obtain target water demand required for urban water use based on the target meteorological data and date data; At least one water plant is numbered according to a preset numbering method, the target water demand and the numbering result are input into a machine learning model to obtain the target water supply of each water plant, and the target scheduling data of the water pump station is generated according to the target water supply of all the water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; According to the target scheduling data and the historical operating parameters of the water pumping station, at least two water pump status parameters corresponding to each of the water plants are obtained, and at least two total operating powers of the water pumping station are generated based on all the water pump status parameters; All the water pump status parameters and the target scheduling data corresponding to the minimum total operating power are sent to the water pump station, so that the water pump station outputs water to all the water plants according to all the corresponding water pump status parameters and the target scheduling data.

2. The method according to claim 1, characterized in that The target meteorological data includes target rainfall and target temperature; Determining target meteorological data and date data within a specified period includes: Evenly divide the specified time period into at least two interval periods, and obtain at least two historical rainfall amounts corresponding to all the interval periods; When there is no corresponding historical rainfall in any of the interval periods, obtaining supplementary rainfall according to all the historical rainfalls corresponding to all the interval periods adjacent to the interval period; According to the arrangement of all the intervals, the supplementary rainfall is updated in all the historical rainfalls, and all the updated historical rainfalls are used as target rainfalls; Obtaining the historical temperature interval corresponding to each of the interval periods, selecting the lowest temperature and the highest temperature from each of the historical temperature intervals as a temperature set, and taking all of the temperature sets as target temperatures; When it is detected that the specified time period includes a date corresponding to a holiday, a preset date feature vector is used as date data.

3. The method according to claim 2, characterized in that The step of obtaining the target water demand for urban water use according to the target meteorological data and the date data includes: Standardizing the target rainfall and the target temperature respectively, and merging the processed target rainfall and the processed target temperature to obtain a meteorological feature vector; The meteorological feature vector and the date data are input into a regularized regression model to obtain the target water demand required for urban water use; wherein the regularized regression model is trained by at least two groups of feature vector samples and the historical water demand corresponding to each group of feature vector samples, and each group of feature vector samples includes a meteorological feature vector sample and a date feature vector sample.

4. The method according to claim 1, characterized in that: The historical water supply data includes historical scheduling data and the historical water supply volume; The target dispatching data of the water pump station is generated according to the target water supply of all the water plants and the historical water supply data corresponding to each of the water plants, including: Obtaining the historical dispatching data and the historical water supply corresponding to each of the water plants in the specified time period, and training the Bayesian optimization model based on all the historical dispatching data and all the historical water supply; The target water supply of each water plant is input into the trained Bayesian optimization model to obtain corresponding water pump scheduling data, and the target scheduling data of the water pump station is generated according to all the water pump scheduling data.

5. The method according to claim 4, characterized in that The water pump scheduling data includes at least one water pump number and a scheduling period corresponding to each water pump number; The generating target dispatching data of the water pump station according to all the water pump dispatching data comprises: Integrate all the water pump scheduling data to obtain at least one scheduling period corresponding to each water pump number; When it is detected that all the scheduling time periods corresponding to any one of the water pump numbers have overlapping time periods, the time period ratio between the overlapping time period and all the corresponding scheduling time periods is calculated; Arrange all the time period proportions according to a preset proportion order, and update all the corresponding scheduling time periods according to the arrangement result, so that there is no overlapping time period between all the scheduling time periods after processing; The target scheduling data of the water pump station is generated according to all the scheduling time periods corresponding to all the water pump numbers.

6. The method according to claim 5, characterized in that The water pump scheduling data also includes the water output corresponding to each water pump number; The at least two water pump status parameters corresponding to each water plant are obtained according to the target scheduling data and the historical operation parameters of the water pump station, including: All the water pump numbers corresponding to each of the water plants are screened out from the target scheduling data, and based on all the water pump numbers corresponding to each of the water plants, historical operating parameters of the water pump stations are divided and processed to obtain corresponding historical operating parameters; The historical dispatching data is divided and processed based on all the water pump numbers corresponding to each of the water plants to obtain the corresponding historical water output, and a decision tree learning model is trained based on the historical operating parameters and the historical water output corresponding to each of the water plants; The water output corresponding to all the water pump numbers of each water plant is input into the trained decision tree learning model to obtain corresponding at least two water pump state parameters.

7. The method according to claim 6, characterized in that The generating at least two total operating powers of the water pump station based on all the water pump state parameters comprises: According to all the water pump status parameters corresponding to each of the water plants, at least two status parameter sets of the water pump station are combined; wherein each of the status parameter sets includes any one of the water pump status parameters corresponding to each of the water plants; According to the water pump state parameter corresponding to each water plant in each state parameter set, all the water pump numbers and the scheduling period corresponding to each water pump number, the operating time corresponding to each water pump number is obtained; The operating time corresponding to each of the water pump numbers and the preset pump station power are multiplied, and all the product results are summed to obtain the total operating power of the corresponding state parameter set.

8. A water plant water volume management device based on machine learning, characterized in that: include: A water demand prediction module is used to determine target meteorological data and date data within a specified period of time, and obtain the target water demand required for urban water use based on the target meteorological data and the date data; A water pump scheduling prediction module, used to number at least one water plant according to a preset numbering method, input the target water demand and the numbering result into a machine learning model to obtain the target water supply of each water plant, and generate target scheduling data for the water pump station according to the target water supply of all the water plants and the historical water supply data corresponding to each water plant; wherein the machine learning model is trained by at least two historical water demands, the water plant number corresponding to each historical water demand, and the historical water supply corresponding to each water plant; A water pump state prediction module, used to obtain at least two water pump state parameters corresponding to each of the water plants according to the target scheduling data and the historical operating parameters of the water pump station, and generate at least two total operating powers of the water pump station based on all the water pump state parameters; The water volume management module is used to send all the water pump status parameters corresponding to the minimum total operating power and the target scheduling data to the water pump station, so that the water pump station outputs water to all the water plants according to all the corresponding water pump status parameters and the target scheduling data.

9. A water plant water volume management device based on machine learning, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the steps of the method according to any one of claims 1 to 7.

Images