AI electric meter operation management method for electric appliance configuration

Abstract

The invention relates to an AI electric meter operation management method for electric appliance configuration. The method comprises the following steps: using a commodity sales prediction model corresponding to a set type of commodity at an AI electric meter to intelligently predict a predicted after-sale time when the set type of commodity is completely sold; and various production-related electric appliances of a production workshop for producing the set-type commodities are configured for the set-type commodities in advance at the AI electric meter, and are automatically started at the predicted after-sale time when the set-type commodities are completely sold. The AI electric meter operation management method for electric appliance configuration is intelligent in control and stable in operation. As the predicted after-sale time when the set type of commodities are completely sold can be intelligently predicted in an artificial intelligence mode at the AI electric meter of the production workshop for producing the set type of commodities, key data can be provided for the AI electric meter to perform electric appliance management and production line management on the production workshop for producing the set type of commodities; and the electric appliances and the automatic production line in the production workshop are prevented from being started too early or too late.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of AI electric meters, in particular to an AI electric meter operation management method for electric appliance configuration. BACKGROUND

[0002] AI electric meters are the product of the deep integration of traditional smart meters and artificial intelligence technology, and have evolved from simple electric energy metering tools to core components of energy management. AI electric meters can be used for data intelligent analysis, such as real-time analysis of current, voltage, power and other parameters through deep learning models, identification of electric appliance characteristics such as step power curve of electric vehicle charging and continuous high load characteristics, establishment of user electricity portrait, accurate differentiation of use modes of air conditioners, lighting and other devices, and optimization of energy consumption strategy. AI electric meters can also be used to perform multi-modal risk prevention and control, such as abnormal early warning, automatic blocking or image linkage.

[0003] However, for AI electric meters installed at production workshops that produce set type commodities, their functions are still limited to monitoring and managing the electricity of various electric devices in the production workshop, and cannot perform deeper data intelligent analysis, for example, cannot predict the predicted sell-out time of the set type commodities at the AI electric meter according to various data to intelligently predict the set type commodities, resulting in the inability to configure the automatic start time of each electric appliance in the production workshop to provide services for production start, and the inability to configure the automatic start time of the automatic production line in the production workshop. SUMMARY

[0004] In order to solve the technical problems in the related art, the present application provides an AI electric meter operation management method for electric appliance configuration, which can intelligently predict the predicted sell-out time of the set type commodities at the AI electric meter of the production workshop that produces set type commodities according to the respective parts of commodity associated information corresponding to each past time of the set type commodities, the multiple related data of the set type commodities in the set type commodity sales area, and the number of similar commodities of the set type commodities in the set type commodity sales area, so as to provide key data for the electric appliance management and production line management of the AI electric meter for the production workshop that produces set type commodities, and avoid the premature or late start of each electric appliance and automatic production line in the production workshop.

[0005] According to the present application, an AI electric meter operation management method for electric appliance configuration is provided, which comprises: Obtain a plurality of related data of a dumping area of a set type commodity at an AI electric meter, and obtain a plurality of pieces of commodity associated information corresponding to each time point before the current time point of the set type commodity, the plurality of related data of the dumping area of the set type commodity is the number of permanent residents of the dumping area of the set type commodity, the area geographical area, the proportion of personnel using the set type commodity, and the number of dumping points, and the commodity associated information corresponding to each time point of the set type commodity is the inventory quantity and the commodity consumption quantity of the set type commodity at the time point, and the AI electric meter is arranged at a production workshop for producing the set type commodity; Obtain a commodity sales prediction model corresponding to the set type commodity at the AI electric meter, the commodity sales prediction model corresponding to the set type commodity is based on the network architecture of a radial basis neural network and is a radial basis neural network after performing each learning action, and the number of learning actions performed by the radial basis neural network is monotonically positively correlated with the number of permanent residents of the dumping area of the set type commodity; Synchronously input the plurality of related data of the dumping area of the set type commodity, the plurality of pieces of commodity associated information corresponding to each time point before the current time point of the set type commodity, and the number of similar commodities of the set type commodity in the dumping area of the set type commodity into the commodity sales prediction model corresponding to the set type commodity at the AI electric meter, to run the commodity sales prediction model corresponding to the set type commodity, and obtain a predicted finished selling time point of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity; The AI electric meter is arranged in advance for the set type commodity, and each type of production associated electric appliance of the production workshop for producing the set type commodity is automatically started at the predicted finished selling time point of the set type commodity. The AI electric meter is arranged in advance for the set type commodity, and each type of production associated electric appliance of the production workshop for producing the set type commodity is automatically started at the predicted finished selling time point of the set type commodity.

[0006] Therefore, the present application has the following several outstanding technical effects: Technical effect A: At the AI electric meter of the production workshop for producing the set type commodity, the predicted finished selling time point of the set type commodity is intelligently predicted according to the plurality of related data of the dumping area of the set type commodity, the plurality of pieces of commodity associated information corresponding to each time point of the set type commodity, and the number of similar commodities of the set type commodity in the dumping area of the set type commodity, thereby providing key data for the AI electric meter to manage the electric appliances and production lines of the production workshop for producing the set type commodity. Technical Effect B: A sales forecasting model corresponding to a set type of goods is introduced to perform intelligent prediction of the time when the set type of goods will be sold out. The sales forecasting model corresponding to the set type of goods is based on the network architecture of radial basis neural network and is a radial basis neural network after each learning action. The number of learning actions performed by the radial basis neural network is monotonically positively correlated with the number of permanent residents in the dumping area of ​​the set type of goods. In addition, the number of selected past moments is proportional to the geographical area of ​​the dumping area of ​​the set type of goods. Thus, different structures of sales forecasting models are customized for different types of goods. Technical Effect C: The AI ​​electricity meter pre-configures various production-related electrical appliances in the production workshop used to produce the specified type of goods. These appliances automatically start when the predicted sell-out time for the specified type of goods is reached. These production-related electrical appliances include air conditioners, lighting equipment, and air compressors in the production workshop used to produce the specified type of goods. The AI ​​electricity meter also pre-configures the automated production line in the production workshop used to produce the specified type of goods. These appliances automatically start when the predicted sell-out time for the specified type of goods is reached, thus completing automated electrical appliance management and production line management in the production workshop based on intelligent prediction results.

[0007] The AI ​​meter operation management method for electrical appliance configuration of the present invention is intelligent and stable in operation. Because it can intelligently predict the sell-out time of a set type of goods at the AI ​​meter in the production workshop using artificial intelligence, it provides key data for the AI ​​meter to manage the electrical appliances and production lines in the production workshop, preventing the electrical appliances and automated production lines in the workshop from starting too early or too late. Detailed Implementation

[0008] The embodiments of the AI ​​meter operation management method for electrical appliance configuration of the present invention will be described in detail below.

[0009] Embodiment one of the present application The AI ​​meter operation management method for appliance configuration shown in Embodiment 1 of the present invention specifically includes the following steps: The AI ​​meter obtains multiple relevant data on the dumping area of ​​a set type of product and the product association information corresponding to each time before the current time. The multiple relevant data on the dumping area of ​​the set type of product include the number of permanent residents in the dumping area, the geographical area of ​​the area, the proportion of people using the set type of product, and the number of dumping points. The product association information corresponding to the set type of product at each time is the inventory quantity and consumption quantity of the set type of product at that time. The AI ​​meter is configured in the production workshop where the set type of product is produced. For example, at the AI electric meter, multiple related data of the dumping area of the set type commodity and each part of the commodity association information corresponding to each time before the current time of the set type commodity are obtained. The multiple related data of the dumping area of the set type commodity is the number of permanent residents in the dumping area of the set type commodity, the area geographical area, the proportion of personnel using the set type commodity, and the number of dumping points. The commodity association information corresponding to each time of the set type commodity is the inventory quantity and the commodity consumption quantity of the set type commodity at that time. The AI electric meter configured at the production workshop for producing the set type commodity includes: the set type commodity can be a consumer commodity such as cola and juice, and the dumping area of the set type commodity is the dumping city bound to the consumer commodity such as cola and juice; At the AI electric meter, a commodity sales prediction model corresponding to the set type commodity is obtained. The commodity sales prediction model corresponding to the set type commodity is based on the network architecture of a radial basis neural network and is a radial basis neural network after performing each learning action. The number of learning actions performed by the radial basis neural network is monotonically positively correlated with the number of permanent residents in the dumping area of the set type commodity; At the AI electric meter, the multiple related data of the dumping area of the set type commodity, each part of the commodity association information corresponding to each time before the current time of the set type commodity, and the number of similar commodities of the set type commodity in its dumping area are synchronously input into the commodity sales prediction model corresponding to the set type commodity to run the commodity sales prediction model corresponding to the set type commodity, and the predicted finished selling time of the set type commodity corresponding to the commodity sales prediction model output is obtained. At the AI electric meter, each type of production association electric appliance of the production workshop for producing the set type commodity is automatically started at the predicted finished selling time of the set type commodity. At the AI electric meter, each type of production association electric appliance of the production workshop for producing the set type commodity is automatically started at the predicted finished selling time of the set type commodity, including: each type of production association electric appliance of the production workshop for producing the set type commodity is each air conditioner, each lighting device, and each air compressor in the production workshop for producing the set type commodity. And wherein, at the AI electric meter, a plurality of related data of the dumping area of the set type commodity is acquired, and a plurality of pieces of commodity associated information corresponding to each time point respectively before the current time point of the set type commodity is acquired, the plurality of related data of the dumping area of the set type commodity is the number of permanent residents in the dumping area of the set type commodity, the area geographical area, the proportion of personnel using the set type commodity, and the number of dumping points, and the commodity associated information corresponding to each time point of the set type commodity is the inventory quantity and the commodity consumption quantity of the set type commodity at the time point.

[0010] Embodiment two of the present application Compared with the first embodiment of the application, the AI electric meter operation management method for electric appliance configuration according to the second embodiment of the application further comprises the following steps: The AI electric meter is configured in advance for the set type commodity to automatically start the automatic production line of the production workshop for producing the set type commodity at the predicted sold-out time point of the set type commodity. Among them, the AI electric meter is configured in advance for the set type commodity to automatically start the automatic production line of the production workshop for producing the set type commodity at the predicted sold-out time point of the set type commodity, comprising: sending a production start signal to the automatic production line of the production workshop for producing the set type commodity at the predicted sold-out time point of the set type commodity, so as to trigger the automatic production line to automatically start at the predicted sold-out time point of the set type commodity.

[0011] Embodiment three of the present application Compared with the first embodiment of the application, the AI electric meter operation management method for electric appliance configuration according to the third embodiment of the application further comprises the following steps: The AI electric meter receives the predicted sold-out time point of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity, and transmits the predicted sold-out time point of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity to the remote commodity sales management server through a mobile communication network. Among them, the AI electric meter receives the predicted sold-out time point of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity, and transmits the predicted sold-out time point of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity to the remote commodity sales management server through a mobile communication network.

[0012] Next, the specific steps of the AI electric meter operation management method for electric appliance configuration according to the application will be further described.

[0013] In the AI electric meter operation management method for electric appliance configuration according to various embodiments of the present application: At the AI electric meter, a commodity sales prediction model corresponding to a set type commodity is acquired, the commodity sales prediction model corresponding to the set type commodity is based on a network architecture of a radial basis neural network and is a radial basis neural network after each learning action is performed, the number of learning actions performed by the radial basis neural network is monotonically positively correlated with the number of permanent residents in the dumping area of the set type commodity, and the content conversion relationship between the number of learning actions performed by the radial basis neural network and the number of permanent residents in the dumping area of the set type commodity is expressed by using a content conversion formula. Specifically, the test and simulation of the data processing process of the content conversion relationship between the number of learning actions performed by the radial basis neural network and the number of permanent residents in the dumping area of the set type commodity expressed by using the content conversion formula can be completed by using the MATLAB toolbox. In the content conversion relationship between the number of learning actions performed by the radial basis neural network and the number of permanent residents in the dumping area of the set type commodity expressed by using the content conversion formula, the number of permanent residents in the dumping area of the set type commodity is the input content of the content conversion formula. In the content conversion relationship between the number of learning actions performed by the radial basis neural network and the number of permanent residents in the dumping area of the set type commodity expressed by using the content conversion formula, the number of learning actions performed by the radial basis neural network is the output content of the content conversion formula.

[0014] In the AI electric meter operation management method for electric appliance configuration according to various embodiments of the present application: At the AI electric meter, a commodity sales prediction model corresponding to a set type commodity is acquired, the commodity sales prediction model corresponding to the set type commodity is based on a network architecture of a radial basis neural network and is a radial basis neural network after each learning action is performed, the number of learning actions performed by the radial basis neural network is monotonically positively correlated with the number of permanent residents in the dumping area of the set type commodity, and the content conversion relationship between the number of learning actions performed by the radial basis neural network and the number of permanent residents in the dumping area of the set type commodity is expressed by using a content conversion formula.

[0015] The method further comprises: synchronously inputting, at the AI electric meter, a plurality of related data of a dumping area of a set type commodity, each portion of commodity association information corresponding to each time point before a current time point of the set type commodity, and a number of same type commodities of the set type commodity in the dumping area of the set type commodity into a commodity sales prediction model corresponding to the set type commodity, to run the commodity sales prediction model corresponding to the set type commodity, and obtain a predicted sold-out time point of the set type commodity in the dumping area of the set type commodity output by the commodity sales prediction model corresponding to the set type commodity, wherein the input content and the output content of the commodity sales prediction model corresponding to the set type commodity are in a numerical representation form in hexadecimal.

[0016] In addition, in the AI electric meter operation management method for electric appliance configuration, the plurality of related data of the dumping area of the set type commodity and the each portion of commodity association information corresponding to each time point before the current time point of the set type commodity are obtained at the AI electric meter, the plurality of related data of the dumping area of the set type commodity are the number of permanent residents in the dumping area of the set type commodity, the area geographical area, the proportion of personnel using the set type commodity, and the number of dumping points, and the commodity association information corresponding to each time point of the set type commodity is the inventory quantity and the commodity consumption quantity of the set type commodity at the time point, and the method further comprises: adopting a numerical mapping function to represent a numerical mapping relationship that the number of time points of each time point before the selected current time point is proportional to the area geographical area of the dumping area of the set type commodity.

[0017] While the foregoing detailed description has set forth what are believed to be the most preferred embodiments of the application, at the same time it should be understood that numerous modifications can be made thereto without departing from the spirit of the application and that individual features of the application can be adapted to different embodiments as the skilled artisan will readily appreciate. Accordingly, the novel system and method are not to be limited to the precise details of methodology or construction set forth above as such can readily vary in light of the above teachings. It is thus contemplated that the application or any other such systems or methods can include any number of the novel features discussed above, which can all independently be combined in any combination.

Claims

1. A method for managing the operation of an AI-powered electricity meter for appliance configuration, characterized in that, The method includes: The AI ​​meter obtains multiple relevant data on the dumping area of ​​a set type of product and the product association information corresponding to each time before the current time. The multiple relevant data on the dumping area of ​​the set type of product include the number of permanent residents in the dumping area, the geographical area of ​​the area, the proportion of people using the set type of product, and the number of dumping points. The product association information corresponding to the set type of product at each time is the inventory quantity and consumption quantity of the set type of product at that time. The AI ​​meter is configured in the production workshop where the set type of product is produced. The AI ​​meter obtains the sales prediction model for the specified type of goods. The sales prediction model for the specified type of goods is based on the radial basis function neural network architecture and is the radial basis function neural network after each learning action. The number of learning actions performed by the radial basis function neural network is monotonically positively correlated with the number of permanent residents in the dumping area of ​​the specified type of goods. At the AI ​​meter, multiple relevant data of the dumping area of ​​the set type of goods, the corresponding product association information of the set type of goods at each time before the current time, and the quantity of the same type of goods of the set type of goods in its dumping area are simultaneously input into the sales prediction model of the set type of goods, so as to run the sales prediction model of the set type of goods and obtain the predicted time when the set type of goods will be dumped. The AI ​​meter pre-configures various production-related electrical appliances in the production workshop used to produce the specified type of goods, which will automatically start when the specified type of goods are predicted to be sold out. Among them, the AI ​​electricity meter pre-configures various production-related electrical appliances in the production workshop used to produce the set type of goods, which will automatically start when the set type of goods are sold out as predicted. These include: various production-related electrical appliances in the production workshop used to produce the set type of goods, such as each air conditioner, each lighting device, and each air compressor in the production workshop used to produce the set type of goods.

2. The AI ​​meter operation management method for electrical appliance configuration as described in claim 1, characterized in that: The AI ​​meter retrieves multiple relevant data points for the dumping area of ​​a set type of product, as well as product association information for each time point before the current time. The multiple relevant data points for the dumping area include the number of permanent residents, the geographical area of ​​the area, the percentage of people using the set type of product, and the number of dumping points. The product association information for each time point includes the inventory quantity and consumption quantity of the set type of product at that time. The number of time points before the selected current time is proportional to the geographical area of ​​the dumping area of ​​the set type of product.

3. The AI ​​meter operation management method for electrical appliance configuration as described in claim 2, characterized in that, The method further includes: The AI ​​meter pre-configures an automated production line for a designated type of product in a production workshop that produces that product. This line automatically starts when the designated type of product is predicted to be sold out.

4. The AI ​​meter operation management method for electrical appliance configuration as described in claim 2, characterized in that, The method further includes: Receive the predicted sell-out time of the set type of goods from the sales forecast model corresponding to the set type of goods, and transmit the predicted sell-out time of the set type of goods to the remote sales management server through the mobile communication network. The process of transmitting the predicted sell-out time of the set type of goods, output by the sales forecast model corresponding to the set type of goods, to the remote sales management server via a mobile communication network includes: the mobile communication network being based on time-division duplex communication mode or frequency-division duplex communication mode.

5. The AI ​​meter operation management method for electrical appliance configuration as described in any one of claims 2-4, characterized in that: The AI ​​meter obtains the sales prediction model for a set type of goods. The sales prediction model for the set type of goods is based on the radial basis function neural network architecture and is a radial basis function neural network after each learning action. The number of learning actions performed by the radial basis function neural network and the number of permanent residents in the dumping area of ​​the set type of goods are monotonically positively correlated. This includes: using a content transformation formula to express the content transformation relationship between the number of learning actions performed by the radial basis function neural network and the number of permanent residents in the dumping area of ​​the set type of goods.

6. The AI ​​meter operation management method for electrical appliance configuration as described in claim 5, characterized in that: The content transformation formula expressing the monotonically positive correlation between the number of learning actions performed by the radial basis function neural network and the number of permanent residents in the dumping area of ​​a set type of commodity includes: in the content transformation formula, the number of permanent residents in the dumping area of ​​the set type of commodity is the input content of the content transformation formula.

7. The AI ​​meter operation management method for electrical appliance configuration as described in claim 6, characterized in that: The content transformation formula expressing the monotonically positive correlation between the number of learning actions performed by the radial basis function neural network and the number of permanent residents in the dumping area of ​​a set type of commodity includes: in the content transformation formula, the number of learning actions performed by the radial basis function neural network that is monotonically positively correlated with the number of permanent residents in the dumping area of ​​a set type of commodity is the output content of the content transformation formula.

8. The AI ​​meter operation management method for electrical appliance configuration as described in any one of claims 2-4, characterized in that: At the AI ​​meter, multiple relevant data points for the dumping area of ​​a set type of product, the product association information corresponding to each set type of product at each time point before the current time, and the quantity of similar products of the set type of product in its dumping area are simultaneously input into the product sales forecasting model corresponding to the set type of product. The product sales forecasting model corresponding to the set type of product is then run to obtain the predicted time when the set type of product will be sold out. This includes: at the AI ​​meter, using a programmable logic device, multiple relevant data points for the dumping area of ​​a set type of product, the product association information corresponding to each set type of product at each time point before the current time, and the quantity of similar products of the set type of product in its dumping area are simultaneously input into the product sales forecasting model corresponding to the set type of product.

9. The AI ​​meter operation management method for electrical appliance configuration as described in claim 8, characterized in that: At the AI ​​meter, multiple relevant data points for the dumping area of ​​a set type of product, the product association information corresponding to each set of products at each time before the current time, and the quantity of similar products of the set type of product in its dumping area are simultaneously input into the product sales forecasting model corresponding to the set type of product. The product sales forecasting model corresponding to the set type of product is then run to obtain the predicted sell-out time of the set type of product after dumping. The product sales forecasting model corresponding to the set type of product also includes the fact that both the input and output contents of the product sales forecasting model corresponding to the set type of product are in hexadecimal numerical representation.