Method and apparatus for determining optimal ai combinations, electronic device, and program product

Abstract

Embodiments of the present application mainly relate to the field of power systems, and relate in particular to a method and apparatus for determining optimal AI combinations, an electronic device, and a program product. The method comprises: acquiring a first group of power data, and determining a switch event present in the first group of power data; according to a preset condition, dividing the first group of power data into at least two groups; on the basis of each group of data among the at least two groups, respectively acquiring at least one specified feature value corresponding to the switch event to obtain at least one specified data set; and combining each specified data set among the at least one specified data set with each training algorithm among multiple training algorithms to perform training, so as to determine an optimal AI combination corresponding to each group among the at least two groups, respectively, each optimal AI combination comprising a specified data set and a training algorithm.

Description

Method, device, electronic device and program product for determining optimal AI combination Technical Field

[0001] The embodiments of the present application mainly relate to the field of power systems, and in particular to a method, device, electronic device, and program product for determining an optimal artificial intelligence (AI) combination. Background Art

[0002] Load monitoring refers to the real-time or periodic monitoring and recording of individual loads in a power system to obtain information such as their power usage, power demand, and energy consumption characteristics. Load monitoring can help users understand load power usage behavior, optimize energy management, improve energy efficiency, and perform load scheduling and capacity planning. Load monitoring can be applied in a variety of scenarios, including residential, commercial, and industrial plants. By understanding load power usage, users can manage and optimize energy consumption, achieving goals such as reducing energy consumption, load scheduling, and equipment capacity planning. Some traditional load monitoring methods require sensors or recording devices installed on the loads for monitoring. To improve accuracy and expand device coverage, these methods require additional functionality, which increases complexity. However, more features and a more complex implementation mean increased workflow and costs.

[0003] Summary of the Invention

[0004] The embodiments of the present application provide a method, device, electronic device, and program product for determining an optimal AI combination. The optimal AI combination determined by the embodiments of the present application facilitates accurate identification of relevant on-site equipment.

[0005] In a first aspect, a method for determining an optimal AI combination is provided, comprising: obtaining a first set of power data and determining a switching event present in the first set of power data; dividing the first set of power data into at least two groups according to preset conditions; based on each set of data in the at least two groups, respectively obtaining at least one specified characteristic value corresponding to the switching event to obtain at least one specified data set; combining each specified data set in the at least one specified data set with each training algorithm in a plurality of training algorithms for training to determine an optimal AI combination corresponding to each group in the at least two groups, wherein the optimal AI combination comprises a specified data set and a training algorithm.

[0006] In a second aspect, a device for determining an optimal AI combination is provided, comprising: a first acquisition module, configured to: acquire a first group of power data, and determine the switching events present in the first group of power data; a grouping module, configured to: divide the first group of power data into at least two groups according to preset conditions; a second acquisition module, configured to: based on each group of data in the at least two groups, respectively acquire at least one specified characteristic value corresponding to the switching event, and obtain at least one specified data set; a training module, configured to: combine each specified data set in the at least one specified data set with each training algorithm in a plurality of training algorithms for training, so as to determine the optimal AI combination corresponding to each group in the at least two groups, wherein the optimal AI combination comprises a specified data set and a training algorithm.

[0007] In a third aspect, an electronic device is provided, comprising: at least one memory configured to store computer-readable code; and at least one processor configured to call the computer-readable code and execute each step of the method provided in the first aspect.

[0008] In a fourth aspect, a computer storage medium is provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the processor executes each step in the method provided in the first aspect.

[0009] In a fifth aspect, a computer program product is provided, which is tangibly stored on a computer-readable medium and includes computer-executable instructions, which, when executed, cause at least one processor to perform the steps in the method provided in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The following figures are intended only to illustrate and explain the embodiments of the present application and are not intended to limit the scope of the embodiments of the present application.

[0011] FIG1 is a schematic diagram of a method for determining an optimal AI combination according to an embodiment of the present application;

[0012] FIG2 is a schematic diagram of load clustering according to an embodiment of the present application;

[0013] FIG3 is a schematic diagram of a device for determining an optimal AI combination according to an embodiment of the present application;

[0014] FIG4 is a schematic diagram of an electronic device according to an embodiment of the present application.

[0015] Description of Reference Numerals

[0016] 100: Method for Determining the Optimal AI Combination 101-104: Method Step 31: First Acquisition Module

[0017] 32: Grouping module 33: Second acquisition module 34: Training module

[0018] 400: Electronic device 401: Processor 402: Communication interface

[0019] 403: Memory 404: Communication bus 405: Program DETAILED DESCRIPTION

[0020] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that discussing these embodiments is merely to enable those skilled in the art to better understand and implement the subject matter described herein, and is not intended to limit the scope of protection, applicability, or examples set forth in the claims. The functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the embodiments of the present application. Various examples may omit, replace, or add various processes or components as needed. For example, the described method may be performed in an order different from the described order, and various steps may be added, omitted, or combined. In addition, the features described relative to some examples may also be combined in other examples.

[0021] As used herein, the term "including" and its variations are open terms meaning "including but not limited to". The term "based on" means "based at least in part on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other definitions may be included below, whether explicit or implicit. Unless the context clearly indicates otherwise, the definition of a term is consistent throughout the specification.

[0022] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0023] FIG1 is a schematic diagram of a method for determining an optimal AI combination according to an embodiment of the present application. As shown in FIG1 , the method 100 for determining an optimal AI combination includes:

[0024] Step 101: Acquire a first set of power data and determine switching events present in the first set of power data.

[0025] Step 102: Divide the first group of power data into at least two groups according to preset conditions.

[0026] Among them, in the first group of power data, clustering can be performed according to the power values ​​corresponding to the switching events to obtain at least two groups.

[0027] Optionally, the preset conditions include: specified power parameters of linearity, nonlinearity, resistance, inductance, low-frequency characteristics, or high-frequency characteristics.

[0028] In one embodiment, as shown in FIG2 , the horizontal axis is active power P, and the vertical axis is reactive power Q1. The first group of power data is clustered according to the power values ​​corresponding to the existing switching events, and two groups of data are obtained. One group D1 is selected from the part between [15W, 90W], such as fans, refrigerators, incandescent bulbs, etc., and the other group D2 can be the rest, such as blenders, refrigerator defrosters, vacuum cleaners, electric kettles, coffee machines, etc.

[0029] Step 103 : Based on each set of data in the at least two groups, respectively obtain at least one designated characteristic value corresponding to the switching event to obtain at least one designated data set.

[0030] Optionally, any designated characteristic value of the at least one designated characteristic value includes: a single power parameter or a combination of multiple power parameters.

[0031] In one embodiment, a single power parameter is such as active power P or reactive power Q, and a combination of multiple power parameters is such as active power P+reactive power Q, or active power P+current I.

[0032] In one embodiment, any one of the at least one specified characteristic value includes: active power / active power and reactive power / reactive power / active power and current.

[0033] Step 104: Combine each of the at least one specified data set with each of the multiple training algorithms for training to determine an optimal AI combination corresponding to each of the at least two groups, wherein the optimal AI combination includes a specified data set and a training algorithm.

[0034] The first set of power data includes historical power data with device name tags and / or relevant power data collected in real time.

[0035] Multiple training algorithms include: nearest neighbor, naive Bayes, regression, support vector machine with Gaussian radial basis kernel function, latent Dirichlet allocation, quadratic discriminant, decision tree, random forest, and adaptive boosting.

[0036] In one embodiment, assume there are a light bulb A and an electric kettle B. By default, light bulb A and electric kettle B are off. During the first time period, light bulb A is turned on first, which is event one. Next, electric kettle B is turned on, which is event two. Again, light bulb A is turned off, which is event three. Finally, electric kettle B is turned off, which is event four. The specified feature values ​​corresponding to events one through four during the first time period are used as the training set. The above example is merely illustrative; the number of samples in the training set is much larger than the example in this embodiment. During the second time period, light bulb A is turned on first, which is event one. Next, electric kettle B is turned on, which is event two. Again, light bulb A is turned off, which is event three. Finally, electric kettle B is turned off, which is event four. The specified feature values ​​corresponding to events one through four during the second time period are used as the test set.

[0037] The embodiment of the present application divides a set of power data into multiple groups, and then trains each group separately to obtain the optimal AI combination, so that when the corresponding optimal AI combination is used to actually predict the name of the relevant equipment, the accuracy can be greatly improved.

[0038] In one embodiment, after step 104, power data for a time period is obtained. Whether a switching event exists is detected. If a switching event exists, an optimal AI combination is determined to output the names of the devices that experienced the switching event during the time period.

[0039] Specifically, a specified characteristic value corresponding to a switching event is determined, where the specified characteristic value includes an active power value. Based on the specified characteristic value, a corresponding group is determined. Based on the group, an optimal AI combination is determined. A specified data set within the optimal AI combination is determined from the power data for the time period to obtain a second specified data set. The second specified data set is then combined with the training algorithm within the optimal AI combination for training. The training algorithm outputs the names of devices that experienced switching events during the time period.

[0040] Optionally, when the jump difference is relatively large, it is determined that a switching event exists. Optionally, the power data of a time period refers to relevant power data of a period of time on site, which may include current, voltage, active power, reactive power, etc.

[0041] Optionally, the second designated data set may be directly extracted from the power data, or may be obtained through calculation.

[0042] The embodiment of the present application determines the corresponding grouping based on the specified characteristic value corresponding to the existing switch event, and then determines the corresponding optimal AI combination. The algorithm model obtained by training the optimal AI combination can accurately identify the device name corresponding to the existing switch event. It can be seen that the embodiment of the present application does not need to deploy additional equipment that will increase the complexity and cost of installation, and adaptively switches the collected field data to the most appropriate load identification path. It is effective for both single-operation and overlapping-operation devices, and therefore has high practicality. In addition, by introducing field data into the training data set, good scalability is achieved. The embodiment of the present application helps to further perform load monitoring on different devices.

[0043] FIG3 is a schematic diagram of a device for determining an optimal AI combination according to an embodiment of the present application. As shown in FIG3 , the device 30 for determining an optimal AI combination includes:

[0044] The first acquisition module 31 is configured to acquire a first set of power data and determine a switching event present in the first set of power data.

[0045] The grouping module 32 is configured to: divide the first group of power data into at least two groups according to a preset condition.

[0046] The second acquisition module 33 is configured to respectively acquire at least one designated characteristic value corresponding to the switching event based on each set of data in the at least two groups, and obtain at least one designated data set.

[0047] The training module 34 is configured to combine each specified data set in at least one specified data set with each training algorithm in a plurality of training algorithms for training to determine an optimal AI combination corresponding to each group in at least two groups, wherein the optimal AI combination includes a specified data set and a training algorithm.

[0048] The optimal AI combination determined by the embodiments of the present application helps to accurately identify relevant on-site equipment.

[0049] FIG4 is a schematic diagram of an electronic device according to an embodiment of the present application. The specific embodiments of the present application do not limit the specific implementation of the electronic device. As shown in FIG4 , the electronic device 400 may include: a processor 401, a communications interface 402, a memory 403, and a communication bus 404.

[0050] The processor 401 , the communication interface 402 , and the memory 403 communicate with each other via the communication bus 404 .

[0051] The communication interface 402 is used to communicate with other electronic devices or servers.

[0052] The processor 401 is configured to execute the program 405 , and specifically may execute the relevant steps in any one of the aforementioned method embodiments.

[0053] Specifically, the program 405 may include program codes, which include computer operation instructions.

[0054] Processor 401 may be a CPU, an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs, or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0055] The memory 403 is used to store the program 403. The memory 403 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0056] The program 405 can be specifically used to enable the processor 401 to execute any one of the multiple method embodiments in the aforementioned embodiments.

[0057] The specific implementation of each step in program 405 can be found in the corresponding descriptions of the corresponding steps and units in the aforementioned embodiment of the method for generating a communication channel, and will not be repeated here. Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding process descriptions in the aforementioned method embodiment, and will not be repeated here.

[0058] The present application also provides a computer-readable storage medium storing instructions for causing a machine to perform any of the multiple method embodiments described herein. Specifically, a system or device equipped with a storage medium can be provided, wherein the storage medium stores software program code that implements the functions of any of the above-described embodiments, and a computer (or CPU or MPU) of the system or device can read and execute the program code stored in the storage medium.

[0059] In this case, the program code read from the storage medium itself can realize the function of any one of the above embodiments, so the program code and the storage medium storing the program code constitute part of this application.

[0060] Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0061] An embodiment of the present application also provides a computer program product, including computer instructions, which instruct a computing device to perform any corresponding operation in the above-mentioned multiple method embodiments.

[0062] It should be pointed out that, according to the needs of implementation, the various components / steps described in the embodiments of the present application can be split into more components / steps, or two or more components / steps or partial operations of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of the present application.

[0063] The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as software or computer code that can be stored in a recording medium (such as CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or can be implemented as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded via a network and will be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a special-purpose processor or programmable or special-purpose hardware (such as ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown here, the execution of the code converts the general-purpose computer into a special-purpose computer for executing the method shown here.

[0064] It should be noted that not all steps and modules in the above processes and system structure diagrams are required, and certain steps or modules can be omitted according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or may be implemented by certain components in multiple independent devices.

[0065] In the above embodiments, the hardware module can be implemented mechanically or electrically. For example, a hardware module can include a permanent dedicated circuit or logic (such as a dedicated processor, FPGA or ASIC) to complete the corresponding operation. The hardware module can also include programmable logic or circuits (such as a general-purpose processor or other programmable processors), which can be temporarily set by software to complete the corresponding operation. The specific implementation method (mechanical method, or dedicated permanent circuit, or temporarily set circuit) can be determined based on cost and time considerations.

[0066] The present invention has been shown and described in detail above through the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the scope of protection of the present invention.

[0067] Nouns and pronouns referring to persons in this patent application are not limited to a specific gender.

Claims

1. A method for determining an optimal AI combination, comprising: - obtaining (101) a first set of power data and determining a switching event present in the first set of power data; - dividing (102) the first set of power data into at least two groups according to a preset condition; - based on each set of data in the at least two groups, respectively obtaining (103) at least one designated characteristic value corresponding to the switching event to obtain at least one designated data set; - combining each of the at least one designated data set with each of the plurality of training algorithms for training (104) to determine an optimal AI combination corresponding to each of the at least two groups, wherein the optimal AI combination includes a designated data set and a training algorithm.

2. The method according to claim 1, wherein The first group of power data is divided (102) into at least two groups according to a preset condition, including: - In the first set of power data, clustering is performed according to the power values corresponding to the switching events to obtain at least two groups.

3. The method according to claim 1, wherein The preset conditions include: specified power parameters of linearity, nonlinearity, resistance, inductance, low-frequency characteristics or high-frequency characteristics.

4. The method according to claim 1, wherein After determining the optimal AI combination corresponding to each of the at least two groups, the method further includes: -Get power data for a period of time; -Detect whether there is a switch event; -If there is a switch event, determine the corresponding optimal AI combination to output the name of the device that has the switch event during the time period.

5. The method according to claim 4, wherein The determining of the corresponding optimal AI combination to output the name of the device having the switch event in the time period includes: -determining a specified characteristic value corresponding to the switching event, wherein the specified characteristic value includes: an active power value; -determining corresponding groups according to the specified feature values; -Determine the corresponding optimal AI combination based on the grouping; -determining a designated data set in the optimal AI combination from the power data of the time period to obtain a second designated data set; - combining the second designated data set with the training algorithm in the optimal AI combination for training, and outputting the names of devices having switching events in the time period through the training algorithm.

6. The method according to claim 1, wherein Any of the at least one specified characteristic value includes: a single power parameter or a combination of multiple power parameters.

7. The method according to claim 1, wherein Any of the at least one specified feature value includes: - Active power / Active power and reactive power / Reactive power / Active power and current.

8. A device for determining an optimal AI combination, comprising: - a first acquisition module (31), configured to: acquire a first set of power data, and determine a switching event present in the first set of power data; - a grouping module (32), configured to: divide the first group of power data into at least two groups according to a preset condition; - a second acquisition module (33), configured to: acquire at least one specified characteristic value corresponding to the switching event based on each set of data in the at least two groups, to obtain at least one specified data set; - A training module (34) configured to: combine each of the at least one designated data set with each of the plurality of training algorithms for training, so as to determine an optimal AI combination corresponding to each of the at least two groups, wherein the optimal AI combination includes a designated data set and a training algorithm.

9. An electronic device (400), comprising: A processor (401), a communication interface (402), a memory (403) and a communication bus (404), wherein the processor (401), the memory (403) and the communication interface (402) communicate with each other via the communication bus (404); The memory (403) is used to store at least one executable instruction, and the executable instruction enables the processor to execute an operation corresponding to the method for determining the optimal AI combination according to any one of claims 1 to 7.

10. A computer storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the method for determining the optimal AI combination according to any one of claims 1 to 7 is implemented.

11. A computer program product, the computer program product being tangibly stored on a computer-readable medium and comprising computer-executable instructions, which when executed cause at least one processor to perform the method for determining an optimal AI combination according to any one of claims 1 to 7.

Images