Device checking method, multi-device system and computer readable storage medium

Abstract

The invention provides a device checking method, a multi-device system and a computer readable storage medium, the system comprises a plurality of devices, and the device checking method comprises the following steps: acquiring basic use data of each device; acquiring energy management data of the system within a preset time; according to the energy management data of the system within the preset time, statistical use data of each device is obtained through a machine learning model; and comparing the basic use data with the statistical use data to judge whether the equipment operates normally or not. According to the embodiment of the invention, the relation between the energy management data and the equipment use condition is established, the operation condition of the equipment is reflected through the energy management data, the operation condition of the equipment can be preliminarily judged, the manual labor is reduced, and the equipment checking cost is reduced.

Description

Technical Field

[0001] The present invention generally relates to the field of Internet of Things technology, and in particular to a device inventory method, a multi-device system and a computer-readable storage medium. Background Art

[0002] In a system with multiple devices, in order to ensure the normal operation of the system, it is necessary to take inventory of multiple devices in the system after a period of time to determine whether the devices can operate normally. By comparing the consistency with the equipment (asset) list in the system, the equipment (asset) status of the system can be known. For a system with a small number of devices, each device can be set to upload its operating status separately, or a dedicated control system can be used to monitor the operating status of each device. However, for a complex system including a large number of devices or a large number of types of devices, it is extremely cumbersome to set up a channel for separate communication with each device, which will greatly increase the complexity of the system. Therefore, it is difficult to accurately judge each device. The method currently commonly used is to manually take inventory by hand, and the operator needs to go to the site to confirm the equipment. Or set an antenna for external communication on the device, and the operator also uses a reading device to read each device on the site to complete the inventory of the equipment. This process requires a lot of manpower and material resources, high cost, and low efficiency, so it is necessary to improve the inventory method of the equipment.

[0003] The contents of the background technology section are merely the technologies known to the inventors and do not necessarily represent the prior art in the field. Summary of the invention

[0004] In view of one or more deficiencies in the prior art, the present invention provides a device inventory method, including multiple devices in a system, and the device inventory method includes:

[0005] Obtaining basic usage data of each of the devices;

[0006] Obtaining energy management data of the system within a preset time;

[0007] Obtaining statistical usage data of each of the devices through a machine learning model based on the energy management data of the system within a preset time;

[0008] The basic usage data and the statistical usage data are compared to determine whether the device operates normally.

[0009] According to one aspect of the present invention, the basic usage data is obtained based on one or more of the historical data of the device, historical data of the same device in other systems, simulation system data and user-set usage data.

[0010] According to one aspect of the present invention, the basic usage data includes the number of times each of the devices is used within a unit time and the duration of each use; the unit time is less than or equal to the preset time.

[0011] According to one aspect of the present invention, the energy management data includes energy consumption of the system in a time series within a preset time.

[0012] According to one aspect of the present invention, the step of obtaining statistical usage data of each of the devices comprises:

[0013] Dividing the preset time into a plurality of sub-intervals;

[0014] Obtain energy management data within each sub-interval;

[0015] According to the energy management data in each sub-interval, the statistical usage data of each device in each sub-interval is obtained through a machine learning model.

[0016] According to one aspect of the present invention, the time lengths of the multiple subintervals are the same and continuous; the time length of the subinterval is not greater than the minimum value of the duration of each use of the device in the basic usage data.

[0017] According to one aspect of the present invention, the statistical usage data includes whether each of the devices is running in each sub-interval; and the step of determining whether the device is running normally includes:

[0018] comparing the number of times the device is operated in the statistical usage data with the number of times the device is used in the basic usage data;

[0019] Based on the comparison results of the operation times and the usage times, it is judged whether the equipment is operating normally.

[0020] According to one aspect of the present invention, the step of determining whether the device is operating normally comprises:

[0021] Calculate the risk rate based on the comparison between the number of runs and the number of uses;

[0022] A risk rate threshold is set, and when the risk rate exceeds the threshold, it is determined that the device is not operating normally.

[0023] According to one aspect of the present invention, the risk rate is represented by a difference and / or a ratio between the number of operations and the number of uses.

[0024] According to one aspect of the present invention, the device inventory method further includes:

[0025] Verify equipment inventory results;

[0026] Input the verification result, return to the step of obtaining the basic usage data of each device to optimize the basic usage data, and / or return to the step of obtaining the statistical usage data of each device to optimize.

[0027] According to one aspect of the present invention, the step of verifying the inventory result of the equipment comprises:

[0028] When it is determined that the equipment is not operating normally, verify the inventory results of the equipment.

[0029] According to one aspect of the present invention, the energy management data is obtained through an energy management module and / or a simulation system.

[0030] According to one aspect of the present invention, the present invention also includes a multi-device system, the multi-device system comprising:

[0031] Multiple devices;

[0032] An energy output terminal, to which the plurality of devices are connected;

[0033] an energy management module, the energy management module being connected to the energy output terminal and configured to acquire energy management data of the multi-device system;

[0034] A processor is in communication with the energy management module and is configured to execute the device inventory method as described above.

[0035] According to one aspect of the present invention, the multi-device system further includes:

[0036] A human-computer interaction interface communicates with the processor and is configured to acquire the basic usage data input by a user.

[0037] According to one aspect of the present invention, the present invention further includes a computer-readable storage medium, including computer executable commands stored thereon, and the executable commands implement the device inventory method as described above when executed by a processor.

[0038] Compared with the prior art, the embodiment of the present invention provides a device inventory method for inventorying multiple devices in a system, using energy management data to obtain statistical usage data of the device, and judging whether the device is operating normally based on the comparison of the statistical usage data and the basic usage data. In this embodiment, a connection between energy management data and device usage is established, and the operating status of the device is reflected by the energy management data, so that a preliminary judgment can be made on the operating status of the device, reducing manual labor and reducing the cost of equipment inventory. The present invention also includes an embodiment of a multi-device system and a computer-readable storage medium for executing the aforementioned device inventory method. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0040] Figure 1 is a schematic flow chart of a device inventory method in one embodiment of the present invention;

[0041] Figure 2A This is the LSTM model structure diagram;

[0042] Figure 2B It is the overall structure diagram of the deep learning model;

[0043] Figure 3 is a flow chart of a device inventory method including a process of dividing sub-intervals within a preset time in one embodiment of the present invention;

[0044] Figure 4 is a flow chart of a device inventory method including a process of comparing the number of runs and the number of uses in one embodiment of the present invention;

[0045] Figure 5 It is a schematic diagram of an arrangement table of the number of equipment shutdowns in one embodiment of the present invention;

[0046] Figure 6 is a schematic flow chart of a device inventory method including a verification process in one embodiment of the present invention;

[0047] Figure 7 It is a structural block diagram of a multi-device system in one embodiment of the present invention. DETAILED DESCRIPTION

[0048] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0049] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, which are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0050] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or can communicate with each other; it can be directly connected, or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0052] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or reference letters in different examples, and this repetition is for the purpose of simplicity and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides various specific examples of processes and materials, but those of ordinary skill in the art can be aware of the application of other processes and / or the use of other materials.

[0053] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0054] Figure 1 The specific process of a device inventory method 100 according to an embodiment of the present invention is shown. The system includes multiple devices. The device inventory method 100 is used to perform a preliminary evaluation of the operating status of the device to determine whether the device is operating normally. Figure 1 The device inventory method 100 is described.

[0055] like Figure 1 As shown, in step S101, basic usage data of each device is obtained, and the basic usage data of multiple devices in the system may be the same or different.

[0056] In actual applications, in most cases, multiple devices for realizing different functions are connected in the system. Specifically, taking a convenience store as an example, the entire system may include multiple different devices such as lighting equipment, food processing equipment, audio and video playback equipment, air conditioners, and monitoring equipment. Similarly, the usage of different devices is also different, so the basic usage data in this embodiment corresponds to the specific device.

[0057] In different embodiments of the present invention, the basic usage data may be a manually set value, or may be obtained through other methods, which will be specifically described in subsequent embodiments.

[0058] In step S102, energy management data of the system within a preset time is obtained, wherein the energy management data can be used to obtain or infer the actual usage of the device in this embodiment. Preferably, the preset time is a time period close to the current time, and the range of the preset time can be set according to the specific usage environment of the system and the device, such as several days or months.

[0059] In conventional applications, equipment inventory is a safeguard for the system, and it is not necessary to frequently perform inventory on the equipment in the system. For example, in most supermarkets, canteens and other systems, the frequency of equipment inventory is generally once or twice a year. The selected preset time is preferably a time unit that is less than the frequency of equipment inventory.

[0060] In a preferred embodiment of the present invention, the basic usage data of the device includes, for example, the number of times the device is used per unit time and the duration of each use of the device. To facilitate comparison in subsequent steps and simplify the calculation process, preferably, the preset time in this step is set to be greater than or equal to the unit time in the basic usage data to ensure that the device is theoretically started at least once within the preset time. For example, if the preset time is 3 days, the energy management data of the system within 3 days is obtained, and accordingly, the number of times the device is used per unit time can be the number of times the device is used in 1 day, or the number of times the device is used in 3 days.

[0061] The energy management data in this step may include, for example, energy consumption with a time series, or may be output power, output time, and other numerical values ​​or parameters that can quantify energy consumption. Specifically, the energy management data may be a time variation curve of the energy consumption of the system within a preset time, such as reading the meter data at a preset time frequency, and drawing a time variation curve of energy consumption based on the readings at each moment. Of course, the reading or sampling time may also be random or have other rules, as long as the energy consumption has its corresponding timestamp.

[0062] In actual situations, it is difficult to set up energy management or energy monitoring programs for all devices in the system, and it is even more unrealistic for systems with a large number of devices or a large number of types of devices. However, it is less difficult to obtain energy consumption values, which can be obtained directly from the energy management module or system. For example, the energy consumption with a time series can be obtained by monitoring the total electric meter in the system. Of course, in different embodiments of the present invention, the energy management data can also be the energy consumption of different devices or device groups in the system (also with a time series). For example, multiple devices in the system can be divided into several groups, and each group can be equipped with a separate electric meter, which not only helps to clarify management, but also improves the safety performance of the system.

[0063] In other embodiments of the present invention, energy management data can also be obtained through a simulation system. For example, when training the machine learning model in the subsequent steps, a virtual simulation model can be established, in which the energy management data used can be obtained by the simulation system without obtaining real energy management, energy consumption and other data. In the simulation system, physical equipment, energy output terminals and energy management devices may not be set, and the basic usage data of the equipment and the energy management data can be set or read through the simulation system.

[0064] In step S103, based on the energy management data of the system within a preset time, the statistical usage data of each device is obtained through a machine learning model. According to the above, the energy management data does not need to be specific to the energy usage of each device, and the total energy consumption of the system (with a time series) or the energy consumption of the group (with a time series) can be substituted into the machine learning model for calculation.

[0065] The machine learning model in this embodiment can be obtained by any machine learning algorithm, such as the aforementioned LSTM (long short-term memory neural network), or other neural network models. In most cases, the use process of the device can be divided into two states: operation (i.e., power on) and shutdown (i.e., shutdown or standby). Therefore, the state of the device can be represented by binary. Furthermore, the machine learning model can adopt deep learning methods, such as recursive neural network + fully connected layer, which can automatically learn from the data to obtain higher-level feature representations without manually constructing features.

[0066] For example, when constructing a machine learning model, energy management data and equipment operation data (such as historical operation data or manually set operation data) are input into the machine learning model, which may specifically include the frequency of use of the equipment, the duration of each use of the equipment, the power consumption of the equipment, and other data. Through model training, the relationship between energy management data and equipment usage is obtained after multiple iterations. The equipment usage includes, for example, whether the equipment is started, the use time of the equipment, the use frequency of the equipment, the relationship between the operation of the equipment and the environment, etc. The machine learning model is trained by a deep learning method to obtain the corresponding relationship between energy pipeline data and equipment operation data. In this embodiment, the operation data of the equipment can be inferred based on the energy management data, and then it can be determined whether the equipment is operating normally. In the actual operation of the system, each device operation (i.e., accessing the system and requesting energy supply) will bring about changes in energy usage. Therefore, the energy management data can reflect the usage of the equipment to a certain extent. At the same time, the acquisition of energy management data is less difficult. Therefore, this embodiment provides a machine learning model to establish a corresponding relationship between energy management data and equipment usage (statistical usage data of the equipment). After obtaining the machine learning model, the energy management data within a preset time is input, and the statistical usage data of the equipment can be output.

[0067] Specifically, the deep learning algorithm automatically extracts multiple features from the data to identify a certain pattern represented by the data. In this embodiment, the features may include: peak power consumption, minimum value, average value, rising slope, falling slope, duration, etc. The above features can be manually constructed and input energy management data. Preferably, they can also be calculated by automatically extracting more complex and abstract features by adjusting weights and biases after inputting energy consumption with a time series. The above features are only examples. When applying different learning algorithms, or even in different iterations, the types of features may be different.

[0068] like Figure 2A and Figure 2B As shown in the figure, first build an LSTM model between energy management data and equipment statistical usage data. Specifically, you can substitute the existing LSTM model and optimize the parameters in the iterative process. After obtaining the LSTM model, you can further build Figure 2B The deep learning model shown in the figure inputs energy management data into the input layer, passes through the LSTM model of the LSTM layer, and then enters the fully connected layer. The fully connected layer includes all the corresponding relationships between the input (the output result of the LSTM model) and the output (the operating status of the equipment). After the operation of the fully connected layer, the status of the equipment is finally output. For multiple devices in the system, the output of the output layer is the set of the operating status of all devices in the system.

[0069] The statistical usage data in this step is obtained based on the energy management data and does not represent the actual usage data of the device. By optimizing the machine learning model, the statistical usage data can be made close to the actual usage data, so that the statistical usage data can reflect the actual usage of the device within a preset time to a certain extent. This embodiment constructs a correspondence between energy management data and device usage, and obtains the usage of the device through energy management data analysis, that is, statistical usage data. The obtained device statistical usage data (which reflects the actual usage of the device to a certain extent) can be applied in subsequent steps.

[0070] In step S104, the basic usage data and the statistical usage data are compared to determine whether the device is operating normally. In this embodiment, the basic usage data represents the theoretical usage of the device set manually or obtained based on historical data, such as the minimum usage of the device or the average usage, while the statistical usage data represents the actual usage of the device. By comparing the two, it is possible to determine whether the device is operating normally, that is, to take inventory of the device.

[0071] Preferably, the data types of the basic usage data and the statistical usage data are set to be the same, for example, the basic usage data and the statistical usage data are both the usage frequency of the device, or the total usage duration within a preset time. Alternatively, the data types of the basic usage data and the statistical usage data are set to data types that can be unified through simple calculations for easy comparison, for example, the basic usage data includes the usage frequency of the device, and the statistical usage data is the total number of times the device is used within a preset time. When the difference between the statistical usage data of one device and its basic usage data is too large (for example, exceeding a set threshold), it can be determined that the operation of the device is abnormal.

[0072] According to the above, in existing applications, especially in complex systems with a large number of devices and a wide variety of devices, it is impossible to ensure that each device is connected to the processor or terminal of the system, that is, it is impossible to achieve that any device can upload the operating status data of the device and summarize it. For complex systems, it is usually configured that the processor or terminal only communicates with the main devices in the system to control the operation of the main devices and obtain the operating status of the main devices. For other secondary devices or small devices, separate signal paths are generally not established to simplify the system and reduce control and computing pressure. Therefore, all or part of the devices in the system cannot upload their operating status in real time, and the existing system cannot automatically obtain the operating status of all devices. When the equipment needs to be counted, it still needs to be done manually by the operator.

[0073] However, each device in the system needs to be connected to the energy supply end to ensure that the device can be used, for example, all devices in the system are connected to the power supply. At the same time, the device needs energy to drive when in use, and its energy usage in the running state and the shutdown state is completely different, that is, the operation of the device is related to the energy usage. Therefore, the embodiment of the present invention proposes a technical solution that uses energy management data to more accurately reflect the actual usage of the device. On this basis, the present invention proposes a device inventory method using a machine learning model, which can be used as an auxiliary device inventory process to reduce manual labor and improve the inventory efficiency of the equipment.

[0074] According to different embodiments of the present invention, the basic usage data of the device can be obtained in a variety of ways, such as based on one or more of the historical data of each device in the system, the historical data of the same device in other systems, simulation system data and user-set usage data.

[0075] According to the above, basic usage data may include: the frequency of use of each device, the duration of each use, the input power of each use, etc. For example, to obtain basic usage data based on the historical data of the device, you can first obtain data in the historical data of the device that is the same as the set basic usage data type, or data that can obtain the basic usage data type through simple calculations, and then use the extreme value, average value, weighted average value, etc. in the historical data as the basic usage data. The selection of specific values ​​and weights can be set according to the actual use requirements and use environment of the device.

[0076] Furthermore, the historical data of the device can also introduce parameters such as the environmental information of the device operation to reflect the difference in the operating status of the device in different environments. For example, for air conditioners, the parameters of ambient temperature and ambient humidity can be introduced in the process of obtaining basic usage data based on historical data to reflect the different characteristics of the frequency of use, duration of use and other data of the air conditioner under different ambient temperature and humidity conditions. Or in other embodiments of the present invention, for places such as convenience stores located in industrial and commercial areas, there are also differences in the historical data of the equipment on weekdays and weekends, that is, the historical data of the equipment is affected by time, so the date feature can be used as a parameter to reflect the changes in the usage of the equipment on different dates.

[0077] On this basis, a learning model can be established based on historical data, and the basic usage data of the equipment can be obtained using the historical data. The historical data, as well as the environment, time and other parameters that affect the historical data, are input into the learning model, and the basic usage data of the equipment is output through the learning model. For example, the basic usage data of the equipment in the current environment can be obtained based on the relationship between historical data and environmental changes.

[0078] Furthermore, historical data can be preprocessed, such as data cleaning, missing value processing, and standardization. The processed data is input into a deep learning model suitable for sequence data, such as LSTM. The learning model can automatically learn the time dependencies and features in the sequence and output the basic usage data of the equipment over a period of time.

[0079] According to different embodiments of the present invention, the basic usage data of the device can also be obtained through simulation system data. Specifically, a simulation system can be established to train a learning model for obtaining the basic usage data of the device, as well as to train a machine learning model in subsequent steps. It is also possible to improve the learning model of existing mature applications, such as adjusting parameters, to build a simulation system suitable for the scenario of this embodiment. The basic usage data of the device can also be set or obtained using the simulation system.

[0080] In the aforementioned process of using historical data to obtain basic equipment usage data, the historical data can also be historical data of equipment in other systems. For example, based on the similarity between other systems and the current system, historical data of other systems can be called, where the similarity can be used as a condition for optimizing the learning model, and the basic equipment usage data in the current system can be obtained through the matching relationship between different systems.

[0081] Alternatively, according to some other specific embodiments of the present invention, the basic usage data of the equipment in the system can also be directly set by the user. For example, for a restaurant, the dishwasher equipment can be manually set to start 1 to 3 times in a day. The basic usage data set by the user can also be obtained by the user based on the historical data of the equipment. Of course, the above methods can be comprehensively used to optimize the equipment in the system to obtain its corresponding basic usage data.

[0082] The frequency and duration of use in basic usage data can be obtained using the aforementioned methods or learning models. For some equipment, the duration of each use can also be directly obtained, such as automatic washing machines, disinfection cabinets and other equipment, the usage time after each start-up is a known value or a fixed value set by the equipment itself.

[0083] Figure 3 The specific process of the device inventory method 200 according to the preferred embodiment of the present invention is shown, which specifically includes the process of dividing the preset time into multiple sub-intervals. In the device inventory method 200, steps S201, S202 and S206 are basically the same as steps S101, S102 and S104 in the device inventory method 100 in the aforementioned embodiment, and are not repeated here.

[0084] This embodiment includes a process of dividing the preset time into multiple sub-intervals and performing calculation optimization, wherein specifically, in step S203, the preset time is divided into multiple sub-intervals. In this step, the time lengths of the sub-intervals are preferably equal, and the sub-intervals are continuous, so as to facilitate subsequent calculations and comparisons. For example, if the preset time is 3 days, the time lengths of the sub-intervals are all set to 15 minutes, that is, within the 3 days of the preset time, 288 continuous sub-intervals are divided.

[0085] Of course, in different embodiments of the present invention, the time lengths of the sub-intervals may be different. For example, in a time period when the device is used less frequently, the granularity of the sub-interval division may be appropriately enlarged. Similarly, the sub-intervals may also be set to be discontinuous. For example, the sub-intervals in a time period when the device is used less frequently may be discarded.

[0086] According to the above, in some embodiments of the present invention, the basic usage data of the device includes the duration of each use of the device. Preferably, the time length of the sub-interval is set to be no greater than the minimum value of the duration of each use of the device.

[0087] Specifically, the system includes four devices, namely, a dishwasher, an air conditioner, a coffee machine, and a lighting lamp, each of which has corresponding basic usage data, and the basic usage data of each device includes the duration of each use of the device. The duration of the dishwasher after each start-up is 20 minutes, the duration of the coffee machine after each start-up is 15 minutes, the duration of the air conditioner after each start-up is 3 hours, and the duration of the lighting lamp after each start-up is 5 hours. In the basic usage data of the above four different devices, the minimum value of the specific duration is 15 minutes. Therefore, the time length of the sub-interval is preferably set to no more than 15 minutes to avoid the sub-interval being too large, resulting in the device switching between operation and shutdown multiple times in a sub-interval, affecting the subsequent calculation accuracy. That is, in this embodiment, the smaller the time length of the sub-interval, the higher the accuracy of the subsequent calculation process, and of course, the calculation complexity also increases accordingly.

[0088] In step S204, energy management data in each sub-interval is obtained. Specifically, the energy management data can be directly retrieved from the energy management module or system, or obtained after a simple calculation, such as recording the electric meter readings at the start and end time of the sub-interval respectively, to obtain the total power consumption of the system in each sub-interval.

[0089] Furthermore, when the time length of each sub-interval is different, the statistical usage data in each sub-interval can be unified into the same time unit using the sub-interval with the smallest time length or the lowest common divisor of the time lengths of all sub-intervals as the unit to facilitate subsequent calculation processes.

[0090] In step S205, based on the energy management data in each sub-interval, the statistical usage data of each device in each sub-interval is obtained through a machine learning model.

[0091] Taking the aforementioned system including four devices, namely a dishwasher, an air conditioner, a coffee machine and a lighting lamp, as an example, the basic usage data of the devices can be set to include the frequency of use of the device and the duration of each use, wherein the duration of each use can refer to the values ​​in the aforementioned embodiments and will not be repeated here.

[0092] The frequency of use of the dishwasher is, for example, three times a day, the frequency of use of the coffee machine is, for example, once every three days, the frequency of use of the air conditioner is, for example, once a day, and the frequency of use of the lighting is, for example, once a day. The above usage frequencies are unified into unit time to ensure that each device is started at least once within the unit time. According to the above usage frequency data, the unit time is three days. Preferably, in step S202, the preset time for obtaining energy management data is set to not less than three days to ensure that each device is theoretically started at least once within the preset time.

[0093] This embodiment divides the preset time into multiple sub-intervals, which can further improve the calculation accuracy so that the obtained statistical usage data is close to the actual usage data of the device. After comparing with the basic usage data, the result of whether the device is operating normally is more accurate.

[0094] Further, such as Figure 4 As shown, in a preferred embodiment of the present invention, the device inventory method 300 includes a method for comparing the number of operations and the number of uses to obtain the device operation status. Figure 4 The device inventory method 300 in this embodiment is described.

[0095] In the device inventory method 300, steps S301, S302, S303 and S304 are substantially the same as steps S201, S202, S203 and S204 in the device inventory method 200 in the aforementioned embodiment, and are not described in detail.

[0096] In step S305, based on the energy management data in each sub-interval, the statistical usage data of each device in each sub-interval is obtained through a machine learning model.

[0097] In this embodiment, the statistical usage data of the device includes whether the device is running in each sub-interval, that is, whether it is powered on and used. Correspondingly, the basic usage data includes the number of times the device is used (or the frequency of use). Still taking the system in the above embodiment including four devices, a dishwasher, an air conditioner, a coffee machine and a lighting lamp as an example, in this system, the four devices all include two states, running and stopping, and therefore can be represented in binary form.

[0098] Specifically, Figure 5 The table shown in the figure shows that the columns in the table represent four different devices, and each row represents the running and stopping status of each device in a sub-interval, with 0 representing stopping and 1 representing running. The preset time is, for example, 3 days, and the time length of the sub-interval is set to 15 minutes. There are 288 sub-intervals in 3 days, and 288 sets of data can be formed. Figure 5 The table shown represents the operation or shutdown status of four different types of equipment in each sub-interval.

[0099] The operation or shutdown status is obtained through a machine learning model. For example, a machine learning model is constructed based on the LSTM algorithm and the fully connected layer, and the energy management data of each sub-interval is input. After passing through the LSTM algorithm and the fully connected layer, the state of the device is output, where the state of the device includes the operation of the device (recorded as 1) or shutdown (recorded as 0), so as to obtain Figure 5 The table shown in . It should be noted that Figure 5 The table shown in is only used to explain the present embodiment, and does not mean that the table needs to be formed or output in the system during the actual execution of the device inventory method 300.

[0100] After obtaining the operation and shutdown status of the equipment in different sub-intervals, in step S306, the operation times of the equipment in the statistical usage data and the usage times of the equipment in the basic usage data are compared. Taking the usage frequencies of the four different equipment described in the above embodiment as an example, the usage frequency of the dishwasher is three times a day, and within the preset time of 3 days, the usage times of the dishwasher is 9 times, and the usage frequency of the coffee machine is once every three days, and within the preset time of 3 days, the usage times of the coffee machine is 1 time. Similarly, within the preset time of 3 days, the usage times of the air conditioner is 3 times and the usage times of the lighting lamp is 3 times.

[0101] The statistical usage data of the device obtained when executing step S305 also includes the number of times the device is run within the preset time of 3 days. By comparing the difference between the number of runs and the number of uses, it can be determined whether the device is running normally within the preset time. For example, the usage frequency in the basic usage data of the device is the minimum usage frequency of the device according to historical data or user settings. When the number of runs in the statistical usage data of the device obtained in step S305 is less than the minimum number of uses, it can be determined that the device is abnormal.

[0102] In order to further improve the accuracy of the device inventory method 300 in judging the device operating status, in a preferred embodiment of the present invention, the device inventory method 300 further includes step S307, calculating the risk rate based on the comparison result of the number of operations and the number of uses. For example, the basic usage data of the device is an average value obtained based on the historical data of the device. There is a high possibility of misjudgment by simply comparing the size. By using the risk rate to represent the probability of abnormal operation of the device, the risk of misjudgment can be reduced, thereby improving the accuracy of the device inventory method in this embodiment.

[0103] Preferably, the risk rate can be expressed by the difference and / or ratio between the number of runs and the number of uses. For example, the calculation formula of the risk rate e can be set as:

[0104] e=(s1-m1) / s1,

[0105] Where s1 represents the number of times the device is used in the basic usage data, and m1 represents the number of operations in the statistical usage data obtained by the machine learning model. Specifically, for example, in the above embodiment, according to the basic usage data of the dishwasher, the number of times s1 is used within the preset time of 3 days is 9 times, and according to the energy management data, the number of times m1 of the dishwasher is run obtained by the machine learning model is, for example, 8 times. Substituting into the above formula, the risk rate e=1 / 9 is obtained.

[0106] In step S308, it is determined whether the risk rate exceeds a threshold, wherein the threshold can be set manually, for example, a reasonable risk rate threshold is set according to the actual application scenario of the system, and the risk rate threshold of each device in the system can be the same or different.

[0107] When the calculated risk rate exceeds the threshold, in step S309, it is determined that the device is not operating normally, otherwise it is determined that the device maintains normal operation.

[0108] For example, in the above embodiment, the threshold of the risk rate is set to 25%, and the risk rate obtained by the above calculation is e=1 / 9<25%, and it is judged that the dishwasher maintains normal operation. Or the threshold of the risk rate is set to 10%, and the risk rate obtained by the above calculation is e=1 / 9>10%, and it is judged that the dishwasher does not operate normally. The threshold can be set by the user, for example, according to the stability requirements of the system, a suitable threshold can be set.

[0109] In a preferred embodiment of the present invention, the equipment inventory method further includes a process of verifying the equipment inventory result. The judgment result of the equipment inventory method is verified through the verification process, specifically, for example, manually verifying and confirming the operation status of the equipment output by the equipment inventory method.

[0110] Furthermore, it also includes inputting verification results. In this embodiment, the verification results are obtained after on-site confirmation by the operator, for example, and can be considered as true results. If there is a deviation from the inventory results obtained according to the equipment inventory method in this embodiment, that is, there is a deviation between the output result of the equipment inventory method and the true result, the equipment inventory method can be corrected according to the verification results to further improve the accuracy of the equipment inventory method. The specific verification process can be carried out during the training of the machine learning model, or after the actual application of the equipment inventory method, a spot check verification can be carried out.

[0111] Specifically, Figure 6 The specific process of a device inventory method 400 including a verification process according to a preferred embodiment of the present invention is shown. In the device inventory method 400, steps S401, S402 and S403 are substantially the same as steps S101, S102 and S103 in the device inventory method 100 in the aforementioned embodiment, and are not described in detail.

[0112] In step S404, the basic usage data and the statistical usage data are compared to determine whether the device is operating normally. The specific determination process can refer to the above embodiment. When it is determined that the device is not operating normally, that is, the result output by the device inventory method 400 is that the device is not operating normally, in step S405, the inventory result of the device is verified.

[0113] Of course, in different embodiments of the present invention, even when it is determined that the equipment is operating normally according to the equipment inventory method, verification may be performed, such as random inspection.

[0114] On the other hand, after the device inventory method 400 determines that the device is not operating normally, it is also necessary to manually confirm the cause of the device failure. Therefore, when verifying the inventory result, troubleshooting can be performed at the same time.

[0115] In step S405, after verifying the inventory results, the verification results are input to optimize the equipment inventory method.

[0116] Specifically, the process may return to step S401 and modify and optimize the basic usage data of the device according to the verification result. When the basic usage data of the device is obtained through a learning model or algorithm, the verification result may be substituted into the learning model or algorithm for optimization.

[0117] It is also possible to return to step S403 and optimize the machine learning model, such as modifying the parameters of the machine learning model so that the obtained equipment statistical usage data is close to the actual usage of the equipment, thereby optimizing the equipment inventory method 400. Regardless of whether the result output by the equipment inventory method 400 is consistent with the actual situation determined after verification, the verification result can be input into the machine learning model to optimize the statistical usage data of the obtained equipment. For example, if the verification result is the same as the result output by the equipment inventory method 400, the confidence of the corresponding relationship between energy management data and the statistical usage data of the equipment in the current machine learning model can be increased. Otherwise, the confidence of the corresponding relationship in the machine learning model is reduced.

[0118] Further, according to Figure 4 In the embodiment shown in , the risk rate threshold can also be adjusted according to the input verification result, so that the final result of determining whether the device is operating normally is more accurate. Figure 7 As shown, the present invention also includes an embodiment of a multi-device system 1, which includes multiple devices 10, an energy output terminal 20, an energy management module 30 and a processor 40. Figure 7 The multi-device system 1 will be described.

[0119] In the multi-device system 1, the multiple devices 10 may be independent of each other or may have a linkage operation relationship, such as air conditioners and electric doors and windows, etc., without limitation. The multiple devices 10 are all connected to the energy output terminal 20, and the energy output terminal 20 provides energy for the operation of the devices 10. Specifically, the energy output terminal 20 is, for example, a power cord.

[0120] The energy management module 30 is connected to the energy output terminal 20 and can obtain energy management data of the multi-device system 1. Specifically, the energy management module 30 includes, for example, an electric meter, which can obtain the power consumption output by the power line.

[0121] The processor 40 communicates with the energy management module 30 and is configured to execute the device inventory method described in the above embodiments.

[0122] In this embodiment, some of the multiple devices 10 may be connected to the energy management module 30 or the processor 40 , for example, to upload power consumption or operating status, or to receive control instructions output by the processor 40 .

[0123] However, it is also possible that some or all of the devices 10 are not connected to the energy management module 30 or the processor 40 to simplify the complexity of the multi-device system 1. The control of each device 10 can be manually operated, or the operation rules can be pre-set for each device, such as setting a dishwasher to automatically run at a specific time.

[0124] This embodiment utilizes the device inventory method in the aforementioned embodiment to inventory multiple devices 10 in the multi-device system 1, obtain the operating status of the devices 10, reduce manual labor intensity, and improve equipment inventory efficiency.

[0125] According to a preferred embodiment of the present invention, the multi-device system 1 further includes a human-machine interface 50, which communicates with the processor 40 and is capable of acquiring basic usage data input by a user for application in the device inventory method. Furthermore, the human-machine interface 50 can also be configured to display the operating status of the device 10 output by the device inventory method for user reference.

[0126] The present invention also includes a computer-readable storage medium, including computer executable commands stored thereon, and when the executable commands are executed by a processor, the device inventory method described in the aforementioned embodiment is implemented. The computer-readable storage medium may specifically be, for example, a hard disk, a magnetic disk, a magnetic tape, an optical disk, etc.

[0127] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A device inventory method, comprising a plurality of devices in a system, the device inventory method comprising: Obtaining basic usage data of each of the devices; Obtaining energy management data of the system within a preset time; Obtaining statistical usage data of each of the devices through a machine learning model based on the energy management data of the system within a preset time; The basic usage data and the statistical usage data are compared to determine whether the device operates normally.

2. The equipment inventory method according to claim 1, wherein the basic usage data is obtained based on one or more of historical data of the equipment, historical data of the same equipment in other systems, simulation system data and user-set usage data.

3. The device inventory method according to claim 1, wherein the basic usage data includes the number of times each of the devices is used within a unit time and the duration of each use; the unit time is less than or equal to the preset time.

4. The equipment inventory method according to claim 3, wherein the energy management data includes energy consumption of the system in a time series within a preset time.

5. The device inventory method according to claim 3, wherein the step of obtaining statistical usage data of each of the devices comprises: Dividing the preset time into a plurality of sub-intervals; Obtain energy management data within each sub-interval; According to the energy management data in each sub-interval, the statistical usage data of each device in each sub-interval is obtained through a machine learning model.

6. The device inventory method according to claim 5, wherein the time lengths of the multiple sub-intervals are the same and continuous; and the time length of the sub-interval is not greater than the minimum value of the duration of each use of the device in the basic usage data.

7. The device inventory method according to claim 6, wherein the statistical usage data includes whether each device is running in each sub-interval; and the step of determining whether the device is running normally includes: comparing the number of times the device is operated in the statistical usage data with the number of times the device is used in the basic usage data; Based on the comparison results of the operation times and the usage times, it is judged whether the equipment is operating normally.

8. The device inventory method according to claim 7, wherein the step of determining whether the device is operating normally comprises: Calculate the risk rate based on the comparison between the number of runs and the number of uses; A risk rate threshold is set, and when the risk rate exceeds the threshold, it is determined that the device is not operating normally.

9. The equipment inventory method according to claim 8, wherein the risk rate is represented by a difference and / or a ratio between the number of operations and the number of uses.

10. The device inventory method according to any one of claims 1 to 9, further comprising: Verify equipment inventory results; Input the verification result, return to the step of obtaining the basic usage data of each device to optimize the basic usage data, and / or return to the step of obtaining the statistical usage data of each device to optimize.

11. The device inventory method according to claim 10, wherein the step of verifying the device inventory result comprises: When it is determined that the equipment is not operating normally, verify the inventory results of the equipment.

12. The equipment inventory method according to any one of claims 1 to 9, wherein the energy management data is obtained through an energy management module and / or a simulation system.

13. A multi-device system comprising: Multiple devices; An energy output terminal, to which the plurality of devices are connected; an energy management module, the energy management module being connected to the energy output terminal and configured to acquire energy management data of the multi-device system; A processor, the processor communicates with the energy management module and is configured to execute the device inventory method according to any one of claims 1-12.

14. The multi-device system according to claim 13, further comprising: A human-computer interaction interface communicates with the processor and is configured to acquire the basic usage data input by a user.

15. A computer-readable storage medium, comprising computer-executable commands stored thereon, wherein the executable commands, when executed by a processor, implement the device inventory method according to any one of claims 1 to 12.

Images