A home appliance system with generative script and control method thereof

Abstract

The present invention provides a home appliance system with generative script, which includes a plurality of home appliances, a control device and a cloud server. the home appliances is used to detect their operating status and generate a status data, the cloud server inputs the status data into the neural network model to calculate and generate at least one home appliance script, and the control device operates a script application to select and execute the corresponding home appliance script and generate a corresponding control request. Afterwards, the cloud server generates at least one control signal based on the control request and transmits the control signal to the corresponding home appliance, thereby outputting a control command based on the control signal and controlling the home appliances to execute the control command.

Description

Technical Field

[0001] This invention relates to a home appliance system and its control method, and more particularly to a home appliance system and its control method with generative scripts. Prior Technology

[0002] With the rapid development of technology, many home appliances now have remote control capabilities. However, when you want to control these appliances to perform the desired actions, you may need to input various commands to make them work as required.

[0003] The aforementioned control methods are not only inconvenient for users to operate, but also lead to energy waste if the appliance is forgotten to be turned off when not in use. Furthermore, if the appliance is used again, the same commands need to be entered again, which will affect the user's experience of operating it.

[0004] Therefore, current technology offers a method where users can set scripts to make multiple home appliances operate accordingly. While this can reduce the inconvenience of users individually controlling multiple home appliances and forgetting to turn off unused appliances, the script settings are fixed processes and cannot be humanized to conform to users' usage habits.

[0005] In view of this, there is an urgent technical need for a home appliance system with generative scripts and its control method to improve the problems existing in the prior art. Summary of the Invention

[0006] The purpose of this invention is to provide a home appliance system with generative scripts, so as to realize scripted control of home appliances and humanize the user's usage habits.

[0007] To achieve the aforementioned objectives, the present invention provides a home appliance system with generative scripting, comprising: a plurality of home appliances, each home appliance including an appliance processing module and an appliance transmission module, the appliance processing module being connected to the appliance transmission module, the appliance processing module recording the operating status of the home appliance and generating status data; a control device having a scripting application installed thereon, the control device including a control processing module and a control transmission module, the control transmission module being connected to the control processing module and receiving at least one appliance script, the control processing module running the scripting application to select and execute the corresponding appliance script, and generating data based on the selected appliance script. The system comprises: a control request; and a cloud server connected to the home appliance and the control device. The cloud server receives the status data and inputs it into a neural network model to calculate and generate the home appliance script. The cloud server transmits the home appliance script to the control device. The cloud server receives the control request and generates at least one control signal based on the control request. It then transmits the control signal to the home appliance transmission module corresponding to the home appliance based on a control flag of the control signal. The home appliance transmission module outputs a control command to the home appliance processing module based on the control signal. The home appliance processing module controls the home appliance to execute the control command.

[0008] Preferably, the neural network model includes an input layer, a plurality of hidden layers, and an output layer. The input layer receives the state data and identifies a plurality of categories of data in the state data. The input layer outputs each category of data to a plurality of nodes in the hidden layers. The hidden layers assign a weight to each node according to a set rule. The output layer receives a prediction data for each category from the configuration output of the hidden layers. The cloud server generates the home appliance script based on the prediction data.

[0009] Preferably, this category of data includes an appliance label, an operating time, an operating data point, or a power consumption.

[0010] Preferably, the setting rules include a comfort rule, a maximization rule, and an energy-saving rule. The weights include a weighted value, a base value, or a subtracted weight. When the hidden layers assign weights to each node according to the comfort rule, the hidden layer assigns the weighted value to the corresponding node based on user data and assigns the base value to other nodes. When the hidden layers assign weights to each node according to the maximization rule, the hidden layer assigns the weighted value to the corresponding node based on maximum usage data and assigns the base value to other nodes. When the hidden layers assign weights according to the energy-saving rule... When assigning weights to each node, the hidden layer makes a judgment based on user data and power consumption. When a node matches the user data and its corresponding power consumption is lower than a power consumption threshold, the node is given a weighted value. When a node matches the user data and its corresponding power consumption is higher than a power consumption threshold, the node is given a base value. When a node does not match the user data and its corresponding power consumption is lower than a power consumption threshold, the node is given a base value. When a node does not match the user data and its corresponding power consumption is higher than a power consumption threshold, the node is given a decremented value.

[0011] Preferably, the user data consists of a plurality of running times and a plurality of running data for each of the home appliances during a training period. An overlap interval of each running time and each running data is determined based on a habit threshold, and the overlap interval above the habit threshold is taken as the user data.

[0012] Preferably, the maximum usage data is a plurality of the running times and a plurality of the running data during a training period, and the maximum value of the running time and the running data is selected as the maximum usage data based on an anomaly exclusion rule.

[0013] Preferably, the home appliance system with generative scripts includes: an identification device connected to the home appliance, the identification device identifying identification data based on user information, and when the identification device confirms that the user information matches the identification data, the identification device activates the home appliance to drive the home appliance to execute the control command.

[0014] Preferably, the appliance transmission module transmits one of the control commands' command codes to the cloud server in MQTT format. Then, when the script application is installed on the control device, the cloud server transmits the command code of the control command to the control device. The appliance script contains the command codes of each control command.

[0015] Preferably, the control device includes a storage module for storing the appliance script. The control device has an appliance control application installed and executes the appliance control application to control the appliance. The storage module stores the control content set in the control settings as a custom script.

[0016] To achieve the aforementioned objectives, the present invention also provides a generative scripting home appliance control method applicable to a home appliance system with generative scripting as described above, comprising: recording the operating states of a plurality of home appliances and generating state data; inputting the state data into a neural network model using a cloud server to calculate and generate at least one home appliance script; receiving the home appliance script via a control device and running the script application to select and execute the corresponding home appliance script; generating a control request based on the selected home appliance script; receiving the control request via the cloud server and generating at least one control signal based on the control request; transmitting the control signal to the corresponding home appliance based on a control flag of the control signal; and outputting a control command based on the control signal to control the home appliance to execute the control command.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the following detailed description is provided in conjunction with the specific embodiments listed in the figures. Simple Explanation of the Diagram

[0018] Figure 1 is a schematic diagram of the configuration relationship of the home appliance system with generative script according to the present invention; Figure 2 is a schematic diagram of the neural network model of the present invention; Figure 3 is a schematic diagram of the overlapping area of ​​the present invention; Figure 4 is a schematic diagram showing the configuration of the identification device of the present invention; Figure 5 is a schematic diagram of the configuration of the storage module of the present invention; Figure 6 is a flowchart of the steps of the home appliance control method with generative script according to the present invention. Implementation

[0019] The advantages, features, and technical methods of this invention will be more readily understood by referring to the exemplary embodiments and accompanying drawings. This invention may be implemented in different forms and should not be construed as being limited to the embodiments set forth herein. Rather, the embodiments provided will make this disclosure more thorough, complete, and fully convey the scope of the invention to those skilled in the art, and the invention will be defined only as provided in the appended claims.

[0020] In addition, the terms “contain” and / or “contain” refer to the existence of said feature, region, whole, step, operation, component and / or component, but do not exclude the presence or addition of one or more other features, region, whole, step, operation, component, component and / or combination thereof.

[0021] In order to facilitate your review of the contents of the present invention, and the efficacy that can be achieved, the specific embodiments enumerated in the diagram are hereby elaborated as follows:

[0022] 1 to 3, which are a schematic diagram of the configuration relationship for a home appliance system with generative scripts, a schematic diagram of a neural network model, and a schematic diagram of an overlap interval of the present invention. As shown, the home appliance system with generative scripts of the present invention comprises a plurality of home appliance devices 110 , control devices 120 , and cloud servers 130 .

[0023] In continuation, the home appliance device 110 described above comprises a home appliance processing module 111 and a home appliance transmission module 112 , which is linked to a home appliance transmission module 112 , wherein the home appliance processing module 111 and the home appliance transmission module 112 may be entered by resorting to PCIE, PCI, UART, etc line link and transmit and receive signal information, and home appliance transmission module 112 has a remote communication unit, such as WIFI, ZIGBEE, RF, Bluetooth, etc. However, the above is an example only and is not limited to this; into status data, or used to control the home appliance device 110 to execute a control command, the home appliance transmission module 112 may be used to transmit the command code for the control command sent in MQTT format to the cloud server 130, which will then be sent to the control device 120 when the script application is installed The command code for the control command The control device 120 is used so that the command code for each control command is included in the home appliance script 131 in order to generate a control signal 132 according to the control requirement 1211 and then to know the control command to be executed according to the control signal 132. This is also an example and is not used to qualify the invention.

[0024] Its control device 120 is installed with a script application, and the control device 120 comprises a control processing module 121 and a control transmission module 122. The control transmission module 122 is linked to and used to receive at least one appliance script 131. When the control processing module 121 operates the script application to select and execute the corresponding appliance script 131 , the generation of the control processing module 121 may be selected by the control script 121 according to the requirements. Whereas appliance script 131 may be generated according to different needs, such as focusing on comfort or energy saving, etc., appliance script 131 may also be a plurality of appliance scripts 131 to provide the user-selected appliance script 131 and generate control requirements 1211 accordingly.

[0025] Its cloud server 130 links the home appliance device 110 and the control device 120. The cloud server 130 is used to receive the status data and input the status data into the neural network model 133 so that it computes and generates the home appliance script 131. The neural network model is capable of including input layer 1331 , a plurality of hidden layers 1332 , and output layer 1333 , while the status data may contain data such as the plural data category of running data, the amount of operating time, and the amount of running time of home appliances. The input layer 1331 is used to receive the status data and identify the various categories of data in the status data, and each category of data is output to a complex number of nodes 13321 in those hidden layers 1332 , which may then be given weights to each node 13321 according to the setting rules.

[0026] For example, the rules can include comfort rules, maximization rules, and energy-saving rules, while the weights can include weighted values, base values, or subtracted weights. When the hidden layers 1332 assign weights to each node 13321 according to the comfort rules, the hidden layer 1332 assigns a weighted value to the corresponding node 13321 based on user data, and assigns a base value to other nodes 13321. It can be understood that the weighted value will generally be higher than the base value, so that the weighted value of the node 13321 that obtains the weighted value is generally higher than the weighted value of the node 13321 that obtains the base value. When the hidden layers 1332 assign weights to each node 13321 according to the maximization rules, the hidden layer 1332 assigns a weighted value to the corresponding node 13321 based on the maximum usage data, and assigns a base value to other nodes 13321. When the hidden layers 1332 assign weights to each node 13321 according to the energy-saving rules, the hidden layer 1332 assigns weights based on user data. The system judges the user data and the power consumption of each node. When node 13321 matches the user data and the corresponding power consumption is lower than the power consumption threshold, a weighted value is given to node 13321. When node 13321 matches the user data and the corresponding power consumption is higher than the power consumption threshold, a base value is given to node 13321. When node 13321 does not match the user data and the corresponding power consumption is lower than the power consumption threshold, a base value is given to node 13321. When node 13321 does not match the user data and the corresponding power consumption is higher than the power consumption threshold, a decremented value is given to node 13321. The power consumption threshold can be set by the user or obtained from a database (for example, if the power consumption of home appliance 110 is 135 kWh under the comfort rule, the cloud server 130 obtains multiple power consumption data of 120 to 145 kWh under the comfort rule, and sets the power consumption threshold based on the corresponding energy-saving data after converting the power consumption data to the corresponding energy-saving data after applying the energy-saving rule (the average value can be taken)).

[0027] It should be noted that the weighted value, base value, and subtraction value mentioned above are the base weight values ​​given to the node. For example, the weighted value is N+0.1, the base value is N, and the subtraction value is N-0.1. Subsequently, the node's weight value can also be changed by calculation or data feeding.

[0028] The user data consists of multiple operating times (daily operating times, e.g., air conditioning from 5 PM to 3 AM) and multiple operating data points (e.g., automatic mode, set temperature of 25 degrees Celsius, fan mode with automatic up-and-down oscillation) for each household appliance 110 during the training period (e.g., one month). The overlap interval OA of each operating time and each operating data point is determined based on a habit threshold. Overlap intervals OA exceeding the habit threshold are considered user data. That is, assuming a habit threshold of 90% overlap, and the overlap interval OA is determined by Monday, Tuesday, and Wednesday, then within one month (assuming at least four complete cycles), if the overlap interval OA is higher than the first Monday to the fourth Monday... If the air conditioning is on from 5 PM to 3 AM on all four Mondays, then the overlapping interval OA can be determined as 100% overlap for the air conditioning from 5 PM to 3 AM on the first to fourth Mondays. However, if the air conditioning is on from 5 PM to 3 AM on the first to third Mondays (spanning into Tuesday) and off on the fourth Monday, and the air conditioning is on from 6 PM to 2 AM on the fourth Monday, then the overlapping interval OA is determined as 100% overlap for the air conditioning from 6 PM to 2 AM on the first to fourth Mondays, but 75% overlap for the air conditioning from 5 PM to 6 PM and from 2 AM to 3 AM on the first to fourth Mondays. (This embodiment is shown in Figure 3) If the air conditioning is turned on from 5 pm to 3 am on the first to the third Monday, and the air conditioning is turned on from 2 pm to 6 pm on the fourth Thursday, then the overlap interval OA is determined to be 100% for the air conditioning from 5 pm to 6 pm on the first to the fourth Monday, but 25% for the air conditioning from 2 pm to 5 pm on the first to the fourth Monday, and so on. In this way, the overlap interval OA that is higher than the customary threshold is used as user data.

[0029] The maximum usage data consists of multiple runtimes and multiple data points during the training period (e.g., one month). The runtime and data points with the highest values ​​are selected as the maximum usage data based on anomaly exclusion rules (i.e., excluding extreme values). In other words, if the air conditioning is on from 5 PM to 3 AM on the first Monday, from 8 PM to 3 AM on the second Monday, from 5 PM to 2 AM on the third Monday, and from 6 PM to 3 AM on the fourth Monday, the runtime from 5 PM to 3 AM on the first Monday is chosen as the maximum usage data because the first Monday has the longest runtime of ten hours. The anomaly exclusion rules are used to exclude extreme cases. For example, if the air conditioning is on from 5 PM to 3 AM on the first to third Mondays, but on the fourth Monday it is on from 8 AM to 3 AM, significantly exceeding the average, the anomaly exclusion rules will exclude the data from the fourth Monday.

[0030] Furthermore, the above explanations are based on operating time, but the same principle applies to operating data. That is, if the air conditioner's set temperature is 25 degrees Celsius on the first Monday, 26 degrees Celsius on the second Monday, and 25 degrees Celsius on the third and fourth Mondays, the user data or maximum usage data can be selected based on the overlapping range OA or the maximum value. It should be noted that since the purpose of air conditioning is cooling, the maximum set temperature can be selected as 25 degrees Celsius, not simply based on the numerical value, but rather on the comfort margin provided by the appliance itself.

[0031] The output layer 1333 receives various types of prediction data from the configuration output of the hidden layers 1332. The cloud server generates corresponding home appliance scripts 131 based on this prediction data. However, the neural network model 133 of this invention can continuously "train" and "infer" based on the daily input data. That is, when the input value is transmitted from one layer node to the next layer node, the neural network model 133 will assign a positive or negative weight to the parameter based on the correctness of the output value. Through repeated cycles of "forward propagation" (or prediction) and "backward propagation" (or feedback), the weight of the parameter can be made more accurate to correctly output the prediction data.

[0032] The cloud server 130 receives the control request 1211 generated by the control device 120, generates a control signal 132 according to the control request 1211, and transmits the control signal 132 to the home appliance transmission module 112 of the home appliance device 110. The home appliance transmission module 112 outputs control commands to the home appliance processing module 111 according to the control signal 132. The home appliance processing module 111 controls the home appliance device 110 to execute the control commands.

[0033] Among them, the home appliance script 131 may include script conditions and action instructions after the script conditions are met.

[0034] The script application selects the execution option of appliance A script 131 through the control device 120. When the user selects the execution option, the script application will cause the control processing module 121 to generate control request 1211 in accordance with appliance A script 131 after triggering the script conditions set by appliance A script 131. Control request 1211 is to require appliance 110 to execute the action instructions set before appliance A script 131. For example, if the script condition for appliance A script 131 is an air conditioning trigger temperature of 30 degrees Celsius and an action command of a set temperature of 25 degrees Celsius, then when the user selects the execution option for appliance A script 131, the control processing module 121 will generate a control request 1211 corresponding to the action command of the set temperature of 25 degrees Celsius. This request is then transmitted to the cloud server 130 to generate a control signal 132. Based on the control flag of the control signal 132, the signal is transmitted to the appliance transmission module 112 of the corresponding appliance device 110. This causes the appliance transmission module 112 to output the control command to the appliance processing module 111 of the appliance device 110, which then controls the appliance device 110 to execute the control command (setting the temperature to 25 degrees Celsius). However, the above is merely an example and is not intended to be limiting.

[0035] Furthermore, the control processing module 121 can generate power consumption assessment values ​​corresponding to the action instructions of the home appliance script 131, and the user can refer to the power consumption assessment values ​​of each home appliance script 131 as the basis for selecting the execution option of the corresponding home appliance script 131. Moreover, the control processing module 121 can store various power consumption data of the home appliance device 110 as the basis for calculating the power consumption assessment value.

[0036] Please refer to Figure 4, which is a schematic diagram of the configuration of the identification device of the present invention. As shown in the figure, the home appliance system with generative scripts includes an identification device 200, which is connected to the home appliance devices 110. The identification device 200 identifies identification data 211 based on user information 210. When the identification device 200 confirms that the user information 210 matches the identification data 211, the identification device 200 activates the home appliance devices 110 to drive them to execute control commands. For example, when the identification device 200 is a door lock, the door lock can determine whether the image or fingerprint (i.e., identification data 211) obtained from outside the door matches the user information 210 stored in the door lock's storage module. If it matches, the door lock can not only start the unlocking procedure but also activate the home appliance devices 110, thereby avoiding energy waste, i.e., the home appliance script 131 continues to run even when no one is home, thus avoiding energy waste.

[0037] Please refer to Figure 5, which is a schematic diagram of the configuration of the storage module of the present invention. As shown in the figure, the control device 120 may include a storage module 123 to store the home appliance script 131. In addition, the control device 120 may also install a home appliance control application. When the user uses the home appliance control application to control the home appliance device 110, the storage module 123 can also store the control content set by the user as a custom script 1231. This way, when the same settings are used again, the user can select the execution option of the custom script 1231 corresponding to the same settings from the script application and directly apply the control content set by the user using the home appliance control application. However, this is only an example and is not limited thereto.

[0038] Please refer to Figure 6, which is a flowchart of the steps of the generative script-based home appliance control method of the present invention. As shown in the figure, the generative script-based home appliance control method of the present invention is applicable to the aforementioned generative script-based home appliance system, and includes the following steps:

[0039] S101: Record the operating status of multiple home appliances and generate status data;

[0040] S102: Utilize a cloud server to input state data into a neural network model to calculate and generate at least one appliance script;

[0041] S103: Receives appliance scripts via the control device and runs the script application to select and execute the corresponding appliance script;

[0042] S104: Generate control requirements based on the selected home appliance script;

[0043] S105: Receives control requests from a cloud server and generates at least one control signal based on the control requests;

[0044] S106: Transmit a control signal to the corresponding home appliance based on the control flag of the control signal; and

[0045] S107: Outputs control commands based on control signals to control home appliances to execute control commands.

[0046] The embodiments disclosed in this invention are preferred embodiments. Any local changes or modifications that originate from the technical concept of this invention and can be easily deduced by those skilled in the art are not outside the scope of the patent rights of this invention.

[0047] In conclusion, this invention, in terms of its purpose, means, and effects, demonstrates technical features that are distinct from those of the conventional invention. Furthermore, its invention is practical and meets all the requirements for a patent. We earnestly request that your esteemed examiner carefully review this matter and grant a patent as soon as possible so that it may benefit society. We would be truly grateful for your assistance.

[0048] 110: Home Appliances 111: Home Appliance Processing Module 112: Home appliance transmission module 120: Control device 121: Control Processing Module 1211: Control Requirements 122: Control Transmission Module 123: Storage Module 1231: Custom Script 130: Cloud Server 131: Home Appliance Script 132: Control signal 133: Neural Network Model 1331: Input Layer 1332: Hidden Layer 13321: Node 1333: Output Layer 200: Identification device 210: User Information 211: Identification Data OA: Overlapping interval S101-S107: Steps

Claims

1. A home appliance system with generative scripting, comprising: a plurality of home appliance devices, each home appliance device including an appliance processing module and an appliance transmission module, the appliance processing module being connected to the appliance transmission module, the appliance processing module recording the operating status of the home appliance device and generating status data; a control device having a scripting application installed thereon, the control device including a control processing module and a control transmission module, the control transmission module being connected to the control processing module and receiving at least one appliance script, the control processing module running the scripting application to select and execute the corresponding appliance script, and generating a control request based on the selected appliance script; and a cloud server connected to the home appliance devices and the control device, the cloud server receiving the status data and inputting the status data into a neural network model to calculate and generate the appliance script, the cloud server transmitting the appliance script to the control device; wherein... The cloud server receives the control request and generates at least one control signal based on the control request. It then transmits the control signal to the corresponding appliance transmission module based on a control flag in the control signal. The appliance transmission module outputs a control command to the appliance processing module based on the control signal. The appliance processing module controls the appliance to execute the control command. The neural network model includes an input layer, multiple hidden layers, and an output layer. The input layer receives the state data and identifies multiple categories of data within the state data. The input layer outputs each category of data to multiple nodes in the hidden layers. The hidden layers assign a weight to each node according to a set rule. The output layer receives predicted data for each category of data from the configuration output of the hidden layers. The cloud server generates the appliance script based on the predicted data. The category data includes an appliance flag, a running time, running data, or power consumption. The setting rules include a comfort rule, a maximization rule, and an energy-saving rule. The weights include a weighted value, a base value, or a subtractive weight. When the hidden layers assign weights to each node according to the comfort rule, the hidden layer assigns the weighted value to the corresponding node based on user data and assigns the base value to other nodes. When the hidden layers assign weights to each node according to the maximization rule, the hidden layer assigns the weighted value to the corresponding node based on maximum usage data and assigns the base value to other nodes. When the hidden layers assign weights according to the energy-saving rule... When assigning weights to each node, the hidden layer makes a judgment based on user data and power consumption. When a node matches the user data and its corresponding power consumption is lower than a power consumption threshold, the node is given a weighted value. When a node matches the user data and its corresponding power consumption is higher than a power consumption threshold, the node is given a base value. When a node does not match the user data and its corresponding power consumption is lower than a power consumption threshold, the node is given a base value. When a node does not match the user data and its corresponding power consumption is higher than a power consumption threshold, the node is given a decremented value.

2. A home appliance system with generative scripting as described in claim 1, wherein, The user data consists of multiple running times and multiple running data for each home appliance during a training period. An overlap interval of each running time and each running data is determined based on a habit threshold, and the overlap interval above the habit threshold is taken as the user data.

3. A home appliance system with generative scripting as described in claim 1, wherein, The maximum usage data is a plurality of runtimes and a plurality of data points during a training period, and the runtime and data points with the largest value are selected as the maximum usage data based on an anomaly exclusion rule.

4. The home appliance system with generative scripts as described in claim 1, comprising: an identification device connected to the home appliance, the identification device identifying identification data based on user information, and when the identification device confirms that the user information matches the identification data, the identification device activating the home appliance to drive the home appliance to execute the control command.

5. A home appliance system with generative scripting as described in claim 1, wherein, The home appliance transmission module transmits one of the control commands' command codes to the cloud server in MQTT format. Then, when the script application is installed on the control device, the cloud server transmits the command codes of the control commands to the control device. The home appliance script contains the command codes of each control command.

6. A home appliance system with generative scripting as described in claim 1, wherein, The control device includes a storage module for storing the appliance script. The control device has an appliance control application installed and executes the appliance control application to set control settings for the appliance. The storage module stores the control content set in the control settings as a custom script.

7. A generative scripting home appliance control method for a home appliance system with generative scripting as described in any one of claims 1 to 6, comprising: recording the operating states of a plurality of home appliance devices and generating state data; inputting the state data into a neural network model using a cloud server to calculate and generate at least one home appliance script; receiving the home appliance script via a control device and running the scripting application to select and execute the corresponding home appliance script; generating a control request based on the selected home appliance script; receiving the control request via the cloud server and generating at least one control signal based on the control request; transmitting the control signal to the corresponding home appliance device according to a control flag of the control signal; and outputting a control command based on the control signal to control the home appliance device to execute the control command.

Images