An intelligent gateway application method and system

Abstract

The application relates to the field of intelligent decision-making, and discloses an intelligent gateway application method, which comprises the following steps: acquiring a gateway application scene and networking equipment of an intelligent gateway to be applied, collecting historical gateway data of the gateway application scene by using the networking equipment, and identifying a gateway factor category of the historical gateway data; according to the gateway factor category, a factor fuzzy rule of the gateway factor is constructed, a gateway monitoring model is constructed according to the factor fuzzy rule and the gateway factor category, the gateway monitoring model is trained according to the historical gateway data, and a trained gateway monitoring model is obtained; real-time gateway data is acquired, the trained gateway monitoring model is used to calculate a gateway evaluation grade corresponding to the real-time gateway data; based on the gateway evaluation grade, a to-be-controlled device in the networking equipment is determined, a pre-constructed device control strategy is sent to the to-be-controlled device, the device control strategy is executed, and an application result of the intelligent gateway to be applied is obtained. The application can improve the control precision of intelligent gateway application.

Description

Technical Field

[0001] This invention relates to the field of intelligent decision-making, and more particularly to an intelligent gateway application method and system. Background Technology

[0002] The smart gateway is the heart of the smart control system. It can realize functions such as system information collection, information input, information output, centralized control, remote control, and linkage control. It is widely used in smart home, industrial control, and real-time monitoring of construction sites.

[0003] Currently, traditional applications of smart gateways mainly focus on monitoring and controlling single control factors in a scenario, without considering the inherent relationships between these control factors. This results in low control accuracy in smart gateway applications. Summary of the Invention

[0004] This invention provides a method and system for smart gateway applications, the main purpose of which is to improve the control accuracy of smart gateway applications.

[0005] To achieve the above objectives, the present invention provides a smart gateway application method, comprising:

[0006] Obtain the gateway application scenarios and network devices of the smart gateway to be applied, collect historical gateway data of the gateway application scenarios using the network devices, and identify the gateway factor categories of the historical gateway data.

[0007] Based on the gateway factor categories, construct factor fuzzy rules for gateway factors; based on the factor fuzzy rules and the gateway factor categories, construct a gateway monitoring model; and train the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model.

[0008] Acquire real-time gateway data and use the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data;

[0009] Based on the gateway evaluation level, the devices to be controlled among the networked devices are determined, and the device control policy is sent to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

[0010] Optionally, constructing the factor fuzzy rules for the gateway factors based on the gateway factor category includes:

[0011] Identify the associated factors and the result factors of the gateway factor category, construct the associated factor space corresponding to the associated factors, and construct the result factor space corresponding to the result factors;

[0012] The association membership function of the association factor is constructed using the following formula:

[0013]

[0014] Where μ(x) represents the association membership function, x represents the association factor, and a, b, and c represent the critical points of the association membership function;

[0015] Based on the association membership function and the pre-constructed fuzzy logic algorithm, a causal mapping relationship is constructed between the association factor space and the result factor space;

[0016] Based on the causal mapping relationship, construct the factor fuzzy rules for the gateway factor.

[0017] Optionally, constructing the causal mapping relationship between the association factor space and the result factor space based on the association membership function and the pre-built fuzzy logic algorithm includes:

[0018] Based on the association membership function, the association factor space is fuzzified to obtain the association membership degree of the association factor space;

[0019] Based on the association membership degree, a fuzzy score function of the result factor space is calculated using a pre-constructed fuzzy logic algorithm;

[0020] Construct the result membership function of the result factor space, and truncate the result membership function according to the fuzzy scoring function to obtain the truncated fuzzy function;

[0021] The truncated fuzzy function is aggregated to obtain a fuzzy aggregate function;

[0022] Based on the fuzzy aggregation function, the causal mapping relationship between the correlation factor space and the result factor space is determined.

[0023] Optionally, constructing the gateway monitoring model based on the factor fuzzy rules and the gateway factor categories includes:

[0024] Obtain the correlation factors of the gateway factor category and the result factors of the gateway factor category, and construct the comparison layer of the gateway monitoring model based on the correlation factors;

[0025] Based on the resulting factors, construct the identification layer of the gateway monitoring model;

[0026] And construct a causal mapping relationship between the correlation factor and the outcome factor;

[0027] The gateway monitoring model is constructed based on the comparison layer, the identification layer, and the causal mapping relationship.

[0028] Optionally, training the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model includes:

[0029] Configure the initial values ​​of the forward weight and the feedback weight of the gateway monitoring model respectively to obtain the initial values ​​of the forward weight and the feedback weight.

[0030] Calculate the activation value of the historical gateway data, and select the maximum activation value among the activation values;

[0031] Calculate the similarity between the historical gateway data and the initial value of the feedback weight. When the similarity is not greater than a preset warning parameter, remove the maximum activation value from the activation values ​​to obtain the remaining activation values. Return the activation values ​​of the historical gateway data and select the maximum activation value from the activation values.

[0032] When the similarity is greater than a preset warning parameter, the forward weight and the feedback weight are adjusted to obtain the forward adjustment weight and the feedback adjustment weight.

[0033] The trained gateway monitoring model is obtained based on the forward adjustment weights and the feedback adjustment weights.

[0034] Optionally, calculating the activation value of the historical gateway data includes:

[0035] The activation value of the historical gateway data is calculated using the following formula:

[0036]

[0037] in, w represents the activation value of the j-th neuron in the recognition layer of the gateway monitoring model for the k-th historical gateway data. ij This represents the current forward weights between the i-th neuron in the comparison layer and the j-th neuron in the recognition layer. This represents the i-th correlation factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer), where i represents the sequence number of the neuron in the comparison layer, j represents the sequence number of the neuron in the recognition layer, n represents the number of neurons in the comparison layer, m represents the number of neurons in the recognition layer, k represents the sample sequence number of the historical gateway data, and q represents the number of samples of the historical gateway data.

[0038] Optionally, calculating the similarity between the historical gateway data and the initial forward weight includes:

[0039]

[0040] in, t represents the similarity between the k-th historical gateway data and the initial value of the feedback weight of the g-th neuron in the recognition layer of the gateway monitoring model. gi This represents the initial feedback weight of the g-th neuron in the recognition layer and the i-th neuron in the comparison layer of the gateway monitoring model. This represents the i-th correlation factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer), where i represents the sequence number of the neuron in the comparison layer, and n represents the number of neurons in the comparison layer.

[0041] Optionally, when the similarity is greater than a preset warning parameter, adjusting the forward weight and the feedback weight to obtain the forward adjustment weight and the feedback adjustment weight includes:

[0042] The forward adjustment weights and feedback adjustment weights are obtained by adjusting the forward weights and feedback weights using the following formulas:

[0043]

[0044] Among them, w ig (t+1) represents the forward weight adjustment of the i-th neuron in the comparison layer and the g-th neuron in the recognition layer of the gateway monitoring model at time (t+1), where t gi (t) represents the feedback weight value between the g-th neuron of the recognition layer and the i-th neuron of the comparison layer at time t. t represents the i-th association factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer). gi (t+1) represents the feedback adjustment weight between the g-th neuron of the recognition layer and the i-th neuron of the comparison layer at time (t+1), where t represents time, i represents the sequence number of the neuron in the comparison layer, and n represents the number of neurons in the comparison layer.

[0045] Optionally, the calculation of determining the devices to be controlled among the networked devices based on the gateway evaluation level includes:

[0046] Based on the preset gateway evaluation criteria, identify the non-compliance levels in the gateway evaluation levels;

[0047] Identify the causal mapping relationship in the gateway monitoring model, and query the associated factors corresponding to the non-compliance level based on the causal mapping relationship;

[0048] Based on the correlation factors, the devices to be controlled in the networked devices are determined.

[0049] To address the above problems, the present invention also provides an intelligent gateway application system, the system comprising:

[0050] The factor category identification module is used to obtain the gateway application scenario and network devices of the smart gateway to be applied, and to collect historical gateway data of the gateway application scenario using the network devices to identify the gateway factor category of the historical gateway data.

[0051] The monitoring model training module is used to construct factor fuzzy rules for gateway factors based on the gateway factor categories, construct a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories, and train the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model.

[0052] The evaluation level calculation module acquires real-time gateway data and uses the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data.

[0053] The application result generation module is used to determine the devices to be controlled among the networked devices based on the gateway evaluation level, and send the device control policy to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

[0054] As can be seen, this embodiment of the invention can determine the actual application scenario and network devices of the smart gateway to be applied by acquiring the gateway application scenario and network devices, providing operational direction for subsequent control of the smart gateway. The historical gateway data collected from the network devices can provide training data for the subsequent construction of the gateway monitoring model. By identifying the gateway factor categories in the historical gateway data, the associated factors of the gateway application can be determined. Based on the gateway factor categories, constructing factor fuzzy rules for gateway factors can establish factor characteristics that are more consistent with the actual application scenario, making the gateway factors more meaningful for real-world interpretation, thus providing a basis for the subsequent construction of the gateway monitoring model. Secondly, this embodiment of the invention can determine the application monitoring model of the smart gateway by constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories, so as to accurately control the network devices corresponding to the smart gateway based on the model results. Furthermore, training the gateway monitoring model based on the historical gateway data can yield a more intelligent gateway monitoring model. This model possesses a self-learning and memory expansion capability lacking in most neural network models. It can update the model in real-time after stabilization, resulting in better and more accurate model fitting. By acquiring real-time gateway data, it can provide real-time monitoring data for the gateway monitoring model, predicting the gateway state in real-time. Using the trained gateway monitoring model, the gateway evaluation level corresponding to the real-time gateway data can be calculated to determine the current evaluation level of the smart gateway, providing a basis for subsequent control of the corresponding network devices to achieve the ideal gateway state. Furthermore, based on the gateway evaluation level, this embodiment of the invention determines the device to be controlled among the network devices, ultimately locating the network device to be controlled, thereby achieving precise control of the smart gateway application. According to a pre-built device control strategy, the device control strategy is sent to the device to be controlled to provide the control premise for subsequent precise identification of the device to be controlled and execution of the corresponding control strategy. Executing the device control strategy yields the application result of the smart gateway to be applied, achieving intelligent and precise control of the smart network to be applied. Therefore, the smart gateway application method and system proposed in this embodiment of the invention can improve the control accuracy of smart gateway applications. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating a smart gateway application method provided in an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of a smart gateway application system provided in an embodiment of the present invention;

[0057] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0058] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0059] This invention provides a smart gateway application method. The execution subject of the smart gateway application method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this invention: a server, a terminal, etc. In other words, the smart gateway application method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0060] Reference Figure 1 The diagram shown is a flowchart illustrating a smart gateway application method according to an embodiment of the present invention. In this embodiment, the smart gateway application method includes:

[0061] S1. Obtain the gateway application scenario and network device of the smart gateway to be applied, and use the network device to collect historical gateway data of the gateway application scenario, and identify the gateway factor category of the historical gateway data.

[0062] In this embodiment of the invention, the gateway application scenario refers to the scenario in which the gateway is used, such as gateway application scenarios in transportation, energy and mining, power, meteorology, industrial automation, finance, water conservancy, environmental protection, medical care, agriculture, forestry and oil, etc. The networked device refers to a processing device that connects to the smart gateway via a communication network and can realize data acquisition, processing, and calculation in the gateway application environment, such as smart switches, smart air conditioners, cameras, smart lighting equipment, etc. Both the gateway application scenario and the networked device can be obtained through data scripts, which can be compiled using the JS scripting language.

[0063] In this embodiment of the invention, the historical gateway data collected by the network device for the gateway application scenario can provide training data for the subsequent construction of the gateway monitoring model.

[0064] This invention, through identifying the gateway factor categories in historical gateway data, can determine the associated factors of gateway applications. These gateway factor categories can be identified through methods such as on-site observation, investigation, and process analysis. The gateway factor category refers to the type of gateway environmental element that has a direct or indirect impact on the gateway application environment, such as temperature, humidity, carbon dioxide concentration, and comfort level.

[0065] S2. Based on the gateway factor category, construct the factor fuzzy rule of the gateway factor; based on the factor fuzzy rule and the gateway factor category, construct the gateway monitoring model; based on the historical gateway data, train the gateway monitoring model to obtain the trained gateway monitoring model.

[0066] This invention, by constructing factor fuzzy rules for gateway factors based on the gateway factor categories, can establish factor characteristics that better reflect real-world application scenarios, making gateway factors more meaningful for practical interpretation and providing a basis for subsequent gateway monitoring model construction. The factor fuzzy rules refer to the regularity representation of gateway factors.

[0067] As an optional embodiment of the present invention, the step of constructing factor fuzzy rules for gateway factors based on the gateway factor category includes:

[0068] Identify the associated factors and the result factors of the gateway factor category, construct the associated factor space corresponding to the associated factors, and construct the result factor space corresponding to the result factors;

[0069] The association membership function of the association factor is constructed using the following formula:

[0070]

[0071] Where μ(x) represents the association membership function, x represents the association factor, and a, b, and c represent the critical points of the association membership function;

[0072] Based on the association membership function and the pre-constructed fuzzy logic algorithm, a causal mapping relationship is constructed between the association factor space and the result factor space;

[0073] Based on the causal mapping relationship, construct the factor fuzzy rules for the gateway factor.

[0074] The association membership function refers to the function represented by A(x) ∈ [0, 1] for any element x in the universe of discourse (the scope of study), when x changes in U. The fuzzy logic algorithm is based on multi-valued logic and uses fuzzy set theory to study fuzzy thinking, language forms, and their patterns.

[0075] Furthermore, as an optional embodiment of the present invention, the correlation factors for identifying the gateway factor categories and the result factors for identifying the gateway factor categories can be identified by analyzing the gateway application scenarios and scenario requirement targets of the smart gateway to be applied. The construction of the correlation factor space corresponding to the correlation factors and the construction of the result factor space corresponding to the result factors can be based on the correlation factors and the result factors, respectively establishing fuzzy sets for the correlation factors and fuzzy sets for the result factors. For example, defining indoor temperatures <4℃ as very cold, 4℃-8℃ as cold, 8℃-13℃ as cool, 13℃-18℃ as refreshingly cool, 18℃-23℃ as comfortable, 23℃-29℃ as warm, and 29℃-35℃ as very hot, etc.

[0076] Further, as an optional embodiment of the present invention, the step of constructing the causal mapping relationship between the correlation factor space and the result factor space based on the correlation membership function and the pre-constructed fuzzy logic algorithm includes: fuzzifying the correlation factor space according to the correlation membership function to obtain the correlation membership degree of the correlation factor space; calculating the fuzzy score function of the result factor space using the pre-constructed fuzzy logic algorithm based on the correlation membership degree; constructing the result membership function of the result factor space; truncating the result membership function according to the fuzzy score function to obtain a truncated fuzzy function; aggregating the truncated fuzzy function to obtain a fuzzy aggregation function; and determining the causal mapping relationship between the correlation factor space and the result factor space based on the fuzzy aggregation function.

[0077] Furthermore, in this embodiment of the invention, by constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories, an application monitoring model for the smart gateway can be determined, so as to accurately control the network devices corresponding to the smart gateway based on the model results.

[0078] Further, as an optional embodiment of the present invention, the step of constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories includes: obtaining the correlation factors of the gateway factor categories and the result factors of the gateway factor categories; constructing a comparison layer of the gateway monitoring model based on the correlation factors; constructing an identification layer of the gateway monitoring model based on the result factors; and constructing a causal mapping relationship between the correlation factors and the result factors; and constructing the gateway monitoring model based on the comparison layer, the identification layer, and the causal mapping relationship.

[0079] Further, as an optional embodiment of the present invention, the comparison layer of the gateway monitoring model constructed based on the correlation factor can be constructed with the same number of correlation factors as the correlation factor, and the comparison neurons are used as the comparison layer. The recognition layer of the gateway monitoring model constructed based on the result factor can be constructed with the same number of recognition neurons as the result factor, and the recognition neurons are used as the recognition layer. The causal mapping relationship between the correlation factor and the result factor can be constructed based on the logical causal relationship between the correlation factor and the result factor.

[0080] Furthermore, by training the gateway monitoring model based on the historical gateway data, this embodiment of the invention can obtain a more intelligent gateway monitoring model. This model possesses a self-learning and memory-expanding capability lacking in most types of neural network models, allowing for real-time model updates after stabilization, resulting in better and more accurate model fitting. The gateway monitoring model is a neural network theoretical model capable of self-organizing and generating encodings of gateway environment awareness.

[0081] Further, as an optional embodiment of the present invention, training the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model includes: configuring initial values ​​for the forward weights and feedback weights of the gateway monitoring model respectively to obtain initial values ​​for the forward weights and feedback weights; calculating the activation values ​​of the historical gateway data and selecting the maximum activation value among the activation values; calculating the similarity between the historical gateway data and the initial value of the feedback weights; when the similarity is not greater than a preset warning parameter, removing the maximum activation value from the activation values ​​to obtain the remaining activation values; returning the calculation of the activation values ​​of the historical gateway data and selecting the maximum activation value among the activation values; when the similarity is greater than the preset warning parameter, adjusting the forward weights and the feedback weights to obtain forward adjustment weights and feedback adjustment weights; and obtaining the trained gateway monitoring model based on the forward adjustment weights and feedback adjustment weights. The warning parameter refers to a key parameter for determining whether the system is functioning normally, and it can be set according to the specific application scenario.

[0082] Further, as an optional embodiment of the present invention, configuring the initial values ​​of the forward weights and feedback weights of the gateway monitoring model to obtain the initial values ​​of the forward weights and feedback weights includes:

[0083] The initial values ​​of the forward weights and feedback weights of the gateway monitoring model are configured using the following formulas to obtain the initial values ​​of the forward weights and feedback weights:

[0084]

[0085] i = 1, 2, ..., nj = 1, 2, ..., m

[0086] Among them, w ij (0) represents the initial forward weight between the i-th neuron in the comparison layer and the j-th neuron in the recognition layer of the gateway monitoring model, t ji (0) represents the initial value of the feedback weight between the j-th neuron in the recognition layer and the i-th neuron in the comparison layer, where i represents the sequence number of the neuron in the comparison layer, j represents the sequence number of the neuron in the recognition layer, n represents the number of neurons in the comparison layer, and m represents the number of neurons in the recognition layer.

[0087] Further, as an optional embodiment of the present invention, calculating the activation value of the historical gateway data includes:

[0088] The activation value of the historical gateway data is calculated using the following formula:

[0089]

[0090] in, w represents the activation value of the j-th neuron in the recognition layer of the gateway monitoring model for the k-th historical gateway data. ij This represents the current forward weights between the i-th neuron in the comparison layer and the j-th neuron in the recognition layer. This represents the i-th correlation factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer), where i represents the sequence number of the neuron in the comparison layer, j represents the sequence number of the neuron in the recognition layer, n represents the number of neurons in the comparison layer, m represents the number of neurons in the recognition layer, k represents the sample sequence number of the historical gateway data, and q represents the number of samples of the historical gateway data.

[0091] Selecting the maximum activation value from the activation values ​​includes:

[0092] The maximum activation value among the activation values ​​is selected using the following formula:

[0093]

[0094] in, This represents the maximum activation value of the k-th historical gateway data in the identification layer of the gateway monitoring model. This represents the activation value of the k-th historical gateway data in the j-th neuron of the recognition layer, where j represents the sequence number of the neuron in the recognition layer, m represents the number of neurons in the recognition layer, and max represents the maximum value symbol.

[0095] Further, as an optional embodiment of the present invention, calculating the similarity between the historical gateway data and the initial forward weight includes:

[0096] The similarity between the historical gateway data and the initial value of the feedback weight is calculated using the following formula:

[0097]

[0098] in, t represents the similarity between the k-th historical gateway data and the initial value of the feedback weight of the g-th neuron in the recognition layer of the gateway monitoring model. gi This represents the initial feedback weight of the g-th neuron in the recognition layer and the i-th neuron in the comparison layer of the gateway monitoring model. This represents the i-th correlation factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer), where i represents the sequence number of the neuron in the comparison layer, and n represents the number of neurons in the comparison layer.

[0099] Further, as an optional embodiment of the present invention, adjusting the forward weight and the feedback weight when the similarity is greater than a preset warning parameter to obtain the forward adjustment weight and the feedback adjustment weight includes:

[0100] The forward adjustment weights and feedback adjustment weights are obtained by adjusting the forward weights and feedback weights using the following formulas:

[0101]

[0102]

[0103] Among them, w ig (t+1) represents the forward weight adjustment of the i-th neuron in the comparison layer and the g-th neuron in the recognition layer of the gateway monitoring model at time (t+1), where t gi (t) represents the feedback weight value between the g-th neuron of the recognition layer and the i-th neuron of the comparison layer at time t. t represents the i-th association factor data of the k-th historical gateway data (corresponding to the neuron of the comparison layer). gi (t+1) represents the feedback adjustment weight between the g-th neuron of the recognition layer and the i-th neuron of the comparison layer at time (t+1), where t represents time, i represents the sequence number of the neuron in the comparison layer, and n represents the number of neurons in the comparison layer.

[0104] S3. Obtain real-time gateway data, and use the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data.

[0105] This invention provides real-time monitoring data for a gateway monitoring model by acquiring real-time gateway data, which can be used to predict the gateway status in real time. This data can be acquired through a data script, which can be compiled using the JS scripting language.

[0106] Furthermore, this embodiment of the invention utilizes the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data, thereby determining the current evaluation level of the smart gateway. This is then used to control the corresponding networked devices to achieve an ideal gateway state. The gateway evaluation level refers to the degree to which the application data of the gateway to be evaluated conforms to the preset gateway evaluation criteria, such as levels like very cold, cold, comfortable, hot, and very hot.

[0107] S4. Based on the gateway evaluation level, determine the devices to be controlled among the networked devices, and send the device control policy to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

[0108] This invention, based on the gateway evaluation level, determines the device to be controlled among the networked devices, and can ultimately locate the networked device to be controlled, thereby achieving precise control of the smart gateway application.

[0109] Further, as an optional embodiment of the present invention, determining the devices to be controlled among the networked devices based on the gateway evaluation level includes: identifying the non-compliance levels in the gateway evaluation level according to a preset gateway evaluation standard; identifying the causal mapping relationship in the gateway monitoring model, querying the correlation factors corresponding to the non-compliance level according to the causal mapping relationship; and determining the devices to be controlled among the networked devices according to the correlation factors. Here, the gateway evaluation standard refers to the value scale and boundaries applied to objects in gateway evaluation activities.

[0110] Furthermore, in this embodiment of the invention, a pre-constructed device control strategy is sent to the device to be controlled to provide a control prerequisite for accurately identifying the device to be controlled and executing the corresponding control strategy. The device control strategy can be sent to the device to be controlled via wireless communication technology. This wireless communication technology refers to a communication method that utilizes the characteristic that electromagnetic waves can propagate in free space for information exchange, including radio frequency (RF) technology, Bluetooth technology, Z-Wave technology, ZigBee technology, WiFi technology, infrared technology, etc. The device control strategy is a strategy and method for controlling a system or instrument. For example, in a smart home scenario, the comfort level of the home environment is related to temperature, humidity, indoor brightness, carbon dioxide concentration, etc. Based on the current comfort level, a smart gateway can be used to control devices such as smart air conditioners, smart humidifiers, smart lights, and smart oxygen generators to achieve a personalized home comfort environment.

[0111] Furthermore, by executing the device control strategy, the application results of the smart gateway to be applied can be obtained in the embodiments of the present invention, thereby realizing intelligent and precise control of the smart network to be applied.

[0112] As can be seen, this embodiment of the invention can determine the actual application scenario and network devices of the smart gateway to be applied by acquiring the gateway application scenario and network devices, providing operational direction for subsequent control of the smart gateway. The historical gateway data collected from the network devices can provide training data for the subsequent construction of the gateway monitoring model. By identifying the gateway factor categories in the historical gateway data, the associated factors of the gateway application can be determined. Based on the gateway factor categories, constructing factor fuzzy rules for gateway factors can establish factor characteristics that are more consistent with the actual application scenario, making the gateway factors more meaningful for real-world interpretation, thus providing a basis for the subsequent construction of the gateway monitoring model. Secondly, this embodiment of the invention can determine the application monitoring model of the smart gateway by constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories, so as to accurately control the network devices corresponding to the smart gateway based on the model results. Furthermore, training the gateway monitoring model based on the historical gateway data can yield a more intelligent gateway monitoring model. This model possesses a self-learning and memory expansion capability lacking in most neural network models. It can update the model in real-time after stabilization, resulting in better and more accurate model fitting. By acquiring real-time gateway data, it can provide real-time monitoring data for the gateway monitoring model, predicting the gateway state in real-time. Using the trained gateway monitoring model, the gateway evaluation level corresponding to the real-time gateway data can be calculated to determine the current evaluation level of the smart gateway, providing a basis for subsequent control of the corresponding network devices to achieve the ideal gateway state. Furthermore, based on the gateway evaluation level, this embodiment of the invention determines the device to be controlled among the network devices, ultimately locating the network device to be controlled, thereby achieving precise control of the smart gateway application. According to a pre-built device control strategy, the device control strategy is sent to the device to be controlled to provide the control premise for subsequent precise identification of the device to be controlled and execution of the corresponding control strategy. Executing the device control strategy yields the application result of the smart gateway to be applied, achieving intelligent and precise control of the smart network to be applied. Therefore, the smart gateway application method and system proposed in this embodiment of the invention can improve the control accuracy of smart gateway applications.

[0113] like Figure 2 The diagram shown is a functional block diagram of the intelligent gateway application system of the present invention.

[0114] The intelligent gateway application system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the intelligent gateway application system may include a factor category identification module 101, a monitoring model training module 102, an evaluation level calculation module 103, and an application result generation module 104. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0115] In this embodiment, the functions of each module / unit are as follows:

[0116] The factor category identification module 101 is used to obtain the gateway application scenario and network device of the smart gateway to be applied, and to collect historical gateway data of the gateway application scenario using the network device to identify the gateway factor category of the historical gateway data.

[0117] The monitoring model training module 102 is used to construct factor fuzzy rules for gateway factors according to the gateway factor categories, construct a gateway monitoring model according to the factor fuzzy rules and the gateway factor categories, and train the gateway monitoring model according to the historical gateway data to obtain a trained gateway monitoring model.

[0118] The evaluation level calculation module 103 acquires real-time gateway data and uses the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data.

[0119] The application result generation module 104 is used to determine the devices to be controlled among the networked devices based on the gateway evaluation level, and send the device control policy to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

[0120] In detail, the modules in the intelligent gateway application system 100 described in this embodiment of the invention employ the same methods as described above. Figure 1 The same technical means are used in the application of the smart gateway described above, and the same technical effects can be produced, so they will not be repeated here.

[0121] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in terms of the scope of the patent invention.

[0122] In the several embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0123] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0124] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0125] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0126] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0127] The embodiments of this invention can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that utilize digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0128] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The term "second class" is used to indicate names and does not indicate any specific order.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart gateway application method, characterized by, The method includes: Obtain the gateway application scenario and network devices of the smart gateway to be applied, collect historical gateway data of the gateway application scenario using the network devices, and identify the gateway factor category of the historical gateway data. The gateway factor categories can be identified through methods such as on-site observation, investigation, and process analysis and judgment. The gateway factor category refers to the type of gateway environment element that has a direct or indirect impact on the application of the gateway in the gateway application environment. Based on the gateway factor categories, construct factor fuzzy rules for gateway factors; based on the factor fuzzy rules and the gateway factor categories, construct a gateway monitoring model; and train the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model. The step of constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories includes: Obtain the associated factors and the result factors of the gateway factor categories, and construct the comparison layer of the gateway monitoring model based on the associated factors; Based on the resulting factors, construct the identification layer of the gateway monitoring model; And construct a causal mapping relationship between the correlation factor and the outcome factor; The gateway monitoring model is constructed based on the comparison layer, the identification layer, and the causal mapping relationship. Acquire real-time gateway data and use the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data; Based on the gateway evaluation level, the devices to be controlled among the networked devices are determined, including: Based on the preset gateway evaluation criteria, identify the non-compliance levels in the gateway evaluation levels; Identify the causal mapping relationship in the gateway monitoring model, and query the associated factors corresponding to the non-compliance level based on the causal mapping relationship; Based on the correlation factors, determine the devices to be controlled among the networked devices; Based on the gateway evaluation level, the devices to be controlled among the networked devices are determined, and the device control policy is sent to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

2. The intelligent gateway application method of claim 1, wherein, The step of constructing the factor fuzzy rules for gateway factors based on the gateway factor categories includes: Identify the associated factors and the result factors of the gateway factor category, construct the associated factor space corresponding to the associated factors, and construct the result factor space corresponding to the result factors; The association membership function of the association factor is constructed using the following formula: wherein, denotes a correlation membership function, denotes a correlation factor, , , denotes a critical point of the correlation membership function; Based on the association membership function and the pre-constructed fuzzy logic algorithm, a causal mapping relationship is constructed between the association factor space and the result factor space; Based on the causal mapping relationship, construct the factor fuzzy rules for the gateway factor.

3. The intelligent gateway application method of claim 2, wherein, The step of constructing a causal mapping relationship between the correlation factor space and the result factor space based on the correlation membership function and a pre-constructed fuzzy logic algorithm includes: Based on the association membership function, the association factor space is fuzzified to obtain the association membership degree of the association factor space; Based on the association membership degree, a fuzzy score function of the result factor space is calculated using a pre-constructed fuzzy logic algorithm; Construct the result membership function of the result factor space, and truncate the result membership function according to the fuzzy scoring function to obtain the truncated fuzzy function; The truncated fuzzy function is aggregated to obtain a fuzzy aggregate function; Based on the fuzzy aggregation function, the causal mapping relationship between the correlation factor space and the result factor space is determined.

4. The intelligent gateway application method of claim 1, wherein, The step of training the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model includes: Configure the initial values ​​of the forward weight and the feedback weight of the gateway monitoring model respectively to obtain the initial values ​​of the forward weight and the feedback weight. Calculate the activation value of the historical gateway data, and select the maximum activation value among the activation values; Calculate the similarity between the historical gateway data and the initial value of the feedback weight. When the similarity is not greater than a preset warning parameter, remove the maximum activation value from the activation values ​​to obtain the remaining activation values. Return the activation values ​​of the historical gateway data and select the maximum activation value from the activation values. When the similarity is greater than a preset warning parameter, the forward weight and the feedback weight are adjusted to obtain the forward adjustment weight and the feedback adjustment weight. The trained gateway monitoring model is obtained based on the forward adjustment weights and the feedback adjustment weights.

5. The intelligent gateway application method of claim 4, wherein, The calculation of the activation value of the historical gateway data includes: The activation value of the historical gateway data is calculated using the following formula: in, Indicates the first The historical gateway data is in the identification layer of the gateway monitoring model. The activation value of each neuron. Indicates the comparison level. The first neuron and the recognition layer The current forward weights between neurons Indicates the first The first historical gateway data Each associated factor data (corresponding to a neuron in the comparison layer). Indicates the sequence number of the neurons in the comparison layer. Indicates the sequence number of the neurons in the recognition layer. This indicates the number of neurons in the comparison layer. This indicates the number of neurons in the recognition layer. This indicates the sample sequence number of the historical gateway data. This indicates the number of samples in the historical gateway data.

6. The smart gateway application method as described in claim 4, characterized in that, The calculation of the similarity between the historical gateway data and the initial forward weight includes: in, Indicates the first The first layer of historical gateway data and gateway monitoring model identification. The similarity between the initial feedback weights of each neuron. Indicates the identification layer number 1 The comparison layer of the neuron and the gateway monitoring model. Initial values ​​of the feedback weights for each neuron. Indicates the first The first historical gateway data Each associated factor data (corresponding to a neuron in the comparison layer). This indicates the sequence number of the neurons in the comparison layer. This indicates the number of neurons in the comparison layer.

7. The intelligent gateway application method of claim 4, wherein, When the similarity is greater than a preset warning parameter, the forward weight and the feedback weight are adjusted to obtain the forward adjustment weight and the feedback adjustment weight, including: The forward adjustment weights and feedback adjustment weights are obtained by adjusting the forward weights and feedback weights using the following formulas: ； ； in, Indicates in Comparison layer of the time gateway monitoring model The recognition layer of the neuron and gateway monitoring model Forward adjustment of weights for each neuron Indicates in The recognition layer at time 1 The neuron and the comparison layer The feedback weight values ​​of each neuron. Indicates the first The first historical gateway data Each associated factor data (corresponding to a neuron in the comparison layer). Indicates in The recognition layer at time 1 The neuron and the comparison layer The feedback of each neuron adjusts the weights. Indicates time, This indicates the sequence number of the neurons in the comparison layer. This indicates the number of neurons in the comparison layer.

8. An intelligent gateway application system characterized by, The system includes: The factor category identification module is used to obtain the gateway application scenario and network devices of the smart gateway to be applied, and to collect historical gateway data of the gateway application scenario using the network devices to identify the gateway factor category of the historical gateway data. The gateway factor categories can be identified through methods such as on-site observation, investigation, and process analysis and judgment. The gateway factor category refers to the type of gateway environment element that has a direct or indirect impact on the application of the gateway in the gateway application environment. The monitoring model training module is used to construct factor fuzzy rules for gateway factors based on the gateway factor categories, construct a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories, and train the gateway monitoring model based on the historical gateway data to obtain a trained gateway monitoring model. The step of constructing a gateway monitoring model based on the factor fuzzy rules and the gateway factor categories includes: Obtain the correlation factors of the gateway factor category and the result factors of the gateway factor category, and construct the comparison layer of the gateway monitoring model based on the correlation factors; Based on the resulting factors, construct the identification layer of the gateway monitoring model; And construct a causal mapping relationship between the correlation factor and the outcome factor; The gateway monitoring model is constructed based on the comparison layer, the identification layer, and the causal mapping relationship. The evaluation level calculation module acquires real-time gateway data and uses the trained gateway monitoring model to calculate the gateway evaluation level corresponding to the real-time gateway data. Based on the gateway evaluation level, the devices to be controlled among the networked devices are determined, including: Based on the preset gateway evaluation criteria, identify the non-compliance levels in the gateway evaluation levels; Identify the causal mapping relationship in the gateway monitoring model, and query the associated factors corresponding to the non-compliance level based on the causal mapping relationship; Based on the correlation factors, determine the devices to be controlled among the networked devices; The application result generation module is used to determine the devices to be controlled among the networked devices based on the gateway evaluation level, and send the device control policy to the devices to be controlled according to the pre-built device control policy to execute the device control policy and obtain the application result of the smart gateway to be applied.

Images