Internet of things self-adaptive networking method and system for business state energy analysis

Abstract

The invention relates to the technical field of business state energy analysis, and provides an Internet of Things adaptive networking method and system for business state energy analysis, and the method comprises the steps: firstly obtaining business state feature data of the Internet of Things in different business states and energy consumption data in the corresponding business states, and building a prediction model according to the obtained data; establishing a networking strategy generation model for generating an energy optimization direction according to a prediction result of the prediction model; when the business state is changed, inputting the business state characteristic data into the prediction model to obtain energy consumption data of subsequent prediction in the corresponding business state, generating a corresponding networking strategy according to the prediction data, and performing adaptive networking according to the networking strategy and the Internet of Things; according to the method, the prediction model is established, the business state characteristic data is input into the model for prediction when business state demands are changed, the influence caused by system response lag is solved, meanwhile, the networking strategy is generated according to the prediction result, and the energy optimization demands of adaptive networking of the Internet of Things in various business states are met.

Description

Technical Field

[0001] The present invention relates to the technical field of business energy analysis, and in particular to an Internet of Things adaptive networking method and system for business energy analysis. Background Art

[0002] Business energy analysis is a comprehensive energy assessment and insight into IoT scenarios under specific business forms. It covers in-depth research on the power required for IoT device operation, energy consumption during data transmission, and energy supply sources and sustainability in different business forms.

[0003] The IoT adaptive networking is an advanced networking method that can dynamically adjust network structure and parameters according to environmental changes and business needs. It uses advanced sensor technology to sense changes in the surrounding physical environment, network environment and business needs, and based on these perceptions, automatically adjusts parameters to adapt to new conditions; it can intelligently adjust networking strategies according to different business needs and environmental conditions, thereby reducing energy consumption.

[0004] The essential purpose of business energy analysis is to conduct an in-depth analysis of the energy of IoT devices under different business formats, including the energy required for device operation, data transmission energy consumption, and sustainability of energy supply. Based on this, the adaptive networking of the IoT can be realized, and the network structure and parameters can be dynamically adjusted according to the specific business energy status to improve the efficiency and reliability of the network.

[0005] For example, in certain periods of low business demand, it automatically reduces unnecessary device connections and data transmission to save energy. At the same time, it can also reasonably allocate network resources according to the energy supply situation to ensure that the network can still operate stably under limited energy conditions. In addition, by sensing environmental changes, adaptive networking can adjust parameters such as transmission power in a timely manner to avoid energy waste and improve energy utilization efficiency. With this networking method, it can better adapt to complex and changeable IoT application scenarios, extend the service life of IoT devices, and reduce operating costs. In practical applications, these can not only optimize the performance of IoT systems in different business scenarios, but also reduce energy consumption, improve energy utilization efficiency, and provide more stable, efficient and sustainable network support for IoT applications in various formats, and promote the widespread application and development of IoT technology in different fields.

[0006] However, the prior art still has the following disadvantages:

[0007] 1. The real-time performance of business energy analysis needs to be further improved. The current analysis method may be difficult to accurately capture real-time energy changes in complex and changeable business scenarios. The lack of real-time performance makes it difficult to perceive energy changes in time when the Internet of Things is networked, resulting in a certain lag in the response of the adaptive networking, which makes it impossible for the equipment to be adjusted to low-energy mode or shut down non-critical functions in time, affecting the normal use of the equipment. Further, the Internet of Things equipment may suddenly lose power or performance due to inaccurate energy analysis at critical business moments, thereby affecting the normal use of related Internet of Things equipment.

[0008] 2. Current business energy analysis often fails to accurately analyze the dynamic energy demand changes of IoT devices in different business scenarios. As a result, the networking strategy generated based on this analysis may not be able to adapt to the energy demand fluctuations in the actual operation of the equipment in a timely manner during energy optimization, resulting in inaccurate and inefficient energy optimization, and unable to fully meet the energy optimization needs of IoT adaptive networking in various complex business scenarios. Summary of the invention

[0009] In order to solve the above technical problems, the present invention provides an Internet of Things adaptive networking method and system for business energy analysis, which performs business energy prediction of the Internet of Things through business energy analysis, and generates an Internet of Things adaptive networking strategy based on the energy optimization direction, and performs Internet of Things adaptive networking to solve the problems in the prior art.

[0010] An Internet of Things adaptive networking method for business energy analysis, the specific steps are as follows:

[0011] S1: Obtaining energy information of IoT business forms, using sensor networks to obtain business form characteristic data of different IoT business forms, as well as energy consumption data of IoT devices under corresponding business forms, where energy consumption data includes IoT device self-consumption data and energy transmission consumption data;

[0012] S2: Establish an energy consumption data prediction model, which can predict energy consumption based on the acquired business characteristic data under different business formats and the energy consumption data of IoT devices under the corresponding business formats, and generate energy consumption prediction results;

[0013] S3: Establish an IoT adaptive networking strategy generation model, which can generate an IoT adaptive networking strategy in the energy optimization direction according to the energy consumption prediction results of the energy consumption data prediction model;

[0014] S4: Adaptive networking optimization of the Internet of Things. When the business demand changes, the changed business characteristic data is input into the energy consumption data prediction model to obtain the subsequent predicted energy consumption data under the corresponding business demand. The predicted energy consumption data is input into the adaptive networking strategy generation model of the Internet of Things to generate the corresponding networking strategy. Then the Internet of Things adjusts the Internet of Things according to the newly generated networking strategy to cope with the changes in business demand.

[0015] Preferably, in S1, the specific steps of obtaining the IoT business energy information are as follows:

[0016] S11: Use sensor networks to obtain business characteristic data under different business forms of the Internet of Things;

[0017] S12: Obtain energy consumption data of IoT devices corresponding to different business forms of IoT;

[0018] S13: Classify the acquired data into business type characteristic data sets according to different business types, each business type characteristic data set includes corresponding business type characteristic data and energy consumption data of IoT devices under the corresponding business type.

[0019] Preferably, in S2, the specific steps of establishing the energy consumption data prediction model are as follows:

[0020] S21: Data presetting, the collected energy consumption data of IoT devices are combined and arranged in time series, assuming that t represents a certain moment, and the t corresponding to each piece of energy consumption data of IoT devices collected constitutes a set (t), i and j represent different IoT devices, and i×j represents that IoT device i is connected to IoT device j; the energy consumption generated by each i×j connection is predicted and calculated;

[0021] S22: Model construction, extract and input corresponding energy consumption data according to time series:

[0022] S221: Data initialization, randomly initialize weight W, bias term b, prediction weight ω and prediction bias β; predict the initial value C of latent number h and memory warehouse, which are initialized to 0 when the data is first calculated;

[0023] S222: Information filtering calculation, information filtering determines which information can be deleted from the memory warehouse:

[0024]

[0025] in, Represents the dot product operation of vectors, f t Indicates the input information, X t Determine whether to filter the result; t-1is the predicted hidden number at time t-1, and is the information transmission carrier for predicting the energy value of the business state at time t. W and b represent the weight and bias terms respectively. σ represents the logical function, and tanh is the hyperbolic tangent function. t and the previous moment prediction implicit number h t-1 The processing is used to determine the degree of information forgetting in the cell state at the previous moment, so as to achieve dynamic adjustment of memory;

[0026] S223: Information update calculation and memory warehouse update value calculation:

[0027]

[0028] Information updating determines what new information is stored in the memory warehouse. First, i t Based on the current input X t and the previous moment prediction implicit number h t-1 By weight W i And the logic function σ determines which of the new input features are important energy information, Update values ​​of the memory warehouse to provide a data basis for subsequent updates of the memory warehouse;

[0029] S224: Update memory warehouse:

[0030]

[0031] S225: Information extraction calculation:

[0032]

[0033] h t =o t ×tanh(C t )

[0034] Among them, C t is the memory store at time t, C t Understood as a vector, it contains information or elements of many dimensions. t The weights, biases, and memory warehouses are used to determine which information is currently needed, that is, to predict the latent number h t , predict the latent number h t for o t and memory warehouse C t The result of multiplication after the hyperbolic tangent function, which means o t Will be in the memory warehouse C t Each element of is scaled if o t The smaller it is, the more important the corresponding element in the memory warehouse is in predicting the latent number h tThe contribution will be smaller; by judging the size of the contribution, we can determine which information is currently needed. The larger the contribution, the easier it is to extract.

[0035] S226: Repeat S221-S224 until all t in the set (t) are calculated;

[0036] S227: Finally use the predicted hidden number h t Conduct business energy forecast:

[0037]

[0038] Where E represents the predicted energy consumption data, That is, the energy consumption data of ij predicted at t+1, X t is the business characteristic data input at time t; ω and β are the prediction weight and prediction bias used in the prediction calculation;

[0039] S23: Model optimization and adjustment, determine the loss function to measure the difference between the predicted value and the actual value, and select the stochastic gradient descent algorithm to update the model parameters.

[0040] Preferably, in S23, the specific process of the model optimization adjustment is as follows:

[0041] S231: Define the loss function and select the mean square error or mean absolute error loss function to measure the difference between the predicted value and the actual value;

[0042] S232: Determine a stochastic gradient descent algorithm as an optimization algorithm;

[0043] S233: Divide the data set, dividing the actual value data into a training data set according to a stratified sampling method;

[0044] S234: Model training, performing multiple iterations of training on the training data set. In each iteration, the predicted value of the energy consumption data is calculated by forward propagation, and then the loss is calculated according to the loss function, and then the gradient is calculated by the back propagation algorithm, and the parameters of the model are updated using the stochastic gradient descent algorithm;

[0045] S235: Parameter adjustment. The validation set is used to adjust the parameters of the model. A grid search or random search method is used to find the parameter combination that minimizes the model result. The parameters include weights, bias terms, prediction weights, and prediction biases. Confirm that the data with the smallest loss function value is selected as the model parameter.

[0046] Preferably, in S3, the steps of establishing an IoT adaptive networking strategy generation model are as follows:

[0047] S31: It is represented as the predicted energy consumption data at time t when i and j are connected. At the same time, the connection mode between i and j is set to k, and R k is the indicator function. When i and j are connected using connection method k, R k is 1, otherwise it is 0;

[0048] S32: constructing a device connection relationship, based on the functional characteristics and spatial position relationship of the devices, constructing a series of device connection relationships, and calculating their total energy consumption, considering only the energy consumption between the actually connected devices and the energy consumption of the used transmission method;

[0049] For each device connection combination i×j, the total energy consumption E 总 for:

[0050]

[0051] Where p is the total number of IoT devices in the connection combination, and q is the total number of connection modes k;

[0052] Connection stability constraint: For any k, if we adjust from connection mode k1 to connection combination k . , only accept the adjustment if the connection is more stable after the adjustment, that is:

[0053]

[0054] Among them, μ represents the stability of the connection mode, which is determined by the specific connection mode;

[0055] S33: gradually optimize the connection relationship, compare the total energy consumption values ​​of different assumed connection combinations, recalculate the total energy consumption after each combination, and accept the adjustment if the new connection combination reduces the total energy consumption; otherwise, continue to try other adjustments.

[0056] Preferably, in S4, the specific steps of the Internet of Things adaptive networking optimization are as follows:

[0057] S41: When the business demand changes, determine the changed business demand;

[0058] S42: inputting the determined business demand into the energy consumption data prediction model in real time, and obtaining subsequent predicted energy consumption data under the corresponding business demand through the long-term and short-term prediction model;

[0059] S43: inputting the predicted energy consumption data into the IoT adaptive networking strategy generation model to generate a corresponding networking strategy;

[0060] S44: The Internet of Things automatically adjusts the connection relationship between the Internet of Things devices according to the generated networking strategy, and each Internet of Things device communicates or connects according to the new strategy to achieve adaptive networking.

[0061] An Internet of Things adaptive networking system for business energy analysis, comprising:

[0062] A business data collection module, which uses a sensor network to obtain business characteristic data under different business forms of the Internet of Things, as well as energy consumption data of Internet of Things devices under corresponding business forms, including consumption data of the Internet of Things devices themselves and energy transmission consumption data;

[0063] A prediction model building module, wherein the prediction model is data-connected to the business format data collection module, and includes a data preprocessing unit, a model building unit, and a model training unit. The data preprocessing unit cleans and denoises the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model building unit establishes an energy consumption data prediction model based on the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model training unit is used to train the constructed model and adjust module parameters.

[0064] A networking strategy model building module, wherein the networking strategy model building module is data-connected with the prediction model building module, and based on the energy consumption data prediction model output by the prediction model building module, generates an IoT adaptive networking strategy in the energy optimization direction according to the predicted energy consumption result;

[0065] The adaptive networking module is connected to the prediction model building module and the networking strategy model building module. By real-time monitoring of the operating status and data of the Internet of Things devices, the monitored Internet of Things related data is input into the prediction model building module, and the networking of the Internet of Things devices is adjusted according to the networking strategy provided by the networking strategy generation module. The specific implementation means include adjusting the connection mode to achieve optimized networking, thereby reducing energy consumption.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. The present invention acquires the business characteristic data and equipment energy consumption data of different business forms of the Internet of Things in real time, and establishes an energy consumption data prediction model. When the business form demand changes, the business form characteristic data can be quickly input into the model for prediction, thereby achieving a rapid response to the business form energy analysis. When the business form demand changes, the business form characteristic data can be quickly input into the prediction model to obtain subsequent predicted energy consumption data; the time lag of the system response is greatly reduced. Through this timely prediction mechanism, the adverse effects caused by the system lag are effectively solved, and the normal use of related Internet of Things equipment is guaranteed.

[0068] 2. The present invention establishes an IoT adaptive networking strategy generation model and generates a networking strategy based on accurate energy consumption data prediction results. The model fully considers the business characteristic data and energy consumption conditions under different business formats, and can timely adapt to the energy demand fluctuations in the actual operation of the equipment, thereby achieving accuracy and efficiency in energy optimization, meeting the energy optimization needs of IoT adaptive networking under various complex business formats, and solving the problems of inaccurate and inefficient energy optimization of traditional analysis methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is a flow chart of the steps of the method of the present invention;

[0070] Figure 2 This is a schematic diagram of a specific process for performing calculations in step S22 of the present invention;

[0071] Figure 3 It is a schematic diagram of the specific process of step S23 of the present invention;

[0072] Figure 4 It is a flowchart of the system work flow of the present invention. DETAILED DESCRIPTION

[0073] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0074] The present invention provides an Internet of Things adaptive networking method for business energy analysis, and the specific steps are as follows:

[0075] S1: Obtaining energy information of IoT business forms, using sensor networks to obtain business form characteristic data of different IoT business forms, as well as energy consumption data of IoT devices under corresponding business forms, where energy consumption data includes IoT device self-consumption data and energy transmission consumption data;

[0076] S2: Establish an energy consumption data prediction model, which can predict energy consumption based on the acquired business characteristic data under different business formats and the energy consumption data of IoT devices under the corresponding business formats, and generate energy consumption prediction results;

[0077] S3: Establish an IoT adaptive networking strategy generation model, which can generate an IoT adaptive networking strategy in the energy optimization direction according to the energy consumption prediction results of the energy consumption data prediction model;

[0078] S4: Adaptive networking optimization of the Internet of Things. When the business demand changes, the changed business characteristic data is input into the energy consumption data prediction model to obtain the subsequent predicted energy consumption data under the corresponding business demand. The predicted energy consumption data is input into the adaptive networking strategy generation model of the Internet of Things to generate the corresponding networking strategy. Then the Internet of Things adjusts the Internet of Things according to the newly generated networking strategy to cope with the changes in business demand.

[0079] An Internet of Things adaptive networking system for business energy analysis, comprising:

[0080] A business data collection module, which uses a sensor network to obtain business characteristic data under different business forms of the Internet of Things, as well as energy consumption data of Internet of Things devices under corresponding business forms, including consumption data of the Internet of Things devices themselves and energy transmission consumption data;

[0081] A prediction model building module, wherein the prediction model is data-connected to the business format data collection module, and includes a data preprocessing unit, a model building unit, and a model training unit. The data preprocessing unit cleans and denoises the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model building unit establishes an energy consumption data prediction model based on the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model training unit is used to train the constructed model and adjust module parameters.

[0082] A networking strategy model building module, wherein the networking strategy model building module is data-connected with the prediction model building module, and based on the energy consumption data prediction model output by the prediction model building module, generates an IoT adaptive networking strategy in the energy optimization direction according to the predicted energy consumption result;

[0083] The adaptive networking module is connected to the prediction model building module and the networking strategy model building module. By real-time monitoring of the operating status and data of the Internet of Things devices, the monitored Internet of Things related data is input into the prediction model building module, and the networking of the Internet of Things devices is adjusted according to the networking strategy provided by the networking strategy generation module. The specific implementation means include adjusting the connection mode to achieve optimized networking, thereby reducing energy consumption.

[0084] Embodiment 1:

[0085] like Figure 1-4 As shown, in this embodiment, an IoT system of a smart factory is provided. There are various sensors and equipment in the production area. Different production needs represent different business formats. An IoT networking strategy is required to optimize energy, maintain stable operation of the factory and reduce energy loss.

[0086] S1, obtain the energy information of the IoT business form, and use the sensor network to obtain the business form characteristic data of different IoT business forms. For example, the operating status data of IoT devices under different production conditions;

[0087] Obtain energy consumption data of IoT devices under corresponding business formats; energy consumption data includes IoT device self-consumption data and energy transmission consumption data;

[0088] The acquired data is classified into business type characteristic data sets according to different business types, and each business type characteristic data set includes corresponding business type characteristic data and energy consumption data of IoT devices under the corresponding business type;

[0089] The data were preprocessed by using a median filter algorithm based on a sliding window to remove noise, and a box plot method to remove outliers.

[0090] S2, establishing a prediction model, and establishing an energy consumption data prediction model based on the acquired business characteristic data under different business formats and the energy consumption data of the IoT devices under the corresponding business formats;

[0091] First, data is preset, and the collected energy consumption data of IoT devices are combined and arranged in time series. Let t represent a certain moment, and all collected t constitute a set (t). i and j represent different IoT devices, and i×j means that IoT device i is connected to IoT device j. The energy consumption generated by each i×j connection is predicted and calculated.

[0092] Model building, extracting and inputting feature vectors according to time series;

[0093] The following calculations are performed at different times t: information filtering, information updating and memory warehouse update values ​​are calculated by inputting the feature vector and the predicted latent number of t; then the memory warehouse is updated and information is extracted to obtain the predicted latent number at the corresponding time, and finally the predicted latent number is used to predict the energy consumption data of t+1;

[0094] Randomly initialize the data, weight W, bias term b, prediction weight ω and prediction bias β; the initial value C of the prediction latent number h and memory warehouse is initialized to 0 when the data is first calculated;

[0095] Information filtering calculation, information filtering determines which information can be deleted from the memory warehouse:

[0096]

[0097] in, Represents the dot product operation of vectors, f t Indicates the input information, X t Determine whether to filter the result; t-1is the predicted hidden number at time t-1, and is the information transmission carrier for predicting the energy value of the business state at time t. W and b represent the weight and bias terms respectively. σ represents the logical function, and tanh is the hyperbolic tangent function. t and the previous moment prediction implicit number h t-1 The processing is used to determine the degree of information forgetting in the cell state at the previous moment, so as to achieve dynamic adjustment of memory;

[0098] Information update calculation and memory warehouse update value calculation:

[0099]

[0100] Information updating determines what new information is stored in the memory warehouse. First, i t Based on the current input X t and the previous moment prediction implicit number h t-1 By weight W i And the logic function σ determines which of the new input features are the energy information, Update values ​​of the memory warehouse to provide a data basis for subsequent updates of the memory warehouse;

[0101] Update the memory repository:

[0102]

[0103] Information extraction calculation:

[0104]

[0105] h t =o t ×tanh(C t )

[0106] Among them, C t is the memory store at time t, C t Understood as a vector, it contains information or elements of many dimensions. t The weights, biases, and memory warehouses are used to determine which information is currently needed, that is, to predict the latent number h t , predict the latent number h t for o t and memory warehouse C t The result of multiplication after the hyperbolic tangent function, which means o t Will be in the memory warehouse C t Each element of is scaled if o t The smaller it is, the more important the corresponding element in the memory warehouse is in predicting the latent number h t The contribution will be smaller; by judging the size of the contribution, we can determine which information is currently needed. The larger the contribution, the easier it is to extract.

[0107] Repeat step S221 to step S224 until all the t in the set (t) are calculated;

[0108] Then use the predicted latent number h t Conduct business energy forecast:

[0109]

[0110] Where E represents the predicted energy consumption data, That is, the energy consumption data of ij predicted at t+1, X t is the business characteristic data input at time t; ω and β are the prediction weight and prediction bias used in the prediction calculation.

[0111] Define the loss function and choose the mean square error or mean absolute error loss function to measure the difference between the predicted value and the actual value;

[0112] Decide on the stochastic gradient descent algorithm as the optimization algorithm;

[0113] Divide the data set, and divide the actual value data into training data sets according to the stratified sampling method;

[0114] Model training, multiple iterations of training are performed on the training data set. In each iteration, the predicted value of energy consumption data is calculated through forward propagation, and then the loss is calculated according to the loss function. The gradient is calculated through the back-propagation algorithm, and the stochastic gradient descent algorithm is used to update the model parameters;

[0115] Parameter adjustment: The validation set is used to adjust the parameters of the model. A grid search or random search method is used to find the parameter combination that minimizes the model results. The parameters include weights, bias terms, prediction weights, and prediction biases. Confirm that the data with the smallest loss function value is selected as the model parameter.

[0116] By establishing a prediction model, we can predict the energy consumption of different equipment in the future based on historical data and current business characteristic data. In this way, when production demand changes, the changed business characteristic data can be input into the prediction model to obtain predicted data on the energy consumption data changing over time under the changed business conditions.

[0117] It provides a basis for adjusting the networking strategy of IoT devices in advance, avoiding energy waste and unstable production caused by delayed system response. For example, when a factory receives a new order and production demand is about to change, the prediction model can predict the energy consumption trend of the equipment in advance, so as to adjust the networking strategy in time to ensure smooth production.

[0118] S3: Establish an IoT adaptive networking strategy generation model, which can generate IoT adaptive networking strategies in the energy optimization direction according to the energy consumption prediction results of the prediction model;

[0119] set up It is represented as the predicted energy consumption data at time t when i is connected to j. At the same time, the connection mode between i and j is set to k, and R k is the indicator function. When i and j are connected using connection method k, R k is 1, otherwise it is 0;

[0120] Construct device connection relationships. Based on the functional characteristics and spatial location relationships of the devices, a series of device connection relationships are constructed and their total energy consumption is calculated, considering only the energy consumption between the actually connected devices and the energy consumption of the transmission method used;

[0121] For each device connection combination i×j, the total energy consumption E 总 for:

[0122]

[0123] Where p is the total number of IoT devices in the connection combination, and q is the total number of connection modes k;

[0124] Connection stability constraint: For any k, if we adjust from connection mode k1 to connection combination k . , only accept the adjustment if the connection is more stable after the adjustment, that is:

[0125]

[0126] Among them, μ represents the stability of the connection mode, which is determined by the specific connection mode;

[0127] Gradually optimize the connection relationship and compare the total energy consumption values ​​of different assumed connection combinations. After each combination, recalculate the total energy consumption. If the new connection combination reduces the total energy consumption, accept the adjustment; otherwise, continue to try other adjustments.

[0128] Establish an adaptive networking strategy generation model, and find the optimal networking solution by continuously optimizing the device connection relationship. Compare the total energy consumption values ​​of different hypothetical connection combinations, and recalculate the total energy consumption after each adjustment to ensure that the new connection combination can reduce energy consumption. This step-by-step optimization method can minimize the energy consumption of the system and improve energy utilization efficiency without affecting production. At the same time, the networking strategy is generated based on the results of the prediction model, which can be dynamically adjusted according to different business needs to ensure that the system is always in the optimal energy consumption state.

[0129] Embodiment 2:

[0130] like Figure 1-4 As shown, in this embodiment, the IoT system of the smart factory in the first embodiment generates business demand due to market changes, and it is necessary to adaptively network the IoT devices under different business formats in the factory to optimize energy, continuously maintain stable operation of the factory and reduce energy loss;

[0131] For example, due to market changes, the scale of production needs to be reduced, and some large equipment may not be adequately utilized for production, so it is necessary to change the transmission method or adjust the equipment layout to reduce energy loss.

[0132] In the first embodiment, a complete prediction model has been established based on historical data, so:

[0133] S41: When the business demand changes, determine the changed business demand. For example, in a smart factory, it may change from large-scale standardized production to small-batch customized production.

[0134] S42: inputting the determined business demand into the long-term and short-term prediction model in real time, and obtaining the subsequent predicted energy consumption data under the corresponding business demand through the long-term and short-term prediction model, and inputting the determined business characteristic data into the long-term and short-term prediction model. The model predicts the subsequent energy consumption data under this business demand based on historical data and current business demand. For example, in the small batch customized production of smart factories, the energy consumption of different devices under different connection modes is predicted;

[0135] S43: Generate a corresponding networking strategy based on the predicted energy consumption data;

[0136] S44: The Internet of Things automatically adjusts the connection relationship between the Internet of Things devices according to the generated networking strategy, and each Internet of Things device communicates or connects according to the new strategy to achieve adaptive networking. For example, in a smart factory, the connection of certain devices is closed or adjusted according to the networking strategy.

[0137] This method can promote better collaboration between different IoT devices. For example, on the production line of a smart factory, after the business format changes, when it is predicted that the energy consumption of equipment in a certain link may increase, the working state of the equipment connected to it can be adjusted to make the energy distribution of the entire production line more reasonable, which not only reduces the overall energy consumption, but also improves production efficiency.

[0138] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An Internet of Things adaptive networking method for business energy analysis, characterized in that: The specific steps are as follows: S1: Obtaining energy information of IoT business forms, using sensor networks to obtain business form characteristic data of different IoT business forms, as well as energy consumption data of IoT devices under corresponding business forms, where energy consumption data includes IoT device self-consumption data and energy transmission consumption data; S2: Establish an energy consumption data prediction model, which can predict energy consumption based on the acquired business characteristic data under different business formats and the energy consumption data of IoT devices under the corresponding business formats, and generate energy consumption prediction results; S3: Establish an IoT adaptive networking strategy generation model, which can generate an IoT adaptive networking strategy in the energy optimization direction according to the energy consumption prediction results of the energy consumption data prediction model; S4: Adaptive networking optimization of the Internet of Things. When the business demand changes, the changed business characteristic data is input into the energy consumption data prediction model to obtain the subsequent predicted energy consumption data under the corresponding business demand. The predicted energy consumption data is input into the adaptive networking strategy generation model of the Internet of Things to generate the corresponding networking strategy. Then the Internet of Things adjusts the Internet of Things according to the newly generated networking strategy to cope with the changes in business demand.

2. The method for adaptive networking of the Internet of Things for business energy analysis according to claim 1, characterized in that: In S1, the specific steps of obtaining the IoT business energy information are as follows: S11: Use sensor networks to obtain business characteristic data under different business forms of the Internet of Things; S12: Obtain energy consumption data of IoT devices corresponding to different business forms of IoT; S13: Classify the acquired data into business type characteristic data sets according to different business types, each business type characteristic data set includes corresponding business type characteristic data and energy consumption data of IoT devices under the corresponding business type.

3. The method for adaptive networking of the Internet of Things for business energy analysis according to claim 1, characterized in that: In S2, the specific steps of establishing the energy consumption data prediction model are as follows: S21: Data presetting, the collected energy consumption data of IoT devices are combined and arranged in time series, assuming that t represents a certain moment, and the t corresponding to each piece of energy consumption data of IoT devices collected constitutes a set (t), i and j represent different IoT devices, and i×j represents that IoT device i is connected to IoT device j; the energy consumption generated by each i×j connection is predicted and calculated; S22: Model construction, extract and input corresponding energy consumption data according to time series: S221: Data initialization, randomly initialize weight W, bias term b, prediction weight ω and prediction bias β; predict the initial value C of latent number h and memory warehouse, which are initialized to 0 when the data is first calculated; S222: Information filtering calculation, information filtering determines which information can be deleted from the memory warehouse: in, Represents the dot product operation of vectors, f t Indicates the input information, X t Determine whether to filter the result; t-1 is the predicted hidden number at time t-1, and is the information transmission carrier for predicting the energy value of the business state at time t. W and b represent the weight and bias terms respectively. σ represents the logical function, and tanh is the hyperbolic tangent function. t and the previous moment prediction implicit number h t-1 The processing is used to determine the degree of information forgetting in the cell state at the previous moment, so as to achieve dynamic adjustment of memory; S223: Information update calculation and memory warehouse update value calculation: Information updating determines what new information is stored in the memory warehouse. First, i t Based on the current input X t and the previous moment prediction implicit number h t-1 By weight W i And the logic function σ determines which of the new input features are important energy information, Update values ​​of the memory warehouse to provide a data basis for subsequent updates of the memory warehouse; S224: Update memory warehouse: S225: Information extraction calculation: h t =o t ×tanh(C t ) Among them, C t is the memory store at time t, C t Understood as a vector, it contains information or elements of many dimensions. t The weights, biases, and memory warehouses are used to determine which information is currently needed, that is, to predict the latent number h t , predict the latent number h t for o t and memory warehouse C t The result of multiplication after the hyperbolic tangent function, which means o t Will be in the memory warehouse C t Each element of is scaled if o t The smaller it is, the more important the corresponding element in the memory warehouse is in predicting the latent number h t The contribution will be smaller; by judging the size of the contribution, we can determine which information is currently needed. The larger the contribution, the easier it is to extract. S226: Repeat S221-S224 until all t in the set (t) are calculated; S227: Finally use the predicted hidden number h t Conduct business energy forecast: Where E represents the predicted energy consumption data, That is, the energy consumption data of ij predicted at t+1, X t is the business characteristic data input at time t; ω and β are the prediction weight and prediction bias used in the prediction calculation; S23: Model optimization and adjustment, determine the loss function to measure the difference between the predicted value and the actual value, and select the stochastic gradient descent algorithm to update the model parameters.

4. The method for adaptive networking of the Internet of Things for business energy analysis according to claim 3, characterized in that: In S23, the specific process of model optimization and adjustment is as follows: S231: Define the loss function and select the mean square error or mean absolute error loss function to measure the difference between the predicted value and the actual value; S232: Determine a stochastic gradient descent algorithm as an optimization algorithm; S233: Divide the data set, dividing the actual value data into a training data set according to a stratified sampling method; S234: Model training, performing multiple iterations of training on the training data set. In each iteration, the predicted value of the energy consumption data is calculated by forward propagation, and then the loss is calculated according to the loss function, and then the gradient is calculated by the back propagation algorithm, and the parameters of the model are updated using the stochastic gradient descent algorithm; S235: Parameter adjustment. The validation set is used to adjust the parameters of the model. A grid search or random search method is used to find the parameter combination that minimizes the model result. The parameters include weights, bias terms, prediction weights, and prediction biases. Confirm that the data with the smallest loss function value is selected as the model parameter.

5. The method for adaptive networking of the Internet of Things for business energy analysis according to claim 3, characterized in that: In S3, the steps of establishing the IoT adaptive networking strategy generation model are as follows: S31: It is represented as the predicted energy consumption data at time t when i and j are connected. At the same time, the connection mode between i and j is set to k, and R k is the indicator function. When i and j are connected using connection method k, R k is 1, otherwise it is 0; S32: constructing a device connection relationship, based on the functional characteristics and spatial position relationship of the devices, constructing a series of device connection relationships, and calculating their total energy consumption, considering only the energy consumption between the actually connected devices and the energy consumption of the used transmission method; For each device connection combination i×j, the total energy consumption E 总 for: Where p is the total number of IoT devices in the connection combination, and q is the total number of connection modes k; Connection stability constraint: For any k, if we adjust from connection mode k1 to connection combination k . , only accept the adjustment if the connection is more stable after the adjustment, that is: Among them, μ represents the stability of the connection mode, which is determined by the specific connection mode; S33: gradually optimize the connection relationship, compare the total energy consumption values ​​of different assumed connection combinations, recalculate the total energy consumption after each combination, and accept the adjustment if the new connection combination reduces the total energy consumption; otherwise, continue to try other adjustments.

6. The method for adaptive networking of the Internet of Things for business energy analysis according to claim 1, characterized in that: In S4, the specific steps of the IoT adaptive networking optimization are as follows: S41: When the business demand changes, determine the changed business demand; S42: inputting the determined business demand into the energy consumption data prediction model in real time, and obtaining subsequent predicted energy consumption data under the corresponding business demand through the long-term and short-term prediction model; S43: inputting the predicted energy consumption data into the IoT adaptive networking strategy generation model to generate a corresponding networking strategy; S44: The Internet of Things automatically adjusts the connection relationship between the Internet of Things devices according to the generated networking strategy, and each Internet of Things device communicates or connects according to the new strategy to achieve adaptive networking.

7. An Internet of Things adaptive networking system for business energy analysis, characterized in that: include: A business data collection module, which uses a sensor network to obtain business characteristic data under different business forms of the Internet of Things, as well as energy consumption data of Internet of Things devices under corresponding business forms, including consumption data of the Internet of Things devices themselves and energy transmission consumption data; A prediction model building module, wherein the prediction model is data-connected to the business format data collection module, and includes a data preprocessing unit, a model building unit, and a model training unit. The data preprocessing unit cleans and denoises the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model building unit establishes an energy consumption data prediction model based on the business format characteristic data and energy consumption data under different business formats provided by the business format data collection module. The model training unit is used to train the constructed model and adjust module parameters. A networking strategy model building module, wherein the networking strategy model building module is data-connected with the prediction model building module, and based on the energy consumption data prediction model output by the prediction model building module, generates an IoT adaptive networking strategy in the energy optimization direction according to the predicted energy consumption result; The adaptive networking module is connected to the prediction model building module and the networking strategy model building module. By real-time monitoring of the operating status and data of the Internet of Things devices, the monitored Internet of Things related data is input into the prediction model building module, and the networking of the Internet of Things devices is adjusted according to the networking strategy provided by the networking strategy generation module. The specific implementation means include adjusting the connection mode to achieve optimized networking, thereby reducing energy consumption.

Images