Distributed energy management method and system based on Internet of Things

By using high-precision sensors and smart meters in the energy management system to collect data, build energy consumption prediction models and optimize energy distribution strategies, the problem of insufficient intelligence of data quality and strategy is solved, efficient energy management and dynamic scheduling are achieved, and improved The overall performance and adaptability of the system.

CN118915503AInactive Publication Date: 2024-11-08JIAXING SHUKAILAI DIGITAL TECH CO LTD

Patent Information

Application Number
CN202410852370.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-28
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology faces the problems of low data quality, unreasonable feature selection and model construction, and insufficient intelligence of energy distribution strategies in energy management, which limits the overall efficiency of the system and its adaptability to dynamic energy consumption requirements.

Method used

By installing high-precision sensors and smart meters, environmental parameters and equipment status data are collected, data processing and feature extraction are used for central control platform, energy consumption prediction models are built, and energy allocation strategies are optimized through loss functions to achieve fine management and dynamic scheduling.

Benefits of technology

It significantly improves the accuracy and stability of energy consumption prediction, ensures the refined level and response speed of energy management, promotes the rational allocation of resources, energy conservation and emission reduction, and realizes the intelligent and adaptive learning mechanism of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915503A_ABST
    Figure CN118915503A_ABST
Patent Text Reader

Abstract

The invention discloses a distributed energy management method and system based on the Internet of Things, and relates to the technical field of energy management of the Internet of Things, and the method comprises the steps: collecting environment parameter data and equipment state data through installing a high-precision sensor and an intelligent electric meter, and obtaining an original data set; processing the original data set through the central control platform to obtain a basic data set, and extracting features from the basic data set to obtain feature vectors; establishing and training an energy consumption prediction model, and inputting the feature vector into the energy consumption prediction model to obtain an energy consumption prediction value; calculating and judging the accuracy of the energy consumption prediction model through a loss function; according to the energy consumption prediction system and method, the high-precision sensor and the intelligent electric meter are integrated, environment and equipment data are accurately collected, key features are processed and extracted through the central control platform, the energy consumption prediction model is constructed and trained, and efficient prediction of energy consumption is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of energy management of the Internet of Things, and in particular to a distributed energy management method and system based on the Internet of Things. Background Art

[0002] In contemporary society, with the deep integration of information technology and intelligent technology, energy management systems are undergoing a key transformation towards intelligence. The rapid development of Internet of Things technology has greatly promoted the intelligence level of distributed energy management, and provided new ways for efficient energy management and optimal allocation through real-time and accurate data collection and analysis. However, existing technologies still face challenges such as low data quality, unreasonable feature selection and model construction, and insufficient intelligence of energy allocation strategies, which limit the overall performance of the system and its adaptability to dynamic energy consumption demands.

[0003] The present invention designs a distributed energy management method and system based on the Internet of Things. While overcoming the defects of the prior art, it ensures the high quality of data and the accuracy of analysis by integrating high-precision sensors and smart meters and combining the precise data processing and feature extraction of the central control platform. The highlight of the invention is that it constructs an energy consumption prediction model including feature interaction terms, deeply mines the relationship between environmental parameters and equipment status, and significantly improves the accuracy and stability of the prediction. At the same time, with the help of real-time data processing and energy allocation strategies guided by loss functions, it realizes fine management and dynamic scheduling of energy consumption, effectively dealing with the problems of inaccurate predictions and rigid strategies of traditional systems, marking a major progress in the field of intelligent energy management. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a distributed energy management method and system based on the Internet of Things to solve the challenges faced by the existing technology, such as low data quality, unreasonable feature selection and model construction, and insufficient intelligence of energy allocation strategy, which limits the overall performance of the system and its adaptability to dynamic energy consumption demands.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a distributed energy management method based on the Internet of Things, which includes collecting environmental parameter data and device status data to obtain an original data set by installing high-precision sensors and installing smart meters;

[0008] The original data set is processed through the central control platform to obtain a basic data set, and features are extracted from the basic data set to obtain a feature vector;

[0009] Establish and train an energy consumption prediction model, and input the feature vector into the energy consumption prediction model to obtain the energy consumption prediction value;

[0010] The accuracy of the energy consumption prediction model is determined by loss function calculation;

[0011] Input real-time data into the energy consumption prediction model and manage energy allocation based on the real-time energy consumption prediction value.

[0012] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, wherein: the high-precision sensor includes a light intensity sensor, a wind speed sensor, a sound sensor and a temperature and humidity sensor;

[0013] The environmental parameter data include light intensity, wind speed and direction, temperature, humidity and environmental noise level;

[0014] The equipment status data includes energy output power, equipment operating temperature, pressure, energy, current intensity, voltage and frequency.

[0015] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, the steps of processing the original data set are as follows:

[0016] The original data set is input into the data quality scoring function, and the data quality scoring function expression is:

[0017] Q i =w1×I comp (i)+w2×I range (i)+w3×I consistency (i)

[0018] Among them, Q represents the original data set, Q i represents the quality score of the i-th data record, w1, w2, and w3 are the weight coefficients of the three indicators of completeness, scope rationality, and consistency, respectively. comp is the integrity index, I range is the scope rationality index, I consistency is the consistency indicator;

[0019] Set the quality score threshold Q i ≥U;

[0020] The original data set is input into the data quality scoring function through the central control platform, and only the data with a score greater than U is retained to obtain the basic data set.

[0021] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, the steps of extracting features from the basic data set to obtain feature vectors are as follows:

[0022] Select features that have a direct impact on energy consumption prediction from the basic data set, including current intensity, voltage, temperature, humidity, frequency, pressure, and energy, and process them to form a feature vector. The feature vector expression is:

[0023]

[0024] Among them, F t represents the processed feature vector, t represents the time point, I, V, T, RH, N, P, and E represent the current intensity, voltage, temperature, humidity, frequency, pressure, and energy respectively, and I t 、V t , T t RH t 、N t , P t 、E t Represents the values ​​of the seven features at time point t, It means that the voltage is multiplied by the negative coefficient α and then the exponential operation is performed, sin(βT t ) represents the temperature multiplied by the coefficient β through the sine function conversion, is a normalization operation. is an inverse proportional function, Represents the pressure to the power of ∈, which can amplify and reduce the impact of pressure.

[0025] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, wherein: the energy consumption prediction model is established and trained, and the characteristic vector is input into the energy consumption prediction model to obtain the energy consumption prediction value, and the specific steps are:

[0026] For the eigenvector F t Each dimension in is normalized to obtain the normalized feature vector

[0027] The output layer of the energy consumption prediction model is transformed into a fully connected layer through nonlinear conversion, and the predicted energy consumption value is output. The expression is:

[0028]

[0029] in, represents the predicted energy consumption value at time t, σ represents the activation function of the output layer, that is, the Sigmoid function, and FC is the fully connected layer.

[0030] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, the accuracy of the energy consumption prediction model is determined by calculating the loss function, and the steps are as follows:

[0031] Input the actual energy consumption value and the energy consumption forecast value into the energy consumption forecast model;

[0032] Using the loss function MSE, the actual energy consumption value of the energy consumption prediction model is compared with the energy consumption prediction value. The expression is:

[0033]

[0034] Among them, E t represents the actual energy consumption value, obtained from actual observation data, Represents the predicted value of energy consumption, n represents the number of training samples, and MSE is a non-negative value.

[0035] As a preferred solution of the distributed energy management method based on the Internet of Things described in the present invention, the steps of inputting real-time data into the energy consumption prediction model are as follows:

[0036] Collect real-time data to form a real-time original data set, and filter the real-time original data set to obtain a high-quality real-time basic data set;

[0037] The real-time feature vector is extracted from the real-time basic data set, normalized, and then input into the energy consumption prediction model to obtain the real-time predicted energy consumption value P E,t ;

[0038] The energy allocation management according to the real-time energy consumption prediction value comprises the following steps:

[0039] Set the historical average energy consumption time to M;

[0040] When P E,t ≤MQ, indicating that energy demand is reduced, energy-saving mode is activated, and excess energy is transferred to high-demand areas;

[0041] When MQ <P E,t ≤MW, indicating that energy demand is stable and the current energy allocation strategy is maintained to ensure supply and demand balance;

[0042] When MW <P E,t ≤ME, indicating that the energy consumption value is slightly higher than the daily level, and the energy supply to the relevant areas should be increased in advance to avoid shortages during peak hours;

[0043] When P E,t >ME, indicating that the predicted energy consumption is significantly higher than the normal level, the emergency energy allocation plan is immediately activated to significantly increase energy supply to ensure that key areas and services are not affected.

[0044] In a second aspect, the present invention provides a distributed energy management system based on the Internet of Things, including a data acquisition module, a data preprocessing module, a feature engineering module, an energy consumption prediction model module, a model optimization module and a management decision module;

[0045] The data acquisition module is used to obtain data collected by high-precision sensors and smart meters;

[0046] The data preprocessing module is used to receive raw data and perform preprocessing;

[0047] The feature engineering module is used to extract features from the basic data set;

[0048] The energy consumption prediction model module is used to run the energy consumption prediction model;

[0049] The model optimization module is used to input the feature vector into the prediction model, obtain the energy consumption prediction value, evaluate the model performance using the mean square error (MSE), and use the particle swarm optimization (PSO) advanced optimization algorithm to tune the model parameters and continuously improve the prediction accuracy;

[0050] The management decision module dynamically adjusts the energy allocation strategy according to real-time data input and energy consumption forecast values. The module can take different management measures according to the predicted energy consumption trend.

[0051] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the distributed energy management method based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the distributed energy management method based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0053] The beneficial effects of the present invention are as follows: the present invention integrates high-precision sensors and smart meters to accurately collect environmental and equipment data, processes and extracts key features through a central control platform, constructs and trains an energy consumption prediction model, and realizes efficient prediction of energy consumption. By comparing the prediction with actual energy consumption in real time, the model is optimized to ensure prediction accuracy, and then energy is intelligently allocated according to the prediction results to achieve a balance between supply and demand. This not only greatly improves the level of refinement and response speed of energy management, but also promotes the rational allocation of resources and energy conservation and emission reduction, and realizes the intelligence and adaptive learning mechanism of the entire chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0055] Figure 1 This is a flow chart of the distributed energy management method based on the Internet of Things in Example 1. DETAILED DESCRIPTION

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0057] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0059] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a distributed energy management method based on the Internet of Things, comprising the following steps:

[0060] By installing high-precision sensors and smart meters, environmental parameter data and equipment status data are collected to obtain the original data set;

[0061] The original data set is processed through the central control platform to obtain a basic data set, and features are extracted from the basic data set to obtain a feature vector;

[0062] Establish and train an energy consumption prediction model, and input the feature vector into the energy consumption prediction model to obtain the energy consumption prediction value;

[0063] The accuracy of the energy consumption prediction model is determined by loss function calculation;

[0064] Input real-time data into the energy consumption prediction model and manage energy allocation based on the real-time energy consumption prediction value.

[0065] High-precision sensors include light intensity sensors, wind speed sensors, sound sensors, and temperature and humidity sensors;

[0066] Environmental parameter data include light intensity, wind speed and direction, temperature, humidity and ambient noise level;

[0067] Equipment status data includes energy output power, equipment operating temperature, pressure, energy, amperage, voltage and frequency.

[0068] The steps for processing the original data set are as follows:

[0069] The original data set is input into the data quality scoring function, and the data quality scoring function expression is:

[0070] Q i =w1×I comp (i)+w2×I range (i)+w3×I consistency (i)

[0071] Among them, Q represents the original data set, Q i represents the quality score of the i-th data record, w1, w2, and w3 are the weight coefficients of the three indicators of completeness, scope rationality, and consistency, respectively. comp is the completeness index, which is used to measure the completeness of the i-th data. If there is no missing data record, then I comp (i)=1, data record is missing, then I comp (i) = 0, I range is a range rationality indicator, which evaluates whether the data value falls within the expected reasonable range. If the value of data point i is within this reasonable range, then I range (i) = 1, otherwise 0, I consistency It is a consistency indicator used to check the logical consistency between data. If the change range is within a reasonable range, the data is consistent. consistency (i)=1, the variation range is not within a reasonable range, and the data are inconsistent. consistency (i) = 0;

[0072] Set the quality score threshold Q i ≥U;

[0073] The original data set is input into the data quality scoring function through the central control platform, and only the data with a score greater than U is retained to obtain the basic data set;

[0074] The central control platform includes,the sensor that automatically collects data once a minute with a timestamp,and sends it to the central control platform through an encrypted channel.

[0075] Extract features from the basic data set to obtain feature vectors. The steps are as follows:

[0076] Select features that have a direct impact on energy consumption prediction from the basic data set, including current intensity, voltage, temperature, humidity, frequency, pressure, and energy, and process them to form a feature vector. The feature vector expression is:

[0077]

[0078] Among them, F t represents the processed feature vector, t represents the time point, I, V, T, RH, N, P, and E represent the current intensity, voltage, temperature, humidity, frequency, pressure, and energy respectively, and I t 、V t 、T t RH t 、N t , P t 、E t Represents the values ​​of the seven features at time point t, It means that the voltage is multiplied by the negative coefficient α and then the exponential operation is performed, sin(βT t ) represents the temperature multiplied by the coefficient β through the sine function conversion, It is a normalization operation that avoids the possibility of division by zero by adding a constant γ, ensuring that all values ​​are between 0 and 1. is an inverse proportional function, Represents the pressure to the power of ∈, which can amplify and reduce the impact of pressure.

[0079] Establish an energy consumption prediction model and train it, input the feature vector into the energy consumption prediction model to obtain the energy consumption prediction value. The specific steps are as follows:

[0080] For the eigenvector F t Each dimension in is normalized to obtain the normalized feature vector Make the feature mean 0 and variance 1;

[0081] The attention mechanism is used to assign weights to each feature, and the expression is:

[0082]

[0083] in, represents the weighted eigenvector, n represents the dimension of the eigenvector, It means the accumulation from index i=1 to index n. represents the relative importance of feature i at the current prediction time t, represents the i-th eigenvalue in the eigenvector at time point t;

[0084] Use the long short-term memory network (LSTM) of deep learning to build an energy consumption prediction model;

[0085] The energy consumption prediction model structure is set to include several LSTM units, each of which contains an input gate i t 、Forget gate t, cell state c t and output gate o t , the expression is:

[0086]

[0087] h t =o t ⊙tanh(c t )

[0088] Among them, c t-1 represents the state of the memory cell at the previous time step, ⊙ represents element-wise multiplication, that is, multiplying two vectors of the same dimension, β t is the adjustment factor used to adjust the input feature vector of the current time step tanh represents the hyperbolic tangent function, h t is the hidden state of the current time step;

[0089] Stack multiple layers of LSTM and receive the output h of the previous layer at each layer t As input, and introducing residual connections between adjacent layers, the expression is:

[0090]

[0091] in, Represents the hidden state of the lth layer, LSTM l represents the operation of the l-th layer LSTM unit, Represents the hidden state of the previous layer;

[0092] The output layer of the energy consumption prediction model is a fully connected layer that undergoes nonlinear transformation to output the predicted energy consumption value. The expression is:

[0093]

[0094] in, represents the predicted energy consumption value at time t, σ represents the activation function of the output layer, that is, the Sigmoid function, and FC is the fully connected layer.

[0095] The accuracy of the energy consumption prediction model is determined by loss function calculation. The steps are as follows:

[0096] Input the actual energy consumption value and the energy consumption forecast value into the energy consumption forecast model;

[0097] Using the loss function MSE, the actual energy consumption value of the energy consumption prediction model is compared with the energy consumption prediction value. The expression is:

[0098]

[0099] Among them, E t represents the actual energy consumption value, obtained from actual observation data, Represents the predicted value of energy consumption, n represents the number of training samples, MSE is a non-negative value, and its calculation formula is the average value of the square of the error. The smaller the value of MSE, the closer the predicted value of the model is to the actual value, and the better the performance of the model. On the contrary, the performance of the model is poor, and then enter the model tuning stage;

[0100] The performance of the constructed energy consumption prediction model is evaluated according to the obtained MSE value. The specific steps are as follows:

[0101] The initial MSE value is set to MSE0. When MSE0 ≥ 0.1, the model tuning phase begins.

[0102] Model tuning is to select particle swarm optimization PSO as the tuning method, including setting the particle position vector to X i =[w1,w2,...,w n ], the velocity vector is V i =[v1,v2,...,v n ], where w n represents the feature weight, n is the number of features;

[0103] The particle swarm optimization PSO expression is:

[0104] V i+1 =w·V i +c1·r1·(P i -X i )+c2·r2·(GX i )

[0105] X i+1 =X i +V i+1

[0106] Among them, w is the inertia weight, c1 and c2 are acceleration constants, r1 and r2 are random numbers between [0,1], P i is the individual best position, G is the global best position, and the particle position is updated iteratively until the parameter combination that minimizes MSE is found;

[0107] When MSE0<0.1, an accurate energy consumption prediction model is obtained.

[0108] To input real-time data into the energy consumption prediction model, the steps are as follows:

[0109] Collect real-time data to form a real-time original data set, and filter the real-time original data set to obtain a high-quality real-time basic data set;

[0110] The real-time feature vector is extracted from the real-time basic data set, normalized, and then input into the energy consumption prediction model to obtain the real-time predicted energy consumption value P E,t ;

[0111] According to the real-time energy consumption forecast value, the energy allocation management steps are as follows:

[0112] Set the historical average energy consumption time to M;

[0113] When P E,t ≤MQ, indicating that energy demand is reduced, energy-saving mode is activated, and excess energy is transferred to high-demand areas;

[0114] When MQ <P E,t ≤MW, indicating that energy demand is stable and the current energy allocation strategy is maintained to ensure supply and demand balance;

[0115] When MW <P E,t ≤ME, indicating that the energy consumption value is slightly higher than the daily level, and the energy supply to the relevant areas should be increased in advance to avoid shortages during peak hours;

[0116] When P E,t >ME, indicating that the predicted energy consumption is significantly higher than the normal level, and the emergency energy allocation plan is immediately activated to significantly increase energy supply to ensure that key areas and services are not affected;

[0117] The value of Q is 0.7;

[0118] The value of W is 1.1;

[0119] The value of E is 1.3.

[0120] This embodiment also provides a distributed energy management system based on the Internet of Things, including: a data acquisition module, a data preprocessing module, a feature engineering module, an energy consumption prediction model module, a model optimization module and a management decision module;

[0121] The data acquisition module is used to obtain data collected by high-precision sensors and smart meters;

[0122] The data preprocessing module is used to receive the original data and perform preprocessing;

[0123] The feature engineering module is used to extract features from the basic dataset;

[0124] The energy consumption prediction model module is used to run the energy consumption prediction model;

[0125] The model optimization module is used to input the feature vector into the prediction model, obtain the energy consumption prediction value, evaluate the model performance using the mean square error (MSE), and use the particle swarm optimization (PSO) advanced optimization algorithm to tune the model parameters and continuously improve the prediction accuracy.

[0126] The management decision module dynamically adjusts the energy allocation strategy based on real-time data input and energy consumption forecast values. The module can take different management measures based on the predicted energy consumption trends.

[0127] This embodiment also provides a computer device, which is suitable for the distributed energy management method based on the Internet of Things, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the distributed energy management method based on the Internet of Things proposed in the above embodiment.

[0128] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0129] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the distributed energy management method based on the Internet of Things proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0130] In summary, the present invention achieves efficient prediction of energy consumption by integrating high-precision sensors and smart meters, accurately collecting environmental and equipment data, processing and extracting key features through a central control platform, building and training an energy consumption prediction model, and optimizing the model by comparing the prediction with actual energy consumption in real time to ensure prediction accuracy, and then intelligently allocating energy according to the prediction results to achieve a balance between supply and demand. This not only greatly improves the level of refinement and response speed of energy management, but also promotes the rational allocation of resources and energy conservation and emission reduction, and realizes a full-chain intelligent and adaptive learning mechanism.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A distributed energy management method based on the Internet of Things, characterized in that: include, By installing high-precision sensors and smart meters, environmental parameter data and equipment status data are collected to obtain the original data set; The original data set is processed through the central control platform to obtain a basic data set, and features are extracted from the basic data set to obtain a feature vector; Establish and train an energy consumption prediction model, and input the feature vector into the energy consumption prediction model to obtain the energy consumption prediction value; The accuracy of the energy consumption prediction model is determined by loss function calculation; Input real-time data into the energy consumption prediction model and manage energy allocation based on the real-time energy consumption prediction value.

2. The distributed energy management method based on the Internet of Things according to claim 1, characterized in that: The high-precision sensors include light intensity sensors, wind speed sensors, sound sensors, and temperature and humidity sensors; The environmental parameter data include light intensity, wind speed and direction, temperature, humidity and environmental noise level; The equipment status data includes energy output power, equipment operating temperature, pressure, energy, current intensity, voltage and frequency.

3. The distributed energy management method based on the Internet of Things according to claim 1 or 2, characterized in that: The steps of processing the original data set are as follows: The original data set is input into the data quality scoring function, and the data quality scoring function expression is: Q i =w1×I comp (i)+w2×I range (i)+w3×I consistency (i) Among them, Q represents the original data set, Q i represents the quality score of the i-th data record, w1, w2, and w3 are the weight coefficients of the three indicators of completeness, scope rationality, and consistency, respectively. comp is the integrity index, I range is the scope rationality index, I consistency is the consistency indicator; Set the quality score threshold Q i ≥U; The original data set is input into the data quality scoring function through the central control platform, and only the data with a score greater than U is retained to obtain the basic data set.

4. The distributed energy management method based on the Internet of Things as claimed in claim 3, characterized in that: The steps of extracting features from the basic data set to obtain a feature vector are as follows: Select features that have a direct impact on energy consumption prediction from the basic data set, including current intensity, voltage, temperature, humidity, frequency, pressure, and energy, and process them to form a feature vector. The feature vector expression is: Among them, F t represents the processed feature vector, t represents the time point, I, V, T, RH, N, P, and E represent the current intensity, voltage, temperature, humidity, frequency, pressure, and energy respectively, and I t 、V t , T t RH t 、N t , P t 、E t Represents the values ​​of the seven features at time point t, It means that the voltage is multiplied by the negative coefficient α and then the exponential operation is performed, sin(βT t ) represents the temperature multiplied by the coefficient β through the sine function conversion, is a normalization operation. is an inverse proportional function, Represents the pressure to the power of ∈, which can amplify and reduce the impact of pressure.

5. The distributed energy management method based on the Internet of Things as claimed in claim 4, characterized in that: The energy consumption prediction model is established and trained, and the characteristic vector is input into the energy consumption prediction model to obtain the energy consumption prediction value. The specific steps are as follows: For the eigenvector F t Each dimension in is normalized to obtain the normalized feature vector The output layer of the energy consumption prediction model is transformed into a fully connected layer through nonlinear conversion, and the predicted energy consumption value is output. The expression is: in, represents the predicted energy consumption value at time t, σ represents the activation function of the output layer, that is, the Sigmoid function, and FC is the fully connected layer.

6. The distributed energy management method based on the Internet of Things according to claim 5, characterized in that: The accuracy of the energy consumption prediction model is determined by calculating the loss function, and the steps are as follows: Input the actual energy consumption value and the energy consumption forecast value into the energy consumption forecast model; Using the loss function MSE, the actual energy consumption value of the energy consumption prediction model is compared with the energy consumption prediction value. The expression is: Among them, E t represents the actual energy consumption value, obtained from actual observation data, Represents the predicted value of energy consumption, n represents the number of training samples, and MSE is a non-negative value.

7. The distributed energy management method based on the Internet of Things according to claim 6, characterized in that: The steps of inputting real-time data into the energy consumption prediction model are as follows: Collect real-time data to form a real-time original data set, and filter the real-time original data set to obtain a high-quality real-time basic data set; The real-time feature vector is extracted from the real-time basic data set, normalized, and then input into the energy consumption prediction model to obtain the real-time predicted energy consumption value P E,t ; The energy allocation management according to the real-time energy consumption prediction value comprises the following steps: Set the historical average energy consumption time to M; When P E,t ≤MQ, indicating that energy demand is reduced, energy-saving mode is activated, and excess energy is transferred to high-demand areas; When MQ <P E,t ≤MW, indicating that energy demand is stable and the current energy allocation strategy is maintained to ensure supply and demand balance; When MW <P E,t ≤ME, indicating that the energy consumption value is slightly higher than the daily level, and the energy supply to the relevant areas should be increased in advance to avoid shortages during peak hours; When P E,t >ME, indicating that the predicted energy consumption is significantly higher than the normal level, the emergency energy allocation plan is immediately activated to significantly increase energy supply to ensure that key areas and services are not affected.

8. A distributed energy management system based on the Internet of Things, based on the distributed energy management method based on the Internet of Things according to any one of claims 1 to 7, characterized in that: Including data acquisition module, data preprocessing module, feature engineering module, energy consumption prediction model module, model optimization module and management decision module; The data acquisition module is used to obtain data collected by high-precision sensors and smart meters; The data preprocessing module is used to receive raw data and perform preprocessing; The feature engineering module is used to extract features from the basic data set; The energy consumption prediction model module is used to run the energy consumption prediction model; The model optimization module is used to input the feature vector into the prediction model, obtain the energy consumption prediction value, evaluate the model performance using the mean square error (MSE), and use the particle swarm optimization (PSO) advanced optimization algorithm to tune the model parameters and continuously improve the prediction accuracy; The management decision module dynamically adjusts the energy allocation strategy according to real-time data input and energy consumption forecast values. The module can take different management measures according to the predicted energy consumption trend.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distributed energy management method based on the Internet of Things described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distributed energy management method based on the Internet of Things described in any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Aircraft ground air conditioner energy consumption prediction method based on IALO-AM-BiLSTM model and storage medium

    CN114065635A

  • Industrial park energy efficiency management and control method based on Internet of Things

    CN116307076A

  • Workshop scheduling method based on reinforcement learning of digital twinning and knowledge graph

    CN117669984A

  • Building energy consumption rapid prediction method

    CN118114816A

  • Method for predicting rotating stall of axial flow compressor on the basis of stacked long short-term memory network

    WO2023056614A1

Cited By

  • Carbon emission surplus and deficiency prediction method and system based on data mining

    CN119539165A

  • Biochemical material storage system based on Internet of Things

    CN119941129A

  • Smart energy consumption control method, device and equipment based on Internet of Things

    CN120178762A

  • Distributed energy resource intelligent scheduling method based on Internet of Things

    CN120373823A