A method for optimizing energy scheduling based on IoT devices

Abstract

The present application relates to the field of energy optimization technology. The present application provides an energy optimization scheduling method based on IoT devices. The technical solution steps include: real-time monitoring of the energy consumption data of each IoT device based on the control terminal of each key node; training a deep learning model using historical device energy consumption data, and outputting the future energy consumption data of the IoT device energy consumption model based on the trained deep learning model; constructing an energy consumption minimization objective function and a load balancing and power grid load optimization objective function based on the future energy consumption data of the IoT device energy consumption model; performing a multi-objective fusion operation on the energy consumption minimization objective function and the load balancing and power grid load optimization objective function to output a comprehensive evaluation function; calculating a comprehensive evaluation value based on the comprehensive evaluation function, and determining the best optimization scheduling plan by minimizing the comprehensive evaluation value. The present application improves the optimization efficiency by establishing and fusion of multi-objective functions for optimizing energy, and considering multiple factors.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy optimization, and in particular relates to an energy optimization scheduling method based on Internet of Things devices. Background Art

[0002] With the continuous development of IoT technology, more and more IoT devices are widely used in various fields, including smart homes, smart cities, industrial automation, etc. However, the energy consumption problem of IoT devices has become increasingly prominent, especially when deployed on a large scale, the control of energy consumption has become increasingly important. The high energy consumption of IoT devices not only increases costs, but may also cause certain pressure on the environment. Therefore, how to optimize the energy use of devices and reduce unnecessary energy waste is a technical problem that needs to be solved urgently.

[0003] In order to improve energy efficiency, energy scheduling based on IoT devices has become an important research direction. Although energy management and optimal scheduling of IoT devices have a certain research basis in theory, the existing technology generally only defines the minimization of energy consumption as the optimization target in energy scheduling, which has limitations, or ignores the relationship between multiple targets for multi-objective optimization targets, and fails to fully consider various factors, resulting in poor optimization effect and difficulty in achieving overall optimality. Summary of the invention

[0004] In view of the above-mentioned defects of the prior art, the present invention proposes an energy optimization scheduling method based on IoT devices. The technical solution steps designed by the present invention include:

[0005] S10: Real-time monitoring of energy consumption data of each IoT device based on the control terminal of each key node;

[0006] S20: Use historical equipment energy consumption data to train a deep learning model, and output future energy consumption data of the IoT equipment energy consumption model based on the trained deep learning model;

[0007] S30: Based on the future energy consumption data of the IoT device energy consumption model, construct the energy consumption minimization objective function and the load balancing and grid load optimization objective function;

[0008] S40: performing a multi-objective fusion operation on the energy consumption minimization objective function and the load balancing and power grid load optimization objective functions to output a comprehensive evaluation function;

[0009] S50: Calculate a comprehensive evaluation value based on the comprehensive evaluation function, and determine the best optimization scheduling plan by minimizing the comprehensive evaluation value.

[0010] Preferably, the S10 includes:

[0011] Identify and classify each IoT device, determine the type, function and priority of each IoT device, deploy control terminals at each key node, monitor the energy consumption data of the equipment in real time, pre-process the real-time monitoring data and store the pre-processed energy consumption data in the central database.

[0012] Preferably, the S20 includes:

[0013] Collect historical equipment energy consumption data and use it to train deep learning models, learn the time series characteristics and influencing factors of historical equipment energy consumption data, predict the energy consumption of each device in the next 24 hours based on the trained deep learning model, and output the future energy consumption data of the IoT equipment energy consumption model.

[0014] Preferably, the energy consumption minimization objective function in S30 is as follows:

[0015]

[0016] In the formula, is the total energy consumption, is the total number of time periods in the scheduling cycle, is the total number of IoT devices, For the Devices at time The power consumption, To indicate the Devices at time A binary variable indicating whether to run.

[0017] Preferably, the load balancing and grid load optimization objective function in S30 is as follows:

[0018]

[0019]

[0020] In the formula, For grid load fluctuations, is the average value of the grid load.

[0021] Preferably, the comprehensive evaluation function in S40 is as follows:

[0022]

[0023] In the formula, is the comprehensive evaluation function, The scaling factor for integrating the energy consumption minimization objective function and the load balancing and grid load optimization objective function is is the offset term.

[0024] Preferably, the comprehensive evaluation value is expressed as follows:

[0025]

[0026] In the formula, is the comprehensive evaluation value, To integrate the objective function of energy consumption minimization and the objective function of load balancing and grid load optimization, is the weight output by the deep learning model, Future energy consumption data for IoT device energy consumption models.

[0027] Preferably, the The calculation formula is as follows:

[0028]

[0029] In the formula, The objective function for minimizing energy consumption is The parameter characteristics and the associated weights of the load balancing and grid load optimization objective functions, The total number of parameter characteristics of the objective function for minimizing energy consumption, is the activation function;

[0030] Said The calculation formula is as follows:

[0031] .

[0032] Beneficial effects:

[0033] 1. The present invention optimizes the operating state and power consumption of IoT devices by establishing an energy consumption minimization objective function, reduces unnecessary energy waste, and preferentially operates devices in time periods with lower electricity prices in a time-sharing electricity price environment, thereby reducing energy costs;

[0034] 2. The present invention optimizes the distribution of equipment operation time, reduces the fluctuation of grid load, and ensures grid load balance by establishing load balancing and grid load optimization objective functions;

[0035] 3. The present invention integrates the two goals of minimizing energy consumption and optimizing load balance into a unified evaluation index through a comprehensive evaluation function and a comprehensive evaluation value, resolves the conflict between the goals, combines future energy consumption forecast data, dynamically adjusts the scheduling plan, and ensures the real-time and accuracy of the optimization results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0037] The embodiments of the present invention are described in detail below. The following embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation methods and specific operation processes are given, but the protection scope of the present invention is not limited to the following embodiments.

[0038] The present invention designs a method for optimizing energy scheduling based on IoT devices. Figure 1 As shown, the technical solution includes the following steps, specifically including:

[0039] S10: Real-time monitoring of energy consumption data of each IoT device based on the control terminal of each key node;

[0040] S20: Use historical equipment energy consumption data to train a deep learning model, and output future energy consumption data of the IoT equipment energy consumption model based on the trained deep learning model;

[0041] S30: Based on the future energy consumption data of the IoT device energy consumption model, construct the energy consumption minimization objective function and the load balancing and grid load optimization objective function;

[0042] S40: Perform multi-objective fusion operation on the energy consumption minimization objective function and the load balancing and power grid load optimization objective functions to output a comprehensive evaluation function;

[0043] S50: Calculate a comprehensive evaluation value based on the comprehensive evaluation function, and determine the best optimization scheduling plan by minimizing the comprehensive evaluation value.

[0044] Preferably, S10 includes:

[0045] Identify and classify each IoT device, determine the type, function and priority of each IoT device, deploy control terminals at each key node, monitor the energy consumption data of the equipment in real time, pre-process the real-time monitoring data and store the pre-processed energy consumption data in the central database.

[0046] Preferably, S20 includes:

[0047] Collect historical equipment energy consumption data and use it to train deep learning models, learn the time series characteristics and influencing factors of historical equipment energy consumption data, predict the energy consumption of each device in the next 24 hours based on the trained deep learning model, and output the future energy consumption data of the IoT equipment energy consumption model.

[0048] Preferably, the energy consumption minimization objective function in S30 is as follows:

[0049]

[0050] In the formula, is the total energy consumption, is the total number of time periods in the scheduling cycle, is the total number of IoT devices, For the Devices at time The power consumption, To indicate the Devices at time A binary variable indicating whether to run.

[0051] Preferably, the load balancing and grid load optimization objective function in S30 is as follows:

[0052]

[0053]

[0054] In the formula, For grid load fluctuations, is the average value of the grid load.

[0055] Specifically, the goal of the energy consumption minimization objective function is to minimize the total energy consumption of all devices in the scheduling period by optimizing the operating status of the equipment. Reduce unnecessary energy waste, specifically To indicate the Devices at time A binary variable indicating whether the equipment is running (1 for running, 0 for shutting down) can reduce energy costs, especially in a time-of-use electricity price environment. It is preferred to run the equipment during time periods with lower electricity prices, which provides a basis for subsequent multi-objective optimization and ensures the minimization of energy consumption. In addition, the goal of the load balancing and grid load optimization objective function is to minimize the fluctuation of grid load and ensure grid load balance. By optimizing the distribution of equipment operation time, it is possible to avoid the grid load being too high or too low in a certain period of time, improve the stability of the grid, reduce the risk of grid failure caused by load fluctuations, and further optimize the operation efficiency of the grid on the basis of minimizing energy consumption.

[0056] Preferably, the comprehensive evaluation function in S40 is as follows:

[0057]

[0058] In the formula, is the comprehensive evaluation function, The scaling factor for integrating the energy consumption minimization objective function and the load balancing and grid load optimization objective function is is the offset term.

[0059] Specifically, the goal of the comprehensive evaluation function is to integrate the two goals of minimizing energy consumption and optimizing load balance into a unified evaluation index, resolve the conflict between the goals in the multi-objective optimization problem, and find the balance point between the two. To fuse the scaling factors of the energy consumption minimization objective function and the load balancing and grid load optimization objective function, the output values ​​of the two objective functions can be mapped to the same magnitude to avoid the deviation caused by different dimensions; is an offset term, which is used to adjust the baseline value of the comprehensive evaluation function. The scaling factor and the offset term are used to eliminate the dimension difference, ensure that the output values ​​of the two objective functions are comparable at the same level, and provide input for the calculation of the comprehensive evaluation value. The function combines the output values ​​of the two objective functions into one vector. The purpose is to normalize and nonlinearly map the outputs of multiple objective functions, better integrate multiple objectives and generate a unified evaluation index.

[0060] Preferably, the comprehensive evaluation value is as follows:

[0061]

[0062] In the formula, is the comprehensive evaluation value, To integrate the objective function of energy consumption minimization and the objective function of load balancing and grid load optimization, is the weight output by the deep learning model, Future energy consumption data for IoT device energy consumption models.

[0063] Specifically, the comprehensive evaluation value is calculated, the goal is to comprehensively consider energy consumption minimization, load balancing optimization and future energy consumption prediction, and determine the best scheduling plan. The function normalizes the comprehensive evaluation value to It is convenient to compare the advantages and disadvantages of different scheduling schemes within the scope of the schedule, and provide a quantitative basis for determining the final scheduling scheme.

[0064] Preferably, The calculation formula is as follows:

[0065]

[0066] In the formula, The objective function for minimizing energy consumption is The parameter characteristics and the associated weights of the load balancing and grid load optimization objective functions, The total number of parameter characteristics of the objective function for minimizing energy consumption, is the activation function;

[0067] The calculation formula is as follows:

[0068] .

[0069] Specifically, is the th parameter feature of the fusion energy consumption minimization objective function and the fusion weight of the load balancing and power grid load optimization objective function, which reflects the correlation between the two objective functions and measures how to affect the power grid load balancing and load fluctuation while optimizing energy consumption. For example, the objective of the energy consumption minimization objective function is to minimize the total energy consumption of all IoT devices, and the objective of the load balancing and power grid load optimization objective function is to minimize the fluctuation of the power grid load and ensure the balance of the power grid load. The th parameter feature is the operating power of a certain device in a specific time period. In order to minimize the total energy consumption, the energy consumption minimization objective function tends to operate the device during the time period with lower electricity price. However, in order to reduce the power grid load fluctuation, the load balancing and power grid load optimization objective function needs to disperse the operating time of the device to different time periods to avoid the peak of the power grid load caused by concentrated power consumption. If the operating power of a device in a certain time period has a greater impact on the power grid load fluctuation (for example, the power grid load is already high during this time period), then will be smaller, indicating that the correlation between this parameter feature and the two objective functions is poor. If the operating power of a device in a certain time period has a smaller impact on the power grid load fluctuation (for example, the power grid load is low during this time period), then will be larger, indicating that the matching of this parameter feature with the two objective functions is better. Therefore, the parameter features include but are not limited to the operating power of a certain device in a specific time period, the power adjustment ability of a certain device (whether the device supports dynamic power adjustment), and the priority of a certain device (whether the device belongs to a critical load).

[0070] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations according to the concept of the present invention without creative labor. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. A method for optimizing energy scheduling based on IoT devices, characterized in that: The following steps are involved: S10: Real-time monitoring of energy consumption data of each IoT device based on the control terminal of each key node; S20: Use historical equipment energy consumption data to train a deep learning model, and output future energy consumption data of the IoT equipment energy consumption model based on the trained deep learning model; S30: Based on the future energy consumption data of the IoT device energy consumption model, construct the energy consumption minimization objective function and the load balancing and grid load optimization objective function; S40: performing a multi-objective fusion operation on the energy consumption minimization objective function and the load balancing and power grid load optimization objective functions to output a comprehensive evaluation function; S50: Calculate a comprehensive evaluation value based on the comprehensive evaluation function, and determine the best optimization scheduling plan by minimizing the comprehensive evaluation value; The energy consumption minimization objective function in S30 is as follows: In the formula, is the total energy consumption, is the total number of time periods in the scheduling cycle, is the total number of IoT devices, For the Devices at time The power consumption, To indicate the Devices at time Binary variable indicating whether to run; The load balancing and grid load optimization objective function in S30 is as follows: In the formula, For grid load fluctuations, is the average value of the grid load; The comprehensive evaluation function in S40 is as follows: In the formula, is the comprehensive evaluation function, The scaling factor for integrating the energy consumption minimization objective function and the load balancing and grid load optimization objective function is is the offset term.

2. According to claim 1, a method for optimizing energy scheduling based on IoT devices is characterized in that: The S10 includes: Identify and classify each IoT device, determine the type, function and priority of each IoT device, deploy control terminals at each key node, monitor the energy consumption data of the equipment in real time, pre-process the real-time monitoring data and store the pre-processed energy consumption data in the central database.

3. The method for optimizing energy scheduling based on IoT devices according to claim 1 is characterized in that: The S20 includes: Collect historical equipment energy consumption data and use it to train deep learning models, learn the time series characteristics and influencing factors of historical equipment energy consumption data, predict the energy consumption of each device in the next 24 hours based on the trained deep learning model, and output the future energy consumption data of the IoT equipment energy consumption model.

4. The method for optimizing energy scheduling based on IoT devices according to claim 1 is characterized in that: The comprehensive evaluation value is as follows: In the formula, is the comprehensive evaluation value, To integrate the objective function of energy consumption minimization and the objective function of load balancing and grid load optimization, is the weight output by the deep learning model, Future energy consumption data for IoT device energy consumption models.

5. The method for optimizing energy scheduling based on IoT devices according to claim 4 is characterized in that: include: Said The calculation formula is as follows: In the formula, The objective function for minimizing energy consumption is The parameter characteristics and the associated weights of the load balancing and grid load optimization objective functions, The total number of parameter characteristics of the objective function for minimizing energy consumption, is the activation function; Said The calculation formula is as follows: 。

Images