Energy consumption equipment operation adjusting method and device, equipment and storage medium

By predicting the energy supply and consumption curve of edge devices and dynamically adjusting task scheduling, the problem of uncertainty in the energy supply of edge devices in the power Internet of Things system is solved, and the stable operation and efficient energy utilization of equipment are achieved.

CN118941045APending Publication Date: 2024-11-12GUANGZHOU KETENG INFORMATION TECH
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202411099617.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the power Internet of Things system, edge-end energy consumption equipment faces complex challenges of energy supply due to the uncertainty of solar power supply. How to achieve long-term stable operation and real-time monitoring needs of equipment under limited and unstable energy supply, become A complex technical problem.

Method used

By obtaining regional meteorological parameters, equipment installation parameters and operating parameters of edge energy equipment, combining pre-trained equipment power generation models and energy consumption models, predicting energy supply and consumption curves, judging the risk of energy deficit, and dynamically adjusting task scheduling according to task priority level , improve energy utilization.

Benefits of technology

It realizes the stable operation of edge equipment and the normal execution of critical tasks, improves energy utilization, and ensures the real-time monitoring needs of the power Internet of Things system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941045A_ABST
    Figure CN118941045A_ABST
Patent Text Reader

Abstract

The invention discloses an energy consumption equipment operation adjusting method and device, equipment and a storage medium. And obtaining regional meteorological parameters, equipment installation parameters and equipment operation parameters corresponding to the edge energy equipment. And determining an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and an equipment power generation model and an energy consumption model obtained by pre-training. And based on the energy supply prediction curve and the energy consumption prediction curve, determining whether the energy consumption equipment has an energy deficiency risk. And under the condition of determining that the energy consumption equipment has the energy deficiency risk, the task operation of the energy consumption equipment is adjusted according to the task priority, so that stable operation of the edge energy consumption equipment can be realized, the energy utilization rate of the edge equipment is effectively improved, and normal execution of equipment tasks is guaranteed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy regulation, and in particular to a method, device, equipment and storage medium for regulating the operation of energy-consuming equipment. Background Art

[0002] In the power Internet of Things system, edge energy-consuming devices face complex energy supply challenges. These devices rely on solar energy for power supply, but the availability of solar energy is affected by multiple factors and is highly uncertain. First, the intensity of sunlight varies between day and night and in the weather, resulting in large fluctuations in energy supply. Second, the change of seasons causes the length of available sunlight to vary, further increasing the instability of energy acquisition. In this case, how to accurately predict and manage energy supply becomes a key issue. Devices need to efficiently collect and store energy when energy is sufficient, and reasonably allocate and use it when energy is scarce. However, due to the dynamic changes in the monitored objects and environment, the energy demand of the equipment also presents unpredictable characteristics. This requires a precise balance between the working requirements of the equipment and the available energy under limited and unstable energy supply.

[0003] In key monitoring tasks, how to ensure the continuous and reliable operation of energy-consuming equipment under energy shortage conditions has become a difficult technical problem. In addition, the differences in energy acquisition under different geographical locations and microclimate conditions also bring challenges to the overall design and optimization of the system. How to achieve long-term stable operation of energy-consuming equipment in such a highly dynamic and uncertain environment while meeting the real-time monitoring needs of the power Internet of Things system is a complex technical problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a method, device, equipment and storage medium for regulating the operation of energy-consuming equipment, so as to achieve stable operation of edge energy-consuming equipment, effectively improve the energy utilization rate of edge equipment, and ensure the normal execution of equipment tasks.

[0005] According to one aspect of the present invention, a method for regulating the operation of energy-consuming equipment is provided. The method comprises:

[0006] Obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment;

[0007] Determine an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training;

[0008] Based on the energy supply prediction curve and the energy consumption prediction curve, determining whether the energy consuming equipment has an energy deficit risk;

[0009] When it is determined that the energy-consuming equipment has an energy deficit risk, the task operation of the energy-consuming equipment is adjusted according to the task priority level.

[0010] According to another aspect of the present invention, a device for regulating the operation of energy-consuming equipment is provided. The device comprises:

[0011] A parameter acquisition module is used to obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment;

[0012] An energy prediction module, used to determine an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training;

[0013] A deficit determination module, used to determine whether there is an energy deficit risk for energy-consuming equipment based on the energy supply prediction curve and the energy consumption prediction curve;

[0014] The operation adjustment module is used to adjust the task operation of the energy-consuming equipment according to the task priority level when it is determined that the energy-consuming equipment has the risk of energy deficit.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the energy-consuming equipment operation adjustment method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the energy-consuming device operation adjustment method described in any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention obtains the regional meteorological parameters, equipment installation parameters and equipment operation parameters corresponding to the edge energy equipment. According to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the pre-trained equipment power generation model and energy consumption model, the energy supply prediction curve and the energy consumption prediction curve are determined. Based on the energy supply prediction curve and the energy consumption prediction curve, the risk of power shortage is judged. In the case of energy shortage risk, the task scheduling strategy is dynamically adjusted according to the task priority level. At the same time, it can also effectively improve the energy utilization rate of edge devices, ensure the execution of key tasks, and realize passive wireless real-time monitoring.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flow chart of a method for adjusting the operation of an energy-consuming device provided according to Embodiment 1 of the present invention;

[0024] Figure 2 is a flow chart of a method for adjusting the operation of an energy-consuming device provided according to Embodiment 2 of the present invention;

[0025] Figure 3 is a structural diagram of an energy-consuming equipment operation adjustment device provided according to Embodiment 3 of the present invention;

[0026] Figure 4 It is a structural schematic diagram of an electronic device for implementing the energy-consuming device operation adjustment method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] Embodiment 1

[0030] Figure 1 A flowchart of a method for regulating the operation of an energy-consuming device is provided for the first embodiment of the present invention. This embodiment is applicable to the case of passive wireless real-time monitoring and regulation of energy-consuming devices at the edge end. The method can be executed by an energy-consuming device operation regulation device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0031] S101. Obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment.

[0032] Among them, edge energy equipment may refer to edge power generation equipment of a clean energy power generation network, such as a solar power generation panel. Regional meteorological parameters may refer to meteorological parameters of the area where the edge energy equipment is located. Regional meteorological parameters may include historical meteorological data, weather forecast information, and current time. Equipment installation parameters may include the longitude and latitude, altitude, terrain feature parameters, and installation angle and orientation of the edge energy equipment at the geographical location of the edge energy equipment. Equipment operation parameters may include the power supply voltage, charge and discharge current, and battery temperature status of the edge energy equipment.

[0033] Specifically, the global positioning system is used to obtain the geographical location longitude and latitude of edge energy equipment, the digital elevation model is used to obtain the altitude, the digital terrain model is used to extract terrain characteristic parameters such as slope and aspect, historical meteorological data and weather forecast information are obtained from the meteorological department, and the seasonal change characteristics of the latitude area are determined according to the current system time. The equipment operation parameters are queried through the equipment label, etc.

[0034] S102, determining an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained through pre-training.

[0035] The energy supply prediction curve may refer to the prediction curve of energy output. Correspondingly, the energy consumption prediction curve may refer to the prediction curve of energy consumption. The equipment power generation model and the energy consumption model may be obtained by training sample parameters in advance. The equipment power generation model is used to predict energy output, and the energy consumption model is used to predict energy consumption.

[0036] Exemplarily, the process of determining the energy supply prediction curve and the energy consumption prediction curve may be as follows:

[0037] The regional environmental parameters corresponding to the edge energy equipment and the equipment installation parameters are input into the equipment power generation model for power generation prediction processing to obtain an energy supply prediction curve. The equipment operation parameters and the historical energy consumption data corresponding to the energy consumption equipment are input into the energy consumption model for energy consumption prediction processing to obtain an energy consumption prediction curve.

[0038] Among them, the regional environmental parameters are environmental parameter information of the area where the edge energy device is located. For example, the regional environmental parameters may include light altitude angle parameters, cloud cover and air pollution, etc.

[0039] Specifically, the regional environmental parameters corresponding to the edge energy equipment (such as the current solar altitude angle parameters) and equipment installation parameters such as the equipment installation angle and orientation are input into the equipment power generation model for power generation prediction processing. At the same time, the installation angle and orientation of the solar panel are dynamically optimized to obtain the energy supply prediction curve.

[0040] Similarly, the equipment operating parameters and historical energy consumption data such as the supply voltage, charge and discharge current and battery temperature state parameters are input into the energy consumption model for energy consumption prediction processing to obtain an energy consumption prediction curve. It should be noted that the energy consumption model can also generate an optimized task scheduling strategy for regulating the task operation of energy-consuming equipment.

[0041] Exemplarily, the training process of the device power generation model includes:

[0042] The regional meteorological parameters are sorted into sample meteorological parameters; based on the sample meteorological parameters and regional geographical parameters, the preset light intensity model is trained and processed to obtain a light intensity prediction model; based on the regional meteorological parameters and the light intensity prediction model, a light intensity prediction curve is obtained; based on the light intensity prediction curve, the equipment installation parameters and the preset power generation model, an equipment power generation model is obtained.

[0043] Among them, the sample meteorological parameters can be obtained by sorting the regional meteorological parameters.

[0044] Specifically, the regional geographical parameters and sample meteorological parameters such as the longitude and latitude, altitude and terrain characteristic parameters of the geographical location of the edge energy equipment are input into the preset light intensity model for training and processing to obtain a light intensity prediction model. The historical meteorological data, weather forecast information and current time are used to predict the regional meteorological parameters such as light intensity for a period of time in the future, and are input into the light intensity prediction model to predict the light intensity and obtain a light intensity prediction curve. The equipment installation angle, orientation and current solar altitude angle parameters are combined with the equipment installation parameters and the light intensity prediction curve and input into the preset power generation model for training to obtain the equipment power generation model.

[0045] In actual situations, the longitude and latitude of the geographical location of the edge energy device are obtained, the altitude is obtained through the digital elevation model, and the terrain characteristic parameters such as slope and aspect are extracted in combination with the digital terrain model. According to the regional geographical parameters such as the longitude and latitude of the geographical location, the altitude and the terrain characteristics, the direct light intensity reaching the ground is calculated. The long-term and short-term memory network is used to train the historical meteorological data. The input features include temperature, humidity, cloud cover and air quality index. The output target is light intensity, and the relationship model between light intensity and meteorological factors (i.e., light intensity model) is obtained. The weather forecast information is input into the trained model to obtain the predicted value of light intensity for a period of time in the future, and the predicted value is corrected by the Holt-Winters exponential smoothing method. The sunshine duration is calculated according to the latitude and date, and a short-term light intensity prediction curve is generated.

[0046] For example, according to the longitude and latitude of the geographical location, the altitude and terrain characteristic parameters, the solar altitude angle is calculated using the solar altitude angle calculation formula, the atmospheric transmittance is calculated in combination with the Beer-Lambert law, and the direct radiation intensity reaching the ground is calculated by the radiation transfer equation I = I0·e^(-τ / cosθ), where I is the ground radiation intensity, I0 is the top of the atmosphere radiation intensity, τ is the atmospheric optical thickness, θ is the solar zenith angle, and the scattered radiation intensity is calculated to obtain the theoretical light intensity. The long short-term memory network is used to train historical meteorological data. The input features include temperature, humidity, cloudiness, air quality index, etc. The output target is light intensity. A relationship model between light intensity and meteorological factors is constructed. The weather forecast information is input into the trained model to obtain the light intensity forecast value for a period of time in the future. The forecast value is corrected by the Holt-Winters exponential smoothing method, where the smoothing parameters α, β, and γ are used to adjust the level, trend, and seasonal factors, respectively. The sunshine duration is calculated according to the latitude and date, and as an additional input feature, a short-term light intensity prediction curve is generated. The short-term light intensity prediction curve is segmented by the sliding time window method. The frequency domain features of each curve are extracted by fast Fourier transform. The time-frequency features of the curve are analyzed by discrete wavelet transform. The features are modeled by autoregressive moving average model to obtain the trend of light intensity changes. During the implementation process, the geographical location and longitude of the edge device are first obtained through the global positioning system, such as A° north latitude and B° east longitude. The digital elevation model is used to obtain an altitude of 50 meters, and the slope is 5° extracted through the digital terrain model, with a southeast slope. Historical meteorological data and weather forecast information are obtained from the meteorological department, including temperature, humidity, cloud cover and other data for the past 30 days and forecasts for the next 7 days. According to the current system time July 31, 2024, it is determined that the latitude area is in summer. The solar altitude angle is calculated to be 65°, the atmospheric transmittance is calculated to be 0.85 using the Beer-Lambert law, and the direct radiation intensity on the ground is calculated to be 800W / m through the radiation transfer equation. 2 , the scattered radiation intensity is 200W / m 2 , the theoretical light intensity is 1000W / m 2 The long short-term memory network is used to train historical meteorological data. The input features include temperature in degrees Celsius, relative humidity %, cloud cover %, air quality index, etc. The output target is light intensity W / m 2. The weather forecast information was input into the trained model to obtain the predicted light intensity for the next 7 days. The predicted value was corrected using the Holt-Winters exponential smoothing method, with smoothing parameters α=0.3, β=0.1, and γ=0.1. The sunshine duration was calculated to be 14 hours based on the latitude of 30.5° and the date of July 31, and was used as an additional input feature to generate a short-term light intensity prediction curve. The prediction curve was segmented using a 24-hour sliding time window, and the frequency domain features of each segment were extracted by fast Fourier transform to obtain the main frequency components. A 5-layer discrete wavelet transform was performed using the db4 wavelet basis function to analyze the time-frequency characteristics of the curve. Finally, the ARMA(2,1) model was used to model the features and obtain the trend of light intensity changes. The prediction results show that the average daily light intensity in the next 7 days will be between 900-1100W / m 2 Fluctuates between.

[0047] On the other hand, an angle sensor is used to obtain the installation angle of the solar panel, an electronic compass is used to obtain the installation orientation, and the geographical latitude is obtained from the global positioning system. A support vector regression model is established based on the installation angle, installation orientation, geographical latitude and light intensity to obtain a photovoltaic power generation efficiency model. The sunrise and sunset times are calculated based on the geographical latitude, the light intensity prediction curve is obtained, and the light intensity peak time period is determined. A particle swarm optimization algorithm is used to search for the optimal installation angle and orientation. The particle swarm optimization algorithm uses the photovoltaic power generation efficiency model as an evaluation function to optimize the cumulative power generation during the light intensity peak time period. A stepper motor is used to control an adaptive adjustment mechanism to adjust the pitch and azimuth angles of the solar panel. The adaptive adjustment mechanism is adjusted according to the optimal installation angle and orientation. According to the optimal installation angle and orientation and the photovoltaic power generation efficiency model, the optimized energy supply prediction curve is calculated.

[0048] Specifically, the angle sensor is used to collect the current installation angle of the solar panel, the installation direction is obtained through the electronic compass, the geographical latitude is obtained by combining the global positioning system, the solar altitude angle is calculated according to the current system time, and the formula sin(h)=sin(φ)sin(δ)+cos(φ)cos(δ)cos(ω) is used, where h is the solar altitude angle, φ is the geographical latitude, δ is the solar declination, and ω is the hour angle to obtain the light intensity prediction curve. The photovoltaic power generation efficiency model is established by the support vector regression algorithm. The input variables include the installation angle of the solar panel, the installation direction, the solar altitude angle, and the light intensity. The output variable is the photovoltaic power generation efficiency. The RBF kernel function is used to optimize the parameters C and γ through grid search. The model is trained using historical operation data containing power generation records under different time and weather conditions. The sunrise and sunset times are calculated according to the geographical latitude and the current date. The peak time period of light intensity is determined by combining the light intensity prediction curve. The optimal solar panel installation angle and orientation are searched by the particle swarm optimization algorithm. The number of particles is set to 50, the maximum number of iterations is 100, the acceleration constant c1=c2=2, and the established photovoltaic power generation efficiency model is used as the evaluation function. The optimization target is the cumulative power generation during the peak time period of light intensity. The adaptive adjustment mechanism controlled by the stepper motor is used to dynamically adjust the pitch and azimuth of the solar panel. According to the optimized installation parameters and the photovoltaic power generation efficiency model, the optimized energy supply prediction curve is calculated. The optimization results are saved through the data storage module, and the optimization strategy is dynamically adjusted according to the geographical latitude and seasonal changes. Different optimization parameter ranges are used in different seasons. In actual applications, the angle sensor collects the current installation angle of the solar panel as 30°, the electronic compass obtains the installation orientation as the south direction, and the global positioning system obtains the geographical latitude as 35° north latitude. According to the current system time 12:00 on August 1, 2024, the solar altitude angle is calculated to be 78.5°. The light intensity prediction curve shows that the highest light intensity of the day is 1000W / m 2, which appears between 13:00 and 14:00. The support vector regression algorithm uses the RBF kernel function to establish a photovoltaic power generation efficiency model, and the optimal parameters C=10, γ=0.1 are determined by grid search. The historical operation data contains the power generation records under different time and weather conditions in the past year, with a total of 10,000 data points. The prediction accuracy of the model after training reaches 95%. According to the geographical latitude of 35° and the date of August 1, the sunrise time is calculated to be 5:30 and the sunset time is 19:30. The particle swarm optimization algorithm sets the number of particles to 50, the maximum number of iterations to 100, the acceleration constant c1=c2=2, and the search range is 0°-90° in pitch angle and -90°-90° in azimuth angle. The optimization results show that the best installation angle is 25°, the best azimuth angle is 5°, and the direction is slightly southeast. The adaptive adjustment mechanism controlled by the stepper motor dynamically adjusts the pitch and azimuth angles of the solar panel with an accuracy of 0.1°. The optimized energy supply forecast curve shows that the daily power generation can reach 28kWh, which is 15% higher than before optimization. The optimization parameters and power generation data are recorded every hour. The system dynamically adjusts the optimization strategy according to different seasons. The optimization parameter range in summer is 15°-45° pitch angle, and in winter is 35°-65° pitch angle to adapt to the seasonal changes in the solar altitude angle.

[0049] Exemplarily, the training process of the energy consumption model includes:

[0050] The equipment operation parameters are sorted into sample operation parameters. According to the sample operation parameters and the historical energy consumption data corresponding to the energy consumption equipment, a preset consumption model is trained to obtain an energy consumption model.

[0051] Specifically, the device operating parameters such as the power supply voltage, charge and discharge current, and battery temperature of the edge device are organized into sample operating parameters, and the preset consumption model is trained using the sample operating parameters to obtain an energy consumption model.

[0052] In the actual process, a voltage sensor is used to obtain the power supply voltage data from the edge energy device, a current sensor is used to obtain the charge and discharge current data, and a temperature sensor is used to measure the battery temperature data. The current device operating status parameters are obtained, and the device energy consumption feature vector is constructed in combination with the historical energy consumption records stored in the historical energy consumption database. According to the device energy consumption feature vector, the device energy consumption is modeled by the particle swarm optimization algorithm to obtain an energy consumption prediction model. The task execution plan for the next twenty-four hours is obtained, and the task execution plan includes the task type, the expected execution time and the task priority. The autoregressive integral moving average model is used to predict the power supply voltage for the next twenty-four hours, and the power outage risk index is calculated in combination with the battery capacity and the discharge characteristic curve. If the power outage risk index is greater than the preset threshold, the task scheduling is optimized. According to the optimized task scheduling strategy and the energy consumption prediction model, an energy consumption prediction curve is generated.

[0053] Specifically, the voltage sensor is used to collect the power supply voltage of the edge device, the current sensor is used to obtain the charge and discharge current data, the temperature sensor is used to measure the battery temperature, the current device operating status parameters are obtained, and the energy consumption feature vector of the device is constructed by combining the energy consumption records of the past six months stored in the historical energy consumption database. The energy consumption of the device is modeled by the particle swarm optimization algorithm, the number of particles is set to 50, the maximum number of iterations is 100, the inertia weight is 0.7, the local learning factor c1 is 1.5, the global learning factor c2 is 1.5, the particle encoding method is the energy consumption model parameter, the speed update formula is v(t+1)=wv(t)+c1r1*(pbest-x(t))+c2r2(gbest-x(t)), the optimization objective function is to minimize the prediction error, and the energy consumption prediction model is obtained. r1 and r2 are two numbers randomly generated from the range of [0,1], pbest is the best position of each particle and gbest is the best position of the group. Obtain the task execution plan within the next 24 hours, including task type, expected execution time and task priority. Use the autoregressive integrated moving average model to predict the power supply voltage within the next 24 hours. Combine the battery capacity and discharge characteristic curve to calculate the power outage risk index, which is defined as (predicted power consumption - available power) / available power. Use the genetic algorithm to optimize the task scheduling. The chromosome encoding method is the task execution order. The population size is set to 100, the crossover probability is 0.8, and the mutation probability is 0.1. The fitness function is defined as w1*task priority score + w2*(1 / energy consumption) + w3*(1 / power outage risk index), where w1, w2, and w3 are weight coefficients. The task priority score is calculated based on the urgency and importance of the task. After 500 iterations, the optimized task scheduling strategy is obtained. The energy consumption prediction curve for the next 24 hours is generated based on the scheduling strategy and the energy consumption prediction model. In actual applications, the voltage sensor collects the power supply voltage of the edge device as 12.5V, the current sensor obtains the charging current of 0.5A and the discharging current of 1.2A, and the temperature sensor measures the battery temperature of 25℃. The device management module shows that the current device is in normal operation and the CPU usage is 60%. The historical energy consumption database stores energy consumption records for the past six months, with an average daily energy consumption of 120Wh. After 100 iterations, the particle swarm optimization algorithm obtained the optimal parameters of the energy consumption prediction model, and the prediction error was reduced to less than 5%. The task queue manager obtained 10 tasks to be executed in the next 24 hours, including 3 high-priority tasks, 5 medium-priority tasks, and 2 low-priority tasks. The autoregressive integrated moving average model predicts that the average supply voltage in the next 24 hours will be 12.3V, the battery capacity will be 5000mAh, and the discharge characteristic curve shows that the available capacity is 80% at 12.3V. The power outage risk index is calculated to be 0.15.After 500 iterations of the genetic algorithm, the optimized task scheduling strategy is to schedule three high-priority tasks to be executed when the battery is fully charged, and postpone two low-priority tasks to a time period with a lower risk of power outage. The weight coefficients in the fitness function are set to w1 = 0.5, w2 = 0.3, and w3 = 0.2. The energy consumption prediction curve generated in the end shows that the total energy consumption of the equipment in the next 24 hours is expected to be 115Wh, which is 4.2% lower than before optimization.

[0054] S103: Determine whether energy-consuming equipment has an energy deficit risk based on the energy supply prediction curve and the energy consumption prediction curve.

[0055] Specifically, the energy supply prediction curve and the energy consumption prediction curve are compared to determine the energy difference in each period between the two. If the energy supply prediction curve is greater than the energy consumption prediction curve and the energy difference is greater than the energy difference threshold, it is determined that the energy-consuming equipment does not have an energy deficit risk. Otherwise, it is determined that the energy-consuming equipment has an energy deficit risk.

[0056] S104: When it is determined that the energy-consuming equipment has an energy deficit risk, the task operation of the energy-consuming equipment is adjusted according to the task priority level.

[0057] Specifically, when it is determined that there is an energy deficit in the energy-consuming device, all execution tasks of the energy-consuming device are determined, and the priorities of all execution tasks are determined. According to the order of task priorities, the task execution status of each energy-consuming device is adjusted.

[0058] Exemplarily, the task processing of the energy-consuming device is adjusted according to the task priority, including:

[0059] Determine the task priority of the equipment operation task; based on the dynamic programming algorithm, determine the task execution order corresponding to the standby operation task according to the task priority; based on the task execution order, adjust the working state of the energy-consuming equipment so that the energy-consuming equipment does not have the risk of energy deficit.

[0060] Specifically, the current task list and its priority information are obtained, and the tasks are classified using the preset threshold rules to obtain critical monitoring tasks, important tasks, and non-critical tasks. The task execution order is rearranged through a dynamic programming algorithm. According to the task execution order, the operating frequency and power consumption of the device are adjusted until the energy surplus in each time period is greater than the preset safety threshold, and the task queue and device working mode configuration are updated.

[0061] Exemplarily, after adjusting the task processing of the energy-consuming equipment according to the task priority level, it also includes: determining the predicted light intensity based on the regional predicted meteorological parameters corresponding to the edge energy equipment and the pre-trained light intensity prediction model; based on a multi-objective genetic algorithm, adjusting the task execution time and task execution frequency of the energy-consuming equipment according to the predicted light intensity.

[0062] Specifically, the light intensity change trend is modeled through a long short-term memory network to obtain the light intensity forecast value for a preset time in the future. The input variables of the long short-term memory network include the light intensity data of the current time and the preset time in the past and the weather forecast information for the preset time in the future. A multi-objective genetic algorithm is used to optimize the task execution time ratio and frequency. The chromosome encoding of the multi-objective genetic algorithm includes the execution time ratio and frequency of critical tasks and non-critical tasks.

[0063] The technical solution of the embodiment of the present invention obtains the regional meteorological parameters, equipment installation parameters and equipment operation parameters corresponding to the edge energy equipment. According to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the pre-trained equipment power generation model and energy consumption model, the energy supply prediction curve and the energy consumption prediction curve are determined. Based on the energy supply prediction curve and the energy consumption prediction curve, the risk of power shortage is judged. In the case of energy shortage risk, the task scheduling strategy is dynamically adjusted according to the task priority level. At the same time, it can also effectively improve the energy utilization rate of edge devices, ensure the execution of key tasks, and realize passive wireless real-time monitoring.

[0064] Exemplarily, the method may further include:

[0065] Based on the geographic location parameters, the changes in the length of day and night and the solar altitude angle are obtained, the historical meteorological data of atmospheric transparency and cloud coverage are analyzed, the light intensity distribution characteristics are obtained, the spatial distribution of solar radiation on the surface is calculated, and the site selection and panel layout plan of the photovoltaic power station are determined.

[0066] The geographical latitude data of the target area is obtained, and the changes in the length of day and night and the solar altitude angle data of different latitudes throughout the year are calculated; based on the changes in the length of day and night and the solar altitude angle data, the time series is analyzed by the autoregressive integrated moving average model to obtain the light intensity distribution characteristics of each region; the spatial distribution of solar radiation on complex terrain is calculated by the Beer-Lambert law; a light resource evaluation model is established in combination with the light intensity distribution characteristics; a genetic algorithm is used to optimize the site selection plan of the photovoltaic power station, wherein the population size is set as a preset value, the crossover rate is set as a preset value, and the mutation rate is set as a preset value, and the fitness function of the algorithm is set in combination with the annual average sunshine hours, terrain slope, vegetation coverage and seasonal factors; the Monte Carlo method is used to simulate the power generation under different battery panel layout schemes, wherein the number of sampling times is set as a preset value, and the random variables include sunshine intensity and duration, which obey the normal distribution, and the optimal battery panel layout scheme is determined by the particle swarm optimization algorithm, and the optimization goal is to maximize the annual power generation.

[0067] Specifically, the geographic latitude data of the target area is obtained from the geographic information system, the changes in the length of day and night and the solar altitude angle of different latitudes throughout the year are calculated by the solar position algorithm, the historical data of atmospheric transparency and cloud coverage are extracted from the meteorological database, and the time series is analyzed by the autoregressive integral moving average model to obtain the distribution characteristics of light intensity in each region. The topographic data of the target area are obtained by the digital elevation model, and the spatial distribution of solar radiation on complex terrain is calculated by the Beer-Lambert law I = I0*exp(-τ*m), where I is the surface radiation intensity, I0 is the top of the atmosphere radiation intensity, τ is the atmospheric optical thickness, and m is the atmospheric mass. Combined with the obtained light intensity distribution characteristics, a light resource evaluation model is established. According to the astronomical calendar, the annual change law of the sun's trajectory is calculated. Combined with the established light resource evaluation model, the genetic algorithm is used to optimize the site selection plan of the photovoltaic power station, and the population size is set to 100, the crossover rate is 0.8, and the mutation rate is 0.1. The fitness function of the algorithm considers the annual average sunshine hours, terrain slope, vegetation coverage and seasonal factors. For the optimized photovoltaic power station site selection, the Monte Carlo method is used to simulate the power generation under different panel layout schemes. The sampling number is set to 10,000. The random variables include sunshine intensity and duration, which obey the normal distribution. Considering the parameters such as panel spacing, inclination, and orientation, the particle swarm optimization algorithm is used to determine the optimal panel layout scheme. The optimization goal is to maximize the annual power generation. In practical applications, the geographic information system obtains the geographic coordinates of the target area at 35° north latitude and 114° east longitude. The solar position algorithm calculates that the length of daylight on the summer solstice in the region is 14 hours and 36 minutes, the length of daylight on the winter solstice is 9 hours and 42 minutes, and the annual average solar altitude angle is 39.5°. The meteorological database provides atmospheric transparency and cloud coverage data for the past 10 years, with an average atmospheric transparency of 0.7 and a cloud coverage of 45%. The results of the autoregressive integrated moving average model analysis show that the average annual sunshine hours in the region are 2,100 hours, and the peak light intensity occurs in June-August, which is 1000W / m 2 The terrain data obtained from the digital elevation model show that the average altitude of the area is 500 meters and the maximum slope is 15°. The calculation results of the Beer-Lambert law show that in clear weather, the surface radiation intensity is 750W / m 2 The light resource assessment model takes the above factors into consideration and concludes that the average annual solar radiation in the area is 1450kWh / m 2 The genetic algorithm optimizes the site selection of the photovoltaic power station. After 500 iterations, the optimal site coordinates are 35.2°N, 114.3°E. The average annual sunshine hours at this location are 2200 hours, the terrain slope is 8°, and the vegetation coverage is 20%. The Monte Carlo method is simulated 10,000 times, and the sunshine intensity follows the mean value of 800W / m 2 、Standard deviation 100W / m 2The sunshine time follows a normal distribution with a mean of 10 hours and a standard deviation of 2 hours. After 1,000 iterations, the particle swarm optimization algorithm determined the best solar panel layout: 4.5 meters spacing, 32° inclination, 5° east of due south, and an estimated annual power generation of 1.8GWh.

[0068] Embodiment 2

[0069] Figure 2 This is a flow chart of a method for adjusting the operation of energy-consuming equipment provided in the second embodiment of the present invention. Based on the above embodiments, this embodiment further refines the determination of whether the energy-consuming equipment has an energy deficit risk based on the energy supply prediction curve and the energy consumption prediction curve. Figure 2 As shown, the method includes:

[0070] S201. Obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment.

[0071] S202, determining an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained through pre-training.

[0072] S203. Based on the trapezoidal integration method, the energy supply prediction curve and the energy consumption prediction curve are discretized to obtain the time period energy result of each time period.

[0073] Specifically, the optimized energy supply forecast curve and energy consumption forecast curve are discretized by using the trapezoidal integration method, 24 hours are divided into 24 1-hour time periods, and the time period energy result of each time period is calculated. For example, the energy surplus or deficit of each time period is obtained by subtraction operation.

[0074] S204: Determine the accumulated energy balance within a preset time period according to the energy result of the time period.

[0075] Among them, the preset time period can be set according to actual needs or according to historical experience. In the present disclosure, the preset time period is preferably 6 hours.

[0076] Specifically, according to the energy surplus or shortage of the energy result of the time period, a preset time period is set, and the accumulated energy surplus of the preset time period is calculated.

[0077] S205: Determine whether the energy-consuming equipment has an energy deficit risk based on the accumulated energy surplus and the total energy consumption of the energy-consuming equipment.

[0078] Specifically, if the accumulated energy surplus is lower than a preset ratio threshold (eg, 20%) of the total energy consumption of the equipment, it is determined that there is an energy deficit risk, and the task scheduling optimization mechanism is triggered. Otherwise, it is determined that there is no energy deficit risk.

[0079] S206: When it is determined that the energy-consuming equipment has an energy deficit risk, the task operation of the energy-consuming equipment is adjusted according to the task priority level.

[0080] In practice, the current task list and its priority information are obtained, and simple threshold rules are used to classify tasks. Priorities greater than 8 are critical monitoring tasks, 5-8 are important tasks, and less than 5 are non-critical tasks. The working mode of the device is determined according to the task category and the current remaining power of the device. The task execution order is rearranged through a dynamic programming algorithm, and the state transition equation is dp[i][j]=max(dp[i-1][j],dp[i-1][j-energy[i]]+value[i]), where dp[i][j] represents the maximum benefit of the first i tasks when the remaining power is j, dp[i-1] represents the maximum benefit when the i-th task is not executed, energy[i] represents the power required by the i-th task, and value[i] represents the benefit that can be obtained by completing the i-th task under given conditions. Prioritize the critical monitoring tasks, delay or suspend the execution of non-critical tasks, adjust the working frequency and power consumption of the device until the energy surplus in each time period is greater than the preset safety threshold of 10Wh, and update the task queue and device working mode configuration. Check regularly whether the energy surplus meets all task requirements. If not, re-trigger the task scheduling optimization. In practical applications, the trapezoidal integration method is used to process the energy supply prediction curve and the energy consumption prediction curve, and 24 hours are divided into 24 1-hour time periods. The calculation results show that in the period of 8:00-9:00, the energy supply is 120Wh, the energy consumption is 100Wh, and the energy surplus is 20Wh. The sliding window mechanism calculates that the cumulative energy surplus for 6 consecutive hours from 6:00 to 12:00 is 80Wh, which is lower than 20% of the total power of 2000Wh of the equipment, that is, 400Wh, triggering the risk of power deficit. There are 10 tasks in the task queue, with priorities of 9, 8, 7, 6, 5, 4, 3, 2, 1, and 0 respectively. After applying the threshold rule, the classification results are 2 critical monitoring tasks, 3 important tasks, and 5 non-critical tasks. The dynamic programming algorithm rearranges the order of tasks, and the state transition equation calculates that when the remaining power is 1500Wh, the maximum benefit of executing the first 5 tasks is 85 points. After optimization, two critical monitoring tasks were scheduled to be executed at 9:00 and 10:00, three important tasks were scheduled to be executed at 11:00, 13:00 and 15:00, and five non-critical tasks were delayed to be executed after 18:00. The operating frequency of the equipment was reduced from 2.4GHz to 1.8GHz, and the power consumption was reduced from 10W to 7W. After optimization, the energy surplus in each time period exceeded the safety threshold of 10Wh. The energy surplus was checked every hour, and if it was less than 5Wh, the task scheduling optimization was triggered again.

[0081] Exemplarily, the method may further include:

[0082] According to the changing trend of light intensity and the remaining power of the equipment, the execution time ratio and frequency of critical tasks and non-critical tasks are adaptively adjusted to maximize the execution efficiency of non-critical tasks while ensuring the critical tasks.

[0083] Obtain the energy surplus at a preset time in the future, which is calculated by the energy supply prediction curve and the energy consumption prediction curve; dynamically adjust the judgment criteria, the adaptive threshold mechanism calculates the mean and standard deviation based on the historical energy surplus data, and sets the threshold to the mean plus one standard deviation. Model the light intensity change trend through a long short-term memory network to obtain the light intensity prediction value at a preset time in the future. The input variables of the long short-term memory network include the light intensity data of the current time, the preset time in the past, and the weather forecast information at the preset time in the future. A multi-objective genetic algorithm is used to optimize the task execution time ratio and frequency. The chromosome encoding of the multi-objective genetic algorithm includes the execution time ratio and frequency of critical tasks and non-critical tasks. The fitness function combines the three goals of critical task completion rate, non-critical task execution efficiency, and energy utilization rate.

[0084] Specifically, according to the energy supply prediction curve and the energy consumption prediction curve, the energy surplus for each hour in the next 24 hours is calculated, and the judgment criteria are dynamically adjusted. The average and standard deviation are calculated based on the energy surplus data of the past 7 days. The threshold is set as the average plus a standard deviation. If the cumulative energy surplus for 6 consecutive hours is greater than the threshold, it is determined that the device has sufficient energy surplus and the hierarchical task scheduling mechanism is triggered. The long short-term memory network is used to model the trend of light intensity changes. The input variables include the current time, the light intensity data of the past 24 hours, and the weather forecast information of the next 24 hours, including cloud cover, precipitation probability, and temperature. The output variable is the predicted value of light intensity in the next 24 hours. Combined with the current remaining power of the device, the total energy available for task execution is determined. The task execution time ratio and frequency are optimized by a multi-objective genetic algorithm. The chromosome encoding includes the execution time ratio and frequency of critical tasks and non-critical tasks. The population size is set to 100, the crossover rate is 0.8, and the mutation rate is 0.1. The fitness function considers the three goals of critical task completion rate, non-critical task execution efficiency, and energy utilization rate. The light intensity prediction result is introduced into the optimization process as a constraint. According to the optimization results, the working mode and task execution plan of the equipment are dynamically adjusted. The task execution status and energy consumption status are recorded through the real-time monitoring mechanism. When the actual energy consumption deviates from the predicted value by more than 10%, the task execution frequency is automatically adjusted. If abnormal fluctuations occur, the task scheduling optimization process is re-triggered until the energy surplus no longer meets the triggering conditions. The optimization effect is evaluated regularly, and the parameters and strategies of each step are adjusted according to the evaluation results to achieve adaptive adjustment. In actual applications, the energy surplus per hour in the next 24 hours is calculated. The average energy surplus in the past 7 days is 200Wh, the standard deviation is 50Wh, and the adaptive threshold is set to 250Wh. It is detected that the cumulative energy surplus from 8:00 to 14:00 for 6 consecutive hours is 300Wh, which exceeds the threshold and triggers hierarchical task scheduling. The LSTM network uses a three-layer structure with 64 neurons in each layer. It inputs the light intensity data of the past 24 hours, with one data point per hour, and the weather forecast for the next 24 hours is 0-100% cloud cover, 0-100% precipitation probability, and temperature -10℃ to 40℃. It outputs the predicted light intensity value for the next 24 hours. The prediction results show that the light intensity is highest from 12:00 to 14:00, reaching 1000W / m 2. The current remaining power of the device is 3000mAh. Combined with the light forecast, the total available energy is determined to be 15000mWh. The multi-objective genetic algorithm sets the population size to 100, the crossover rate to 0.8, the mutation rate to 0.1, and the iterations to 500 times. The optimization results show that the critical task execution time ratio is 60%, the non-critical task is 40%, the critical task execution frequency is 2 times per hour, and the non-critical task is 1 time per hour. Real-time monitoring shows that the actual energy consumption from 13:00 to 14:00 is 15% lower than the predicted value. The system automatically adjusts the execution frequency of non-critical tasks to 1.2 times per hour. The optimization effect is evaluated every 6 hours. According to the evaluation results, the learning rate of the long short-term memory network is adjusted from 0.01 to 0.008, and the crossover rate of the multi-objective genetic algorithm is adjusted from 0.8 to 0.85 to achieve adaptive adjustment.

[0085] Set the priority weight of key tasks for each task in the key task set, and obtain the priority list of non-key tasks by establishing the task value model. Take the total energy budget as the constraint condition and the priority weight of key tasks as the objective function to solve the optimal execution order of key tasks, calculate the remaining energy after executing all key tasks, and dynamically adjust the execution strategy of non-key tasks.

[0086] A critical task set and a non-critical task set are obtained from the task manager. A hierarchical analysis method is used to assign a priority weight to each critical task according to the critical task set. The priority weight is obtained by constructing a judgment matrix and calculating a characteristic value. A task value model is constructed for the non-critical task set. A multi-factor scoring method is used to calculate a value score for the non-critical task. The value score is determined by factors such as task completion time, resource consumption, and expected benefits. The total energy budget of the equipment is obtained. A regression analysis method based on historical data is used to estimate the expected energy consumption of each task. The expected energy consumption is combined with the critical task priority weight to determine the execution order of the critical tasks. A genetic algorithm is used to solve the optimal execution order of the critical tasks according to the critical task priority weight and the expected energy consumption. The fitness function of the genetic algorithm is defined as the sum of the product of the priority weight and the task completion degree minus the weighted value of the energy consumption. If the remaining energy after executing all critical tasks is less than a preset multiple of the expected consumption, an energy emergency plan is triggered. According to the remaining energy and the non-critical task value score, a dynamic programming algorithm is used to optimize the non-critical task execution strategy. The dynamic programming algorithm obtains the optimal execution strategy by setting a state transition equation for iterative calculation.

[0087] Specifically, we obtain a set of critical tasks and a set of non-critical tasks, and use the hierarchical analysis method to assign priority weights to each critical task. When constructing the judgment matrix, we consider the urgency, importance, and dependency of the task, and calculate the weight vector using the eigenvalue method. We build a task value model, and use the multi-factor scoring method to calculate the value score for non-critical tasks. The factors include task completion time, resource consumption, and expected benefits. The weights of each factor are determined by the Delphi method to obtain the non-critical task execution priority list. We obtain the total energy budget of the equipment, and use the regression analysis method based on historical data to estimate the expected energy consumption of each task. Combined with the priority weights of critical tasks, we use the genetic algorithm to solve the optimal execution order of critical tasks. The population size is set to 100, the crossover rate is 0.8, the mutation rate is 0.1, and the iteration is 500 times. The fitness function is defined as Σ(task priority weight*task completion)-α*energy consumption, where α is the balance coefficient. According to the optimal execution order, we simulate the execution of critical tasks, calculate the remaining energy after executing all critical tasks, introduce the sliding window mechanism, and calculate the expected energy consumption in the next 6 hours. If the remaining energy is less than 1.2 times the expected consumption, the energy emergency plan is triggered and all non-critical tasks are suspended. Combined with the remaining energy situation, the dynamic programming algorithm is used to optimize the execution strategy of non-critical tasks. The state transfer equation dp[i][j] is set to represent the maximum value of the first i tasks when the remaining energy is j. The constraints include the task deadline and the upper limit of energy consumption. The optimal execution strategy is obtained through iterative calculation, and the execution order and time allocation of non-critical tasks are dynamically adjusted. A feedback mechanism is added to dynamically adjust the expected execution time and energy consumption estimation parameters of the task according to the actual task execution situation and energy consumption. An exception handling mechanism is set to re-trigger the task scheduling optimization process when the task execution fails or is interrupted. In actual application, 5 critical tasks and 10 non-critical tasks are obtained. The judgment matrix constructed by the hierarchical analysis method is tested for consistency, and the weights of the five critical tasks are calculated to be 0.35, 0.25, 0.20, 0.12 and 0.08 respectively. The multi-factor scoring method is used to evaluate non-critical tasks. The weights of task completion time, resource consumption and expected benefits are 0.4, 0.3 and 0.3 respectively, and the value score ranking of non-critical tasks is obtained. The total energy budget of the equipment is 10,000 mWh. The regression analysis estimates the energy consumption of each task. The average energy consumption of critical tasks is 800 mWh, and the average energy consumption of non-critical tasks is 400 mWh. After 500 iterations of the genetic algorithm, the optimal execution order of critical tasks is 2-1-3-5-4, and the balance coefficient α in the fitness function is set to 0.01. After simulating the execution of critical tasks, the remaining energy is 6,000 mWh. The sliding window mechanism predicts that the energy consumption in the next 6 hours will be 5,000 mWh, and the energy emergency plan is not triggered.The dynamic programming algorithm optimizes the execution strategy of non-critical tasks. The state space is a two-dimensional array of 10*6000. The optimal execution order obtained by iterative calculation is 7-3-1-9-5-2-8-4-6-10. The feedback mechanism shows that the actual execution time of task 3 is 20% less than expected, and the system automatically adjusts the expected execution time of subsequent tasks. During the execution process, task 9 is interrupted, re-triggering the task scheduling optimization, and the adjusted execution order is 5-2-8-4-6-10-9.

[0088] Furthermore, in view of the fact that light intensity is affected by cloud cover and atmospheric pollution, the light intensity prediction model is corrected in real time, and the energy budget strategy is dynamically adjusted. According to the differences in energy supply status between devices, energy sharing and balance between nearby devices are achieved through wireless communication networks.

[0089] The online gradient descent algorithm is used to perform real-time correction on the light intensity prediction model to obtain the corrected light intensity prediction result. According to the corrected light intensity prediction result, the light intensity change amplitude is mapped to the weight adjustment coefficient through the sigmoid function; the new task weight coefficient is calculated to dynamically adjust the weights and priorities of critical tasks and non-critical tasks. Based on the adjusted task weights and priorities, combined with historical energy consumption data and future task plans; the time series prediction algorithm is used to estimate future energy demand and obtain future energy demand forecasts. The dynamic programming algorithm is used to optimize the energy allocation strategy; the optimal energy allocation plan is obtained through iterative calculation. An energy status information exchange mechanism is established between adjacent devices through a wireless communication network; the Raft consensus algorithm is used to calculate the average energy level of the device group. If the energy level of a single device is lower than the preset value, the energy sharing process is triggered.

[0090] Specifically, the light intensity data is collected in real time by the light sensor, combined with the cloud coverage and air pollution index information provided by the meteorological department, and the online gradient descent algorithm is used to correct the light intensity prediction model in real time. The cloud coverage and air pollution index are used as additional feature inputs to update the model parameters and improve the prediction accuracy. According to the corrected light intensity prediction results, the weights and priorities of key tasks and non-key tasks are dynamically adjusted, and the sigmoid function is used to map the light intensity change amplitude to the weight adjustment coefficient to calculate the new task weight coefficient. Based on the adjusted task weights and priorities, combined with historical energy consumption data and future task plans, the time series prediction algorithm is used to estimate future energy demand, and the dynamic programming algorithm is used to optimize the energy allocation strategy. The state transfer equation dp[i][j] is set to represent the maximum benefit of the first i tasks when the remaining energy is j. The constraints include the minimum energy reserve and the task completion time limit. The optimal energy allocation plan is obtained through iterative calculation. The energy status information exchange mechanism between adjacent devices is established through the wireless communication network. The Raft consensus algorithm is used to calculate the average energy level of the device group. If the energy level of a single device is lower than 80% of the average value, the energy sharing process is triggered. The multi-hop energy relay mechanism is used to realize long-distance energy transmission. The radio frequency energy harvesting technology is used to power the passive sensor to realize real-time monitoring. The local abnormal factor algorithm is used to identify abnormal conditions such as sudden changes in light intensity and equipment failures, and the corresponding emergency processing process is triggered. In actual applications, the light sensor collects light intensity data once a minute, and the current reading is 800 lux. The cloud coverage rate provided by the meteorological department is 60%, and the atmospheric pollution index is 150. The online gradient descent algorithm updates the model parameters with a learning rate of 0.01, and takes the cloud coverage rate and pollution index as new features. The corrected light intensity prediction value is 750 lux. The sigmoid function maps the light change of -50 lux to a weight adjustment coefficient of 0.8, the critical task weight is adjusted from 0.7 to 0.56, and the non-critical task weight is adjusted from 0.3 to 0.44. The time series prediction algorithm predicts the energy demand for the next 4 hours to be 2000mWh based on the energy consumption data of the past 24 hours. The dynamic programming algorithm iterates and calculates in 10,000 state spaces to obtain the optimal energy allocation plan: 1200mWh for critical tasks, 600mWh for non-critical tasks, and 200mWh as a reserve. The five adjacent devices in the wireless communication network elect device 1 as the leader through the Raft algorithm, and the average energy level is calculated to be 4000mWh. The energy level of device 3 is 3000mWh, which is lower than 80% of the average value, triggering energy sharing. With device 2 as a relay, device 1 transmits 500mWh of energy to device 3. The RF energy harvesting circuit converts the 2.4GHz wireless signal into direct current to power the passive temperature sensor, realizing data collection and transmission every 5 minutes.The local anomaly factor algorithm detected that the light intensity dropped suddenly from 800 lux to 200 lux within 1 minute, triggering the emergency lighting system.

[0091] The technical solution of the present invention discretizes the energy supply prediction curve and the energy consumption prediction curve by using a trapezoidal integration method to obtain the energy surplus or energy shortage in each time period; according to the energy surplus or energy shortage, a sliding window mechanism is set to calculate the cumulative energy surplus of continuous preset vision, and if the cumulative energy surplus is lower than the preset proportion of the total power of the equipment, it is determined that there is a risk of power deficit; the current task list and its priority information are obtained, and the tasks are classified according to the preset threshold rules to obtain key monitoring tasks, important tasks and non-critical tasks; the task execution order is rearranged through a dynamic programming algorithm; according to the task execution order, the working frequency and power consumption of the equipment are adjusted until the energy surplus in each time period is greater than the preset safety threshold, and the task queue and equipment working mode configuration are updated.

[0092] The present invention establishes a light intensity prediction model by analyzing the geographical location of the equipment, meteorological data and the trajectory of the sun, and optimizes the solar energy conversion efficiency in combination with the equipment installation parameters. At the same time, an energy consumption model is established according to the equipment status and task plan. By comparing the energy supply and consumption prediction curves, the risk of power deficit is judged, and the task scheduling strategy is dynamically adjusted. When the energy is sufficient, a hierarchical mechanism is used to balance the execution of critical and non-critical tasks. In addition, the present invention also considers the impact of factors such as cloud cover on the prediction model, and introduces an energy sharing mechanism between devices. This method can effectively improve the energy utilization of edge devices, ensure the execution of critical tasks, and realize passive wireless real-time monitoring, which is suitable for scenarios such as the power Internet of Things.

[0093] Embodiment 3

[0094] Figure 3 This is a schematic diagram of the structure of an energy-consuming equipment operation adjustment device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device comprises:

[0095] The parameter acquisition module 301 is used to obtain regional meteorological parameters, equipment installation parameters and equipment operation parameters corresponding to the edge energy equipment;

[0096] The energy prediction module 302 is used to determine the energy supply prediction curve and the energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training;

[0097] A deficit determination module 303, for determining whether an energy-consuming device has an energy deficit risk based on the energy supply prediction curve and the energy consumption prediction curve;

[0098] The operation adjustment module 304 is used to adjust the task operation of the energy-consuming equipment according to the task priority level when it is determined that the energy-consuming equipment has an energy deficit risk.

[0099] Optionally, the device further includes:

[0100] The power generation model training module is used to organize the regional meteorological parameters into sample meteorological parameters; train the preset light intensity model according to the sample meteorological parameters and regional geographical parameters to obtain a light intensity prediction model; obtain a light intensity prediction curve according to the regional meteorological parameters and the light intensity prediction model; and obtain an equipment power generation model according to the light intensity prediction curve, the equipment installation parameters and the preset power generation model.

[0101] Optionally, the device further includes:

[0102] The consumption training module is used to organize the equipment operation parameters into sample operation parameters; according to the sample operation parameters and the historical energy consumption data corresponding to the energy-consuming equipment, the preset consumption model is trained to obtain the energy consumption model.

[0103] Optional, energy prediction module, specifically used for:

[0104] Inputting the regional environmental parameters corresponding to the edge energy equipment and the equipment installation parameters into the equipment power generation model to perform power generation prediction processing to obtain an energy supply prediction curve;

[0105] The equipment operating parameters and the historical energy consumption data corresponding to the energy-consuming equipment are input into the energy consumption model to perform energy consumption prediction processing to obtain an energy consumption prediction curve.

[0106] Optionally, the deficit determination module 303 is specifically configured to:

[0107] Based on the trapezoidal integration method, the energy supply prediction curve and the energy consumption prediction curve are discretized to obtain the time period energy result of each time period;

[0108] Determine the accumulated energy surplus within the preset time period according to the energy result of the time period;

[0109] It is determined whether the energy consuming equipment has an energy deficit risk based on the accumulated energy surplus and the total energy consumption of the energy consuming equipment.

[0110] Optionally, the operation adjustment module 304 is specifically configured to:

[0111] Determine the task priority of equipment operation tasks;

[0112] Based on a dynamic programming algorithm, determining the task execution order corresponding to the standby running task according to the task priority level;

[0113] Based on the task execution sequence, the working state of the energy-consuming equipment is adjusted so that the energy-consuming equipment does not have the risk of energy deficit.

[0114] Optionally, the operation adjustment module is also used to:

[0115] Determine the predicted light intensity based on the regional predicted meteorological parameters corresponding to the edge energy device and the pre-trained light intensity prediction model;

[0116] Based on a multi-objective genetic algorithm, the task execution time and the task execution frequency of the energy-consuming equipment are adjusted according to the predicted light intensity.

[0117] The energy-consuming equipment operation adjustment device provided in the embodiment of the present invention can execute the energy-consuming equipment operation adjustment method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0118] Embodiment 4

[0119] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0120] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0121] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0122] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as the method for regulating the operation of energy-consuming equipment.

[0123] In some embodiments, the method of energy consumption device operation adjustment can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method of energy consumption device operation adjustment described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the method of energy consumption device operation adjustment in any other appropriate manner (for example, by means of firmware).

[0124] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0125] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0126] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0127] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0128] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0129] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0130] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0131] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for regulating the operation of energy-consuming equipment, characterized in that: include: Obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment; Determine an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training; Based on the energy supply prediction curve and the energy consumption prediction curve, determining whether the energy consuming equipment has an energy deficit risk; When it is determined that the energy-consuming equipment has an energy deficit risk, the task operation of the energy-consuming equipment is adjusted according to the task priority level.

2. The method according to claim 1, characterized in that The training process of the equipment power generation model includes: Arranging the meteorological parameters of the region into sample meteorological parameters; According to the sample meteorological parameters and regional geographical parameters, a preset light intensity model is trained to obtain a light intensity prediction model; Obtaining a light intensity prediction curve according to the regional meteorological parameters and the light intensity prediction model; A device power generation model is obtained according to the light intensity prediction curve, the device installation parameters and the preset power generation model.

3. The method according to claim 1, characterized in that The training process of the energy consumption model includes: Arranging the equipment operating parameters into sample operating parameters; According to the sample operating parameters and the historical energy consumption data corresponding to the energy consuming equipment, the preset consumption model is trained to obtain an energy consumption model.

4. The method according to claim 1, characterized in that: The step of determining the energy supply prediction curve and the energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training includes: Inputting the regional environmental parameters corresponding to the edge energy equipment and the equipment installation parameters into the equipment power generation model to perform power generation prediction processing to obtain an energy supply prediction curve; The equipment operating parameters and the historical energy consumption data corresponding to the energy-consuming equipment are input into the energy consumption model to perform energy consumption prediction processing to obtain an energy consumption prediction curve.

5. The method according to claim 1, characterized in that The determining whether the energy consumption equipment has an energy deficit risk based on the energy supply prediction curve and the energy consumption prediction curve includes: Based on the trapezoidal integration method, the energy supply prediction curve and the energy consumption prediction curve are discretized to obtain the time period energy result of each time period; Determine the accumulated energy surplus within the preset time period according to the energy result of the time period; It is determined whether the energy consuming equipment has an energy deficit risk based on the accumulated energy surplus and the total energy consumption of the energy consuming equipment.

6. The method according to claim 1, characterized in that The step of adjusting the task processing of the energy-consuming device according to the task priority level includes: Determine the task priority of equipment operation tasks; Based on a dynamic programming algorithm, determining the task execution order corresponding to the standby running task according to the task priority level; Based on the task execution sequence, the working state of the energy-consuming equipment is adjusted so that the energy-consuming equipment does not have the risk of energy deficit.

7. The method according to claim 1, characterized in that After adjusting the task processing of the energy-consuming equipment according to the task priority level, the method further includes: Determine the predicted light intensity based on the regional predicted meteorological parameters corresponding to the edge energy device and the pre-trained light intensity prediction model; Based on a multi-objective genetic algorithm, the task execution time and the task execution frequency of the energy-consuming equipment are adjusted according to the predicted light intensity.

8. An energy-consuming equipment operation regulating device, characterized in that: include: A parameter acquisition module is used to obtain regional meteorological parameters, equipment installation parameters, and equipment operation parameters corresponding to edge energy equipment; An energy prediction module, used to determine an energy supply prediction curve and an energy consumption prediction curve according to the regional meteorological parameters, the equipment installation parameters, the equipment operation parameters, and the equipment power generation model and energy consumption model obtained by pre-training; A deficit determination module, used to determine whether there is an energy deficit risk for energy-consuming equipment based on the energy supply prediction curve and the energy consumption prediction curve; The operation adjustment module is used to adjust the task operation of the energy-consuming equipment according to the task priority level when it is determined that the energy-consuming equipment has the risk of energy deficit.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the energy-consuming equipment operation adjustment method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the energy-consuming equipment operation adjustment method according to any one of claims 1 to 7 when executed.

Citation Information

Cited By

  • Self-adaptive dormancy scheduling method for wireless sensing node powered by solar energy

    CN120343687A

  • Basin flood disaster early prediction method based on machine learning

    CN120409822A

  • A machine learning-based early prediction method for flood disasters in a watershed

    CN120409822B

  • Energy consumption dynamic management method in renewable resource production process

    CN121212756A