A park electric energy intelligent management regulation and control method based on an internet of things

Abstract

This invention discloses a smart management and control method for park electricity based on the Internet of Things (IoT), belonging to the field of community power grid construction platform technology. The steps for managing and controlling park electricity consumption through a management and control platform include: first, installing smart meters to deploy a data acquisition and monitoring system; second, training a load prediction model using historical load data; and finally, predicting future time-of-use electricity consumption by residents based on the trained load prediction model. Simultaneously, a benchmark electricity price model is established based on future weather forecast data and through linear regression to predict future time-of-use benchmark electricity prices. Once all basic data is prepared, a park electricity supply and demand model is constructed using linear programming to minimize total electricity costs. In subsequent power supply management, manual intervention steps are added, and visualization components are used to display each resident's electricity consumption percentage, cost analysis, and energy storage status. Thresholds are set; when the detected deviation exceeds the threshold, administrator intervention is prompted.

Description

Technical Field

[0001] The invention relates to the field of community power grid construction platform technology, specifically to a smart management and control method for park power based on the Internet of Things. Background Technology

[0002] With the continuous development of the Internet, artificial intelligence, and green energy-saving industries, power grid construction is increasingly moving towards intensification, modularization, and integration. Many communities or villages are building independent renewable energy power supply facilities, such as wind power, photovoltaics, and energy storage devices. In some industrial parks, there are not only manufacturing enterprises but also many related service entities. Therefore, in order to manage the various enterprises and entities within the community in a unified manner, it is necessary to ensure the safety and stability of their electricity use in order to minimize electricity costs and reduce the pressure on the power grid.

[0003] However, due to the complex electricity usage conditions of the park's owners, it is not yet possible to effectively reduce electricity costs. In view of this, the applicant proposes a new technical solution that combines large-scale model technology with a manual intervention mechanism to regulate the park's electricity management system. This will ensure balanced electricity usage among the park's owners, minimize the park's electricity costs, and allow for timely allocation of electricity to new owners within the park. Summary of the Invention

[0004] To address this issue, an Internet of Things-based intelligent management and control method for park power is provided to solve the problem of excessively high electricity costs caused by the complex electricity usage of park owners in existing technologies.

[0005] To achieve the above objectives, the invention provides the following technical solution:

[0006] This invention discloses an IoT-based intelligent management and control method for power supply in industrial parks. The control targets include photovoltaic power stations and energy storage power stations within the park, including:

[0007] Establish electricity usage tags for property owners and collect historical load data based on these tags. Then, use this historical load data to train a load prediction model. Finally, based on the trained load prediction model, predict the time-of-use electricity consumption of property owners in the park during a specific time period t within the next 24 hours.

[0008] Calculate the time-of-use output power of the photovoltaic power station based on the weather forecast data for the next 24 hours.

[0009] A benchmark electricity price model is established using linear regression to predict the time-of-use benchmark electricity price for the next 24 hours.

[0010] A linear programming approach is used to construct a power supply and demand model for the industrial park to minimize the total electricity cost. The applicable formula is:

[0011]

[0012] in, Let C be the base electricity price at time t. pu (t) represents the cost of photovoltaic power generation at time t, C ess For energy storage cycle costs, P dis (t)-P ch (t) represents the difference in energy storage discharge / charge power during time period t, and constraints are set based on this.

[0013] Set administrator permissions and allow administrators to manually set priorities to adjust the charging and discharging plans of the energy storage power station. Based on this, the system will be optimized every 15 minutes to respond to photovoltaic fluctuations or load changes. After a new owner is connected, the load prediction model will be initialized to allocate temporary electricity quotas and then be gradually optimized in the future.

[0014] The system uses visualization components to display each owner's electricity consumption percentage, cost analysis, and energy storage status. It also sets thresholds, and prompts administrators to intervene when the detected deviation exceeds the threshold.

[0015] Furthermore, the constraints include power balance constraints, energy storage constraints, and fairness constraints;

[0016] The power balance constraint formula is:

[0017]

[0018] Let be the load of the i-th owner during time period t;

[0019] The energy storage constraints are as follows:

[0020] SOC min ≤SOC(t)≤SOC max ;

[0021]

[0022] The fairness constraint formula is as follows:

[0023]

[0024] Where SOC(t) is the state of charge of the energy storage during time period t, E ess Total energy storage capacity;

[0025] δ is the allowable deviation threshold, P alloc,i (t) represents the power allocated to the i-th owner during time period t, Pavg,i Let η be the historical average electricity consumption percentage of the i-th homeowner. ch Energy storage charging conversion efficiency, η dis This represents the energy storage discharge conversion rate.

[0026] Furthermore, the steps for training the load prediction model include:

[0027] The historical load data is preprocessed and key features are extracted, including hourly features H(t), weekday features D(t), weather features W(t), and historical load H. hist (t-24t);

[0028] Establish a linear regression model for load forecasting and apply the formula:

[0029]

[0030] Wherein, β0 is the intercept term, β1 is the hourly coefficient, β2 is the weekday coefficient, β3 is the weather coefficient, and β4 is the historical load coefficient;

[0031] The results are fitted using the least squares method, and a regularization term is added to prevent overfitting. The loss function formula is as follows:

[0032]

[0033] Where λ is the regularization strength coefficient, the regularization term is used to limit the complexity of the model, thereby improving the model's generalization ability on unknown data.

[0034] Furthermore, the output power of the photovoltaic power station The calculation process includes:

[0035] Based on future weather forecasts, the predicted light intensity G(t) for time period t;

[0036] A photovoltaic power generation prediction model is established, using the following formula:

[0037] Where A is the area of ​​the photovoltaic panel, η pv For photoelectric conversion efficiency, T amb (t) represents the ambient temperature during time period t;

[0038] The load forecasting error is corrected to obtain the historical load for the same period.

[0039] Furthermore, a benchmark electricity price model is established, and the benchmark electricity price is predicted using the benchmark electricity price model. The process includes:

[0040] Regression coefficients were fitted using historical data, including the base electricity price coefficient. Temperature coefficient Supply and demand elasticity coefficient Fuel cost coefficient and renewable energy coefficient

[0041] A linear regression model for the benchmark electricity price is established, using the following formula:

[0042]

[0043] Among them, R supply / demand (t) represents the supply-demand ratio, C fuel (t) Fuel cost, R renew (t) Cost of renewable energy.

[0044] Furthermore, the administrator can manually set the priority by following these steps:

[0045] Step 1: Login security verification. The administrator logs in to the intervention interface through two-factor authentication, and the operation records are automatically archived for auditing.

[0046] Step 2: Select the intervention mode, which includes parameter coverage, weight adjustment and rule optimization skip. When temporarily adjusting the quota of a single owner, select the parameter coverage mode. When it is necessary to modify the priority of the optimization algorithm, select the weight adjustment mode. When there is an emergency power outage, select the rule optimization skip mode.

[0047] Step 3: The system simulates the supply and demand balance and cost changes after the intervention in real time, and generates a comparative report;

[0048] Step 4: After manual confirmation, the system immediately implements the new strategy and highlights the intervention area on the visualization component. Then, it continuously monitors the deviation between the actual load and the intervention target. If the deviation exceeds the threshold, it will issue another alarm.

[0049] Furthermore, in step 2, when the weight adjustment mode is selected, the load curve for the higher priority part can be adjusted by graphical dragging, and the input expression can be used to quantify the adjustment of the intervention parameters.

[0050] Furthermore, in step 1, the administrator sets hierarchical permissions, which include:

[0051] Operations and maintenance engineers have the authority to select parameter coverage modes and adjust the energy storage load ratio;

[0052] System administrator privileges allow modification of optimization algorithm weights and selection of parameter overriding modes;

[0053] With super administrator privileges, all intervention modes can be selected. Among them, the reason for intervention must be entered when implementing critical operations, and for operations involving grid power purchases exceeding 500kW, it is mandatory to upload approval documents.

[0054] Furthermore, the visualization component includes:

[0055] A circular dashboard displays the current self-sufficiency rate of the park, which is the ratio of the sum of photovoltaic and energy storage power to the total load.

[0056] A dynamic flow graph displays the current flow direction in real time, which is the flow direction between the power grid, the park, photovoltaic, load, and energy storage charging and discharging.

[0057] The circular dashboard and dynamic flow chart are both displayed on a digital dashboard, which also displays the owner's load allocation matrix.

[0058] Furthermore, the super administrator privileges allow entry into a fully manual mode by inputting a dynamic verification code, thereby enabling free allocation of power supply priorities.

[0059] The invention has the following advantages:

[0060] This invention discloses an IoT-based intelligent management and control method for park electricity. It utilizes smart meters to collect data such as electricity consumption, power factor, and load curves, while simultaneously collecting data on the output power and solar irradiance of photovoltaic power plants and the charging and discharging power limits of energy storage power plants. With comprehensive data collection, it employs large-scale artificial intelligence modeling technology to predict grid electricity costs, generator-side power supply, and owner electricity loads. Based on this prediction, linear programming is used to model and comprehensively determine the lowest electricity cost. Compared to existing technologies, the technical solution disclosed in this invention integrates multiple factors, meets the electricity needs of parks, and features fast response, low computational cost, and the ability to effectively combine large-scale modeling technology with human intervention mechanisms to regulate park electricity management. This ensures balanced electricity consumption among owners within the park and guarantees the lowest possible electricity cost. Attached Figure Description

[0061] To more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0062] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size should still fall within the scope of the technical content disclosed in the invention, provided that they do not affect the effectiveness and purpose that the invention can achieve.

[0063] Figure 1 A control flowchart for an IoT-based smart management and control method for park power supply provided for the invention;

[0064] Figure 2 Flowchart of load prediction model training provided for the invention;

[0065] Figure 3 A flowchart of manual intervention provided for the invention;

[0066] Figure 4 Electricity price forecast data table provided for the invention. Detailed Implementation

[0067] The following specific embodiments illustrate the implementation of the invention. Those skilled in the art can easily understand other advantages and effects of the invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the invention. Based on the embodiments of the invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the invention.

[0068] Please refer to this as well. Figures 1-4 This invention discloses an IoT-based smart management and control method for park power. The control targets include photovoltaic power stations and energy storage power stations in the park. It mainly combines large-scale model technology with manual intervention mechanism to ensure balanced power consumption among owners in the park and to ensure the lowest power cost in the park. At the same time, it can promptly allocate power to new owners in the park.

[0069] The technical solution disclosed in this invention will be described below with specific embodiments, and based on this, the detailed implementation steps, related calculation formulas, and parameters of the park's electricity management and control system will be explained. First, smart meters need to be installed to deploy a data acquisition and monitoring system. Specifically, this involves establishing resident electricity usage tags and collecting historical load data from residents based on these tags. Then, the historical load data is used to train a load prediction model. Finally, based on the trained load prediction model, the time-of-use electricity consumption of park residents will be predicted for the next 24 hours. Simultaneously, the output power of the photovoltaic power station is calculated based on the weather forecast data for the next 24 hours. Simultaneously, a benchmark electricity price model is established using linear regression, and this model is used to predict the time-of-use benchmark electricity price for the next 24 hours.

[0070] Then, after all the basic data is prepared, a power supply and demand model for the park is constructed using linear programming to minimize the total electricity cost. The specific application formula is as follows:

[0071]

[0072] in, Let C be the base electricity price at time t. pu (t) represents the cost of photovoltaic power generation at time t, C ess For energy storage cycle costs, P dis (t)-P ch (t) represents the difference in energy storage discharge / charge power during time period t.

[0073] Finally, in subsequent power supply management, a manual intervention step is added. Specifically, administrator permissions need to be set, and administrators need to be allowed to manually set priorities to adjust the charging and discharging plans of the energy storage power station. On this basis, rolling optimization is performed every fifteen minutes to respond to photovoltaic fluctuations or load changes. After a new owner is connected, the load prediction model is initialized to allocate temporary power quotas and then gradually optimize it. During this process, the power consumption ratio, cost analysis and energy storage status of each owner are displayed through visualization components. At the same time, thresholds are set, and when the detected deviation exceeds the threshold, the administrator is prompted to intervene.

[0074] In this embodiment, a linear programming approach is primarily used to construct the power supply and demand model for the industrial park. The marginal constraints of the model include power balance constraints, energy storage constraints, and fairness constraints. The power balance constraint formula is as follows:

[0075]

[0076] Let N be the load of the i-th owner during time period t, and N be the number of owners.

[0077] The energy storage constraints are:

[0078] SOC min ≤SOC(t)≤SOC max ;

[0079]

[0080] Where SOC(t) is the state of charge of the energy storage during time period t, E ess Total energy storage capacity;

[0081] The formula for fairness constraints is:

[0082]

[0083] δ is the allowable deviation threshold, P alloc,i (t) represents the power allocated to the i-th owner during time period t, P avg,i The historical average electricity consumption percentage of the i-th homeowner;

[0084] It should be noted that the maximum value of δ does not exceed 5%.

[0085] In a specific embodiment of this invention, the steps for training the load prediction model are as follows: First, historical load data needs to be acquired and preprocessed, with time periods ranging from 15 minutes to 1 hour, based on different types of property owners, such as office buildings, factories, and canteens, to determine the actual load power. Then, key features are extracted from the historical load data to obtain features including hourly features H(t), weekday features D(t), weather features W(t), and historical load P. hist Features such as (t-24t) can be categorized into different types. For example, hourly features H(t) can be the hour, day of the week, and whether it is a weekday / holiday, and holidays need to be distinguished between statutory holidays and weekends. Weather features W(t) include data such as temperature, humidity, and light intensity. These data are strongly correlated with the temperature sensitivity of factory loads and the air conditioning load of office buildings.

[0086] Then, with sufficient data prepared, the features are formalized to establish a linear regression model for load forecasting, and the formula is applied:

[0087]

[0088] Where β0 is the intercept term, β1 is the hourly coefficient, β2 is the weekday coefficient, β3 is the weather coefficient, and β4 is the historical load coefficient. After obtaining the preliminary results, the least squares method needs to be used to fit the results, and a regularization term needs to be added to prevent overfitting. The loss function formula is as follows:

[0089]

[0090] Among them, P real (t) represents the actual load power of the owner. By calculating the loss function L, the gap between the actual result and the predicted result can be represented, so that the predicted result is closer to the actual result.

[0091] For example, predicting the office building's load at 15:00 on a certain day.

[0092] Input features: H(t) = 15 (3 PM), D(t) = 2 (Tuesday), W(t) = 28℃ (temperature), P hist (t-24t)=320kW (load at 15:00 yesterday)

[0093] Calculation process:

[0094] In some embodiments, the output power of a photovoltaic power station The calculation process mainly includes: first, establishing a photovoltaic power generation prediction model; then, based on future weather forecasts, predicting the solar irradiance G(t) for time period t, using the following formula:

[0095]

[0096] Where A is the area of ​​the photovoltaic panel, η pv For photoelectric conversion efficiency, T amb (t) represents the ambient temperature during time period t. Finally, the load prediction error is corrected to obtain the historical load for the same period.

[0097] The formula for correcting load forecasting errors is:

[0098]

[0099] α is the smoothing coefficient, P hist,i (t) represents the historical load for the same period.

[0100] In a specific embodiment of the present invention, the process of establishing a benchmark electricity price model using linear programming and predicting the benchmark electricity price using the benchmark electricity price model mainly includes:

[0101] Regression coefficients were fitted using historical data, including the base electricity price coefficient. Temperature coefficient Supply and demand elasticity coefficient Fuel cost coefficient and renewable energy coefficient Based on this, a linear regression model for the benchmark electricity price is established, using the following formula:

[0102]

[0103] Among them, R supply / demand (t) represents the supply-demand ratio, C fuel (t) Fuel cost, R renew (t) Renewable energy costs. Specifically, R... supply / demand (t) represents the ratio of regional electricity supply to demand, C fuel (t) Fuel costs are the price fluctuations of power generation fuels such as coal and natural gas. It should be noted that this model is primarily suitable for short-term forecasts within 24 hours; please refer to [reference needed] for detailed calculations. Figure 4 Furthermore, in the process of using the model to calculate and predict electricity prices, the accuracy of the model is improved through dynamic correction and error processing.

[0104] In a specific embodiment of this invention, the power management system for the park is mainly managed through manual intervention to ensure the flexibility and reliability of power supply. This manual intervention can cover multiple dimensions, including strategy adjustment, emergency response, and data correction. The detailed intervention steps, scenario classifications, and operation procedures are as follows: The administrator manually sets priorities, including: Step 1: Login security verification. The administrator logs into the intervention interface through two-factor authentication, and operation records are automatically archived for auditing; Step 2: Selecting an intervention mode. Intervention modes include parameter coverage, weight adjustment, and rule optimization skip. When temporarily adjusting a single owner's quota, the parameter coverage mode is selected; when modifying the priority of the optimization algorithm, the weight adjustment mode is selected; and when an emergency power outage occurs, the rule optimization skip mode is selected; Step 3: The system simulates the supply and demand balance and cost changes after the intervention in real time, generating a comparison report; Step 4: After manual confirmation, the system immediately executes the new strategy and highlights the intervention area on the visualization component. It then continuously monitors the deviation between the actual load and the intervention target. If the deviation exceeds a threshold, an alarm is triggered again.

[0105] In this embodiment, the administrator settings for hierarchical permissions in step 1 include: Operations engineer permissions, allowing selection of parameter coverage modes and adjustment of energy storage load ratios; system administrator permissions, allowing modification of optimization algorithm weights and selection of parameter coverage modes; and super administrator permissions, allowing selection of a full intervention mode, where implementing critical operations requires inputting the intervention reason, and operations involving grid power purchases exceeding 500kW require mandatory uploading of approval documents. The super administrator can also enter a fully manual mode by inputting a dynamic verification code, thereby freely allocating power supply priorities.

[0106] In this embodiment, when selecting the weight adjustment mode in step 2, for the higher priority part, the load curve can be adjusted by graphical dragging on the visualization component, and the intervention parameters can be quantitatively adjusted by inputting expressions.

[0107] Building upon this foundation, the visualization components include a circular dashboard and a dynamic flow graph. The circular dashboard displays the current self-sufficiency rate of the industrial park, which is the ratio of the sum of photovoltaic and energy storage power to the total load. Both the circular dashboard and the dynamic flow graph are displayed on a digital dashboard, which also shows the owner's load allocation matrix. The dynamic flow graph displays the current flow direction in real time, showing the relationship between the grid, the industrial park, photovoltaic systems, loads, and energy storage charging and discharging.

[0108] Although the invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, such modifications or improvements made without departing from the spirit of the invention are all within the scope of the claims.

Claims

1. A smart management and control method for power supply in a park based on the Internet of Things (IoT), wherein the control targets include photovoltaic power stations and energy storage power stations within the park, characterized in that, Includes the following steps: The process involves establishing electricity usage tags for property owners, collecting historical load data based on these tags, training a load forecasting model using this historical load data, and finally using the trained load forecasting model to predict a specific time period within the next 24 hours. Time-of-use electricity consumption of park owners ; Calculate the time-of-use output power of the photovoltaic power station based on the weather forecast data for the next 24 hours. ; A benchmark electricity price model is established using linear regression to predict the time-of-use benchmark electricity price for the next 24 hours. ; A linear programming approach is used to construct a power supply and demand model for the industrial park to minimize the total electricity cost. The applicable formula is: ； in, The base electricity price at time t. Let t be the cost of photovoltaic power generation. For energy storage cycle costs, Let t be the difference between the energy storage discharge / charge power during time period t, and set constraints based on this; Set administrator permissions and allow administrators to manually set priorities to adjust the charging and discharging plans of the energy storage power station. Based on this, the system will perform rolling optimization every 15 minutes to respond to photovoltaic fluctuations or load changes. After a new owner is connected, the load prediction model will be initialized to allocate temporary electricity quotas and then be gradually optimized in the future. The system uses visualization components to display the electricity consumption ratio, cost analysis, and energy storage status of each property owner. It also sets thresholds, and prompts the administrator to intervene when the detected deviation exceeds the threshold. The constraints include power balance constraints, energy storage constraints, and fairness constraints; The power balance constraint formula is: ； Let be the load of the i-th owner during time period t; The energy storage constraints are as follows: ； ； The fairness constraint formula is as follows: ； in, The state of charge (SOC) of the energy storage during time period t. min For minimum energy storage load, SOC max For the maximum energy storage load, Total energy storage capacity; The allowable deviation threshold, Let be the power allocated to the i-th owner during time period t. The historical average electricity consumption percentage of the i-th homeowner. Energy storage charging conversion rate, This represents the energy storage discharge conversion rate.

2. The IoT-based intelligent management and control method for park power as described in claim 1, characterized in that, The steps for training the load prediction model include: The historical load data is preprocessed and key features, including hourly features, are extracted. Weekly characteristics Weather characteristics and historical load ; Establish a linear regression model for load forecasting and apply the formula: ； in, For the intercept term, Hourly coefficient, Weekly coefficient, Weather coefficient, Historical load factor; The results are fitted using the least squares method, and regularization is used to prevent overfitting. The loss function formula is as follows: ； in, This represents the actual load power of the property owner.

3. The IoT-based intelligent management and control method for park power as described in claim 1, characterized in that, The output power of the photovoltaic power station The calculation process includes: Based on future weather forecasts, the predicted light intensity for time period t is... ; A photovoltaic power generation prediction model is established, using the following formula: Where A is the area of ​​the photovoltaic panel. For photoelectric conversion efficiency, The ambient temperature during time period t; The load forecasting error is corrected to obtain the historical load for the same period.

4. The IoT-based intelligent management and control method for park power as described in claim 3, characterized in that, Establish a benchmark electricity price model and predict the benchmark electricity price using the benchmark electricity price model. The process includes: Regression coefficients were fitted using historical data, including the base electricity price coefficient. Temperature coefficient Supply and demand elasticity coefficient Fuel cost coefficient and renewable energy coefficient ; A linear regression model for the benchmark electricity price is established, using the following formula: ; in, For supply and demand ratio, Fuel costs, Cost of renewable energy.

5. The IoT-based intelligent management and control method for park power as described in claim 1, characterized in that, The steps for an administrator to manually set priorities are as follows: Step 1: Login security verification. Administrators log in to the intervention interface through two-factor authentication, and operation records are automatically archived for auditing. Step 2: Select the intervention mode, which includes parameter coverage, weight adjustment and rule optimization skip. When temporarily adjusting the quota of a single owner, select the parameter coverage mode. When it is necessary to modify the priority of the optimization algorithm, select the weight adjustment mode. When there is an emergency power outage, select the rule optimization skip mode. Step 3: The system simulates the supply and demand balance and cost changes after the intervention in real time, and generates a comparative report; Step 4: After manual confirmation, the system immediately implements the new strategy and highlights the intervention area on the visualization component. Then, it continuously monitors the deviation between the actual load and the intervention target. If the deviation exceeds the threshold, it will issue another alarm.

6. The IoT-based intelligent management and control method for park power as described in claim 5, characterized in that, In step 2, when the weight adjustment mode is selected, the load curve of the higher priority part can be adjusted by graphical dragging, and the input expression can be used to quantify the adjustment of the intervention parameters.

7. The IoT-based intelligent management and control method for park power as described in claim 6, characterized in that, In step 1, the administrator sets hierarchical permissions, which include: Operations and maintenance engineers have the authority to select parameter coverage modes and adjust the energy storage load ratio; System administrator privileges allow modification of optimization algorithm weights and selection of parameter overriding modes; With super administrator privileges, all intervention modes can be selected. Among them, the reason for intervention must be entered when implementing critical operations, and for operations involving grid power purchases exceeding 500kW, it is mandatory to upload approval documents.

8. The IoT-based intelligent management and control method for park power as described in claim 7, characterized in that, The visualization components include: A circular dashboard displays the current self-sufficiency rate of the park, which is the ratio of the sum of photovoltaic and energy storage power to the total load. A dynamic flow graph displays the current flow direction in real time, which is the flow direction between the power grid, the park, photovoltaic, load, and energy storage charging and discharging. The circular dashboard and dynamic flow chart are both displayed on a digital dashboard, which also displays the owner's load allocation matrix.

9. The IoT-based intelligent management and control method for park power as described in claim 8, characterized in that, The super administrator privileges allow entry into a fully manual mode by inputting a dynamic verification code, enabling free allocation of power supply priorities.

Images