Flexible load adjusting method for intelligent flexible control terminal

Abstract

The invention relates to the technical field of power system load regulation, and discloses an intelligent flexible regulation and control terminal flexible load regulation method which comprises the following steps: acquiring operation data of an air conditioning system in real time, and preprocessing the operation data; based on a machine learning algorithm, the preprocessed data are classified and recognized, and the operation state of the air conditioning system is determined; combining historical data and external environment factor data to determine a power demand change trend in a period of time in the future; and the current power grid demand state is obtained, and a corresponding regulation and control strategy is generated and executed in combination with the operation state of the air conditioner system and the power demand change trend so as to control the temperature, the wind speed and the on-off state of the air conditioner system. Flexible adjustment of the load of the air conditioning system is achieved through an intelligent means, the flexibility and efficiency of power grid operation are improved, and safe and stable operation of the power grid is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power system load regulation, and in particular to a flexible load regulation method of an intelligent flexible control terminal. Background Art

[0002] With the continued growth of global energy consumption and rising environmental awareness, the development of smart grid technology is gaining increasing attention. Smart grids achieve efficient operation and optimized management of power systems through the integration of information, communication, and control technologies. Air conditioning systems, as a significant component of electricity consumption, often place significant load on the power grid during peak summer months. Traditional air conditioning control systems mostly use fixed temperature setpoints and on / off times, which cannot be flexibly adjusted to actual power demand. This not only leads to unnecessary energy waste but also increases the burden on the power grid.

[0003] In recent years, the application of machine learning algorithms in prediction and classification has matured, offering new insights into addressing these challenges. Machine learning can more accurately predict changing trends in electricity demand and intelligently control air conditioning systems based on real-time data, ultimately achieving energy savings and reducing consumption. However, few existing air conditioning control solutions combine machine learning with intelligent, flexible control to dynamically identify and precisely control the operating status of air conditioning systems.

[0004] In view of this, a flexible load adjustment method of intelligent flexible control terminal is proposed to achieve flexible load adjustment, thereby ensuring the stable operation of the power grid. Summary of the Invention

[0005] The present invention proposes a flexible load regulation method for an intelligent flexible control terminal, which realizes flexible regulation of the air-conditioning system load through intelligent means, improves the flexibility and efficiency of power grid operation, and ensures the safe and stable operation of the power grid.

[0006] In order to achieve the above objectives, in a first aspect, the present invention provides a method for adjusting flexible load of an intelligent flexible control terminal, comprising:

[0007] Acquire the operating data of the air conditioning system in real time and pre-process the operating data;

[0008] classifying and identifying the pre-processed data based on a machine learning algorithm to determine the operating status of the air conditioning system;

[0009] Combine historical data with external environmental factors to determine the trend of electricity demand changes in the future;

[0010] The current grid demand status is obtained, and combined with the operating status of the air-conditioning system and the power demand change trend, a corresponding control strategy is generated and executed to control the temperature, wind speed and switch status of the air-conditioning system.

[0011] Preferably, the real-time acquisition of the operating data of the air-conditioning system and the pre-processing of the operating data are specifically as follows:

[0012] The intelligent flexible control unit collects the operating data of the air-conditioning system in real time and sends the operating data to the intelligent flexible control terminal via the communication network. The operating data includes power consumption, indoor temperature, indoor humidity, wind speed, startup status and fault alarm information;

[0013] The intelligent flexible control terminal pre-processes the received operation data, including data cleaning and data formatting.

[0014] Preferably, the classifying and identifying the pre-processed data based on a machine learning algorithm to determine the operating status of the air-conditioning system is specifically as follows:

[0015] Inputting the real-time collected and pre-processed operating data into a trained machine learning model;

[0016] The machine learning model is based on a machine learning algorithm and determines and outputs the operating status results of the air-conditioning system according to the input data characteristics. The operating status prediction results include normal operation, mild overload and severe overload.

[0017] Preferably, the combination of historical data and external environmental factor data to determine the trend of power demand changes in the future period is specifically as follows:

[0018] Acquire the historical data and external environmental factor data, and perform data preprocessing and feature extraction, the historical data including the historical operation data of the air conditioning system, historical power grid data, and historical user behavior data, and the external environmental factor data including meteorological data, holiday data, and special event data;

[0019] Through the trained prediction model, based on the extracted data features, the electricity demand change trend in the future period is determined, and the electricity demand change trend includes increasing demand, stable demand and decreasing demand.

[0020] Preferably, the obtaining of the current grid demand state is specifically:

[0021] Obtaining current grid load and available power generation capacity to determine grid demand status;

[0022] If the current grid load is much lower than the available power generation capacity, determining that the grid demand is in a low demand state;

[0023] If the current grid load is close to the available power generation capacity, determining that the grid demand is in a medium demand state;

[0024] If the current grid load exceeds the available power generation capacity, it is determined that the grid demand is in a high demand state.

[0025] Preferably, the current grid demand state is obtained, and a corresponding control strategy is generated and executed in combination with the operating state of the air-conditioning system and the power demand change trend, specifically:

[0026] Dynamically adjust the corresponding weights based on the specific conditions of the grid demand state, the operating state, and the power demand change trend;

[0027] Determine the current comprehensive score based on the average of the adjusted weights;

[0028] The current risk situation is judged based on the comprehensive score, and a corresponding control strategy is generated and executed according to the risk situation.

[0029] Preferably, the corresponding weights are dynamically adjusted based on the specific conditions of the grid demand state, the operating state, and the power demand change trend, specifically as follows:

[0030] When the grid demand states are respectively the low demand state, the medium demand state, and the high demand state, dynamically adjusting the corresponding weights to be the first weight, the second weight, and the third weight;

[0031] When the operating states are respectively the normal operation, the mild overload, and the severe overload, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight;

[0032] When the power demand change trend is the demand increase, the demand stability, and the demand decrease, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight;

[0033] The first weight is smaller than the second weight, and the second weight is smaller than the third weight.

[0034] Preferably, the current risk situation is judged based on the comprehensive score, specifically:

[0035] If the comprehensive score is less than a preset first threshold, it is determined that the current risk situation is low;

[0036] If the comprehensive score is greater than or equal to a preset first threshold and less than a preset second threshold, it is determined that the current risk situation is medium;

[0037] If the comprehensive score is greater than or equal to a preset second threshold, it is determined that the current situation is high-risk.

[0038] Preferably, the generating and executing of corresponding control strategies according to the risk situation is specifically as follows:

[0039] When the risk situation is low, a first control strategy is generated and executed, that is, slightly increasing the set temperature, maintaining the current wind speed, and not enabling periodic on / off control;

[0040] When the risk situation is a medium risk situation, a second control strategy is generated and executed, that is, a moderate increase in the set temperature, a decrease in the wind speed, and a disabling of periodic on / off control;

[0041] When the risk situation is a high-risk situation, a third control strategy is generated and executed, that is, significantly increasing the set temperature, reducing the current wind speed, and enabling periodic on / off control.

[0042] In a second aspect, the present invention provides an intelligent flexible control terminal flexible load regulation system, comprising:

[0043] An acquisition module is used to obtain the operating data of the air conditioning system in real time and pre-process the operating data;

[0044] An analysis module, configured to classify and identify the pre-processed data based on a machine learning algorithm to determine an operating status of the air conditioning system;

[0045] The forecasting module is used to combine historical data and external environmental factor data to determine the trend of electricity demand changes in the future;

[0046] The execution module is used to obtain the current grid demand status, combine the operating status of the air-conditioning system and the power demand change trend, generate and execute corresponding control strategies to control the temperature, wind speed and switch status of the air-conditioning system.

[0047] The present application discloses an intelligent flexible control terminal flexible load regulation method. This method monitors the operating data of the air-conditioning system in real time and uses a machine learning algorithm to classify and identify the data. This method can more accurately understand the operating status of the air-conditioning system, and then take targeted control measures to effectively reduce energy consumption and improve energy efficiency. The present invention can achieve energy-saving goals through intelligent control of the air-conditioning system without affecting user comfort. By comprehensively considering the grid demand status, the operating status of the air-conditioning system, and the trend of power demand changes, the control strategy is dynamically adjusted. The power output of the air-conditioning system can be appropriately limited during peak power consumption periods to avoid grid overload and ensure the safe and stable operation of the grid. Combined with historical data and external environmental factors such as meteorological conditions and holiday arrangements, a trained prediction model is used to determine the trend of future power demand changes, thereby improving the accuracy of future power demand forecasts and making control decisions more scientific and reasonable. The entire control process is highly intelligent and automated, and can complete the process from data collection, analysis, prediction to the execution of the final control strategy without human intervention, greatly improving work efficiency and reducing operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0049] Figure 1 A flowchart of a method for adjusting flexible load of an intelligent flexible control terminal provided by an embodiment of the present invention;

[0050] Figure 2 A schematic structural diagram of an intelligent flexible control terminal flexible load regulation system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0052] like Figure 1As shown, in some embodiments of the present application, this embodiment provides a method for adjusting flexible load of an intelligent flexible control terminal. Specifically, the method includes the following steps:

[0053] Step S101: Acquire operating data of the air-conditioning system in real time and pre-process the operating data.

[0054] As mentioned above, various sensors and monitoring devices, such as temperature sensors, humidity sensors, and power meters, collect data from the air conditioning system reflecting its current operating status. This includes, but is not limited to, the air conditioning system's power consumption, indoor temperature, indoor humidity, wind speed, power-on status, and fault alarm information. To ensure the timeliness and accuracy of the control strategy, data collection should be as frequent as possible to capture subtle changes in the air conditioning system's operating status and ensure real-time data. Preprocessing refers to a series of processing steps performed on the raw collected data before it is used for further analysis to improve data quality and usability. These steps include data cleaning (noise removal, missing value processing), data formatting (unifying data formats, unit conversion), and data normalization (adjusting values to a specific range). The purpose of preprocessing is to eliminate outliers and inconsistencies in the data to ensure that the data used for subsequent analysis and machine learning model training is accurate and reliable. In addition, preprocessing can reduce the amount of data, improve computing efficiency, and make subsequent processing more efficient.

[0055] For example, imagine a smart office building with multiple air conditioning systems installed, each equipped with a variety of sensors. Each air conditioner's temperature sensor records indoor temperature changes every 10 seconds, its humidity sensor records humidity every minute, and a power meter continuously monitors the air conditioner's energy consumption. All data is transmitted to a central server via wireless communication modules. When certain temperature sensors occasionally display abnormal readings, such as suddenly displaying extremely low or high temperatures, these are identified as outliers. These outliers can be identified and eliminated by setting appropriate thresholds or corrected using smoothing algorithms (such as moving averages).

[0056] It should be noted that in a specific implementation scenario, a multi-device integration solution can be adopted on the basis of the above solution, that is, it is not only applicable to a single air-conditioning system, but can also be expanded to a comprehensive energy management system that integrates multiple different types of equipment. For example, in addition to air conditioning, it can also include lighting systems, hot water supply systems, etc., through a unified data collection and preprocessing platform, to achieve comprehensive energy efficiency management of the entire building or area; adopt a local preprocessing solution, that is, in order to reduce the load of the central server and improve the response speed, edge computing nodes can be deployed close to the data source (such as inside the air-conditioning controller). These nodes can complete preliminary data preprocessing tasks locally, such as simple filtering and aggregation, and then send the processed results to the central server. This not only reduces network bandwidth usage, but also improves the overall performance of the system. The above optional solutions all fall within the scope of protection of this application.

[0057] Step S102: classify and identify the pre-processed data based on a machine learning algorithm to determine the operating status of the air-conditioning system.

[0058] As described above, a pre-trained machine learning model is used to analyze pre-processed air conditioning system operating data to identify and classify the system's current operating status. By learning patterns and features from a large amount of historical data, the machine learning model automatically identifies different operating states and outputs corresponding classification results. Based on the output of the machine learning model, the air conditioning system's operating status is divided into several categories, such as normal operation, mild overload, and severe overload. Each category corresponds to different operating conditions and potential problems, providing a basis for subsequent control strategies.

[0059] For example, in a smart home environment, multiple smart air conditioning systems are installed, each equipped with various sensors to monitor indoor and outdoor parameters such as temperature, humidity, and wind speed. After preprocessing this data, a trained random forest model is used to classify and identify the operating status of the air conditioning systems. Assume that the model has learned to distinguish between normal operation, mild overload, and severe overload. If the power consumption of an air conditioning system suddenly increases, exceeding a preset threshold, and the rate of change in indoor temperature also accelerates, the model will classify it as "mildly overloaded." The system will then automatically adjust its operating parameters, such as reducing wind speed or slightly increasing the set temperature, to alleviate the overload. If the model further detects that power consumption continues to rise, reaching the severe overload threshold, an alarm will be triggered, notifying the user to check the air conditioning system for problems.

[0060] It should be noted that in a specific implementation scenario, an adaptive threshold adjustment scheme can be adopted on the basis of the above scheme, that is, the threshold used for classification and identification can be dynamically adjusted according to historical data and external environmental factors (such as temperature, humidity, holidays, etc.), so that the model can accurately identify the operating status of the air-conditioning system under different conditions; a multimodal data fusion scheme can be adopted, that is, in addition to the operating data of the air-conditioning system itself, other relevant data sources such as meteorological data, power grid data, user behavior data, etc. can also be integrated to enhance the effect of classification and identification. For example, combined with meteorological information such as outdoor temperature, humidity and wind speed, the actual workload of the air-conditioning system can be more accurately judged. Combined with the real-time electricity price and available power generation capacity of the power grid, the operation strategy of the air-conditioning system can be optimized to achieve energy saving and consumption reduction. The above optional schemes all fall within the scope of protection of this application.

[0061] Step S103: Determine the power demand change trend in the future by combining historical data and external environmental factor data.

[0062] As mentioned above, by integrating historical air conditioning system operating data with various external environmental factors (such as weather conditions, holiday schedules, and special events), we can comprehensively analyze the factors affecting electricity demand. Based on this preprocessed comprehensive data, appropriate forecasting algorithms or models are used to infer the changing trends of electricity demand over a period of time. This allows for the proactive planning of air conditioning system operating strategies and avoids problems caused by imbalances in electricity supply and demand.

[0063] For example, in a smart office building, historical operating data for the air conditioning system shows that between 9:00 AM and 5:00 PM on weekdays, indoor temperatures rise rapidly due to the high concentration of people, leading to a significant increase in air conditioning load. Furthermore, meteorological data indicates that high summer temperatures further exacerbate this phenomenon. Therefore, based on the trained LSTM model and this information, it is possible to predict that the power demand of the air conditioning system will increase significantly during this time period on future weekdays, especially during high summer temperatures. Based on this prediction, the air conditioning system's set temperature can be adjusted in advance on weekends, reducing the frequency of power supply startup to cope with the upcoming high load period.

[0064] It should be noted that in specific implementation scenarios, a cross-platform data integration solution can be adopted based on the above solution. That is, in order to improve the accuracy of the prediction, more types of external data sources can be introduced, such as traffic flow data, social media popularity, economic indicators, etc. These data can be accessed through API interfaces, Web services, etc., and integrated with the operating data and meteorological data of the air conditioning system for analysis. For example, traffic flow data can reflect the flow of people in the surrounding area, which helps to predict the flow of people in shopping malls or office buildings; social media popularity can capture upcoming popular events and help prepare for power demand in advance. A scenario simulation solution can be adopted, that is, scenario simulation technology is introduced to better cope with uncertainty and emergencies. By constructing different hypothetical scenarios (such as extreme weather and sudden large-scale events), possible changes in power demand in the future are simulated and their impact on the air conditioning system and the power grid is evaluated. For example, simulating a sudden rainstorm may cause a sharp drop in the number of people in the shopping mall, thereby affecting the power consumption of the air conditioning system. Based on these simulation results, emergency plans can be formulated in advance to ensure the stable operation of the system. The above optional solutions all fall within the scope of protection of this application.

[0065] Step S104: Obtain the current grid demand state, combine the operating state of the air-conditioning system and the power demand change trend, generate and execute a corresponding control strategy to control the temperature, wind speed and switch state of the air-conditioning system.

[0066] As mentioned above, real-time monitoring and evaluation of the current grid load and available power generation capacity are used to determine the grid's demand status. Based on this grid demand status, a comprehensive analysis, combined with the current operating status of the air conditioning system and future trends in power demand, helps to more accurately assess the current risk situation. Based on this comprehensive analysis, appropriate control strategies are generated and implemented to control the temperature, wind speed, and on / off status of the air conditioning system.

[0067] For example, a smart office building uses an API interface with the power company to obtain real-time grid load and available generation capacity data every five minutes. For example, at 3 p.m. on a given day, the grid load is 800 MW and the available generation capacity is 1000 MW, indicating low grid demand. Based on this information, the system can appropriately relax the air conditioning system's control requirements, allowing it to operate freely within a certain range. For example, on a hot day, the air conditioning system is slightly overloaded most of the time. Furthermore, weather forecasts predict continued temperature increases and rising electricity demand over the next few days. Based on this information, the system determines that the building is currently experiencing a medium-risk situation. Based on this medium-risk situation, the system generates a control strategy, raising the air conditioning set temperature from 24°C to 26°C and adjusting the fan speed from high to medium. This ensures indoor comfort while effectively reducing the air conditioning system's energy consumption and alleviating the burden on the grid.

[0068] It should be noted that in specific implementation scenarios, a real-time adjustment solution can be adopted based on the above solution. That is, in order to allow the control strategy to adapt to the ever-changing grid demand and environmental conditions, a dynamic control mechanism can be introduced to enable the system to adjust the control strategy in real time based on the latest grid data, air conditioner operating status, and external environmental factors. For example, if the grid load suddenly drops, the system can automatically relax the control measures and restore the air conditioning setting to a more comfortable level; conversely, if the load rises sharply, the system can immediately strengthen the control and quickly respond to grid demand. A user behavior modeling solution can be adopted. That is, in order to more accurately formulate control strategies, a user behavior model can be established to analyze the user's energy usage habits and preferences. For example, through smart meters and smart home devices, users' historical electricity usage data can be collected to understand their air conditioning usage in different time periods. Based on this information, users' future behavior patterns can be predicted and the air conditioning system's operating strategy can be adjusted accordingly. The above optional solutions all fall within the scope of protection of this application.

[0069] In some embodiments of the present application, in order to ensure that the received operating data is accurate, complete, and easy to analyze, and to provide a solid foundation for subsequent classification, identification, prediction, and control strategy generation, the real-time acquisition of the operating data of the air-conditioning system and the pre-processing of the operating data are specifically as follows:

[0070] The intelligent flexible control unit collects the operating data of the air-conditioning system in real time and sends the operating data to the intelligent flexible control terminal via the communication network. The operating data includes power consumption, indoor temperature, indoor humidity, wind speed, startup status and fault alarm information;

[0071] The intelligent flexible control terminal pre-processes the received operation data, including data cleaning and data formatting.

[0072] As described above, the intelligent flexible control unit uses various sensors and monitoring devices installed on the air conditioning system to collect various operating data of the air conditioning system in real time, including but not limited to: power consumption data, which records the electrical energy consumed by the air conditioning system during operation; indoor temperature data, which monitors the current temperature of the air conditioning environment; indoor humidity data, which measures the air humidity of the air conditioning environment; wind speed data, which detects the wind speed setting data at the air conditioning outlet; power-on status data, which records whether the air conditioning is on or off; and fault alarm information data, which captures any faults or abnormalities in the air conditioning system and generates corresponding alarm information data. The intelligent flexible control unit collects data at a high frequency (e.g., every minute or shorter) to ensure that it can capture subtle changes in the operating status of the air conditioning system in a timely manner. The collected data is transmitted to the intelligent flexible control terminal via a communication network (such as Wi-Fi, Zigbee, LoRa, etc.).

[0073] After receiving the operating data sent from the intelligent flexible control unit, the intelligent flexible control terminal pre-processes it, including data cleaning and data formatting. Data cleaning includes noise removal, that is, identifying and eliminating abnormal data points caused by sensor failure or signal interference; processing missing values, that is, using interpolation or other reasonable methods to fill in the data that has not been successfully collected in a certain time period to ensure data continuity; smoothing, that is, smoothing data with large fluctuations to reduce the impact of short-term fluctuations on subsequent analysis. Data formatting includes unified data format, that is, converting operating data from different sources and types into a unified format for subsequent processing and analysis; unit conversion, that is, converting data in different units (such as Fahrenheit to Celsius, feet / second to meters / second) to ensure that all data uses consistent measurement units; data normalization, that is, adjusting the values to a specific range (such as between 0 and 1) to improve the training efficiency and accuracy of subsequent machine learning algorithms.

[0074] In some embodiments of the present application, in order to accurately identify the current operating status of the air-conditioning system based on the pre-processed operating data and provide a reliable basis for the subsequent generation of control strategies, the pre-processed data is classified and identified based on a machine learning algorithm to determine the operating status of the air-conditioning system, specifically:

[0075] Inputting the real-time collected and pre-processed operating data into a trained machine learning model;

[0076] The machine learning model is based on a machine learning algorithm and determines and outputs the operating status results of the air-conditioning system according to the input data characteristics. The operating status prediction results include normal operation, mild overload and severe overload.

[0077] As described above, the intelligent flexible control terminal transmits preprocessed operating data to the machine learning model via an API interface or file transfer. Before data input, meaningful features must be extracted from the raw data to improve the model's classification and recognition accuracy. These features can be physical quantities directly obtained from sensor readings (such as temperature, humidity, and power), or statistical or derived quantities calculated from the raw data (such as mean, standard deviation, and rate of change). Upon receiving new operating data, the trained model performs classification and recognition based on the input data features, calculates the probability of each possible operating state, and selects the state with the highest probability as the final classification result. Common operating states include: normal operation, where the air conditioning system operates stably within the design parameters without overload or other abnormalities; mild overload, where certain air conditioning system parameters (such as power consumption and temperature change rate) are slightly above the normal range, but not yet reaching the level of severe overload, posing a potential risk; and severe overload, where the air conditioning system's operating parameters significantly exceed the normal range, potentially causing equipment damage or affecting user comfort, requiring immediate action.

[0078] In some embodiments of the present application, in order to accurately predict the trend of electricity demand changes in the future and provide a reliable basis for the subsequent generation of control strategies, the combination of historical data and external environmental factor data to determine the trend of electricity demand changes in the future is specifically as follows:

[0079] Acquire the historical data and external environmental factor data, and perform data preprocessing and feature extraction, the historical data including the historical operation data of the air conditioning system, historical power grid data, and historical user behavior data, and the external environmental factor data including meteorological data, holiday data, and special event data;

[0080] Through the trained prediction model, based on the extracted data features, the electricity demand change trend in the future period is determined, and the electricity demand change trend includes increasing demand, stable demand and decreasing demand.

[0081] As mentioned above, the system first collects various historical data on the air conditioning system, including but not limited to: historical operating data on the air conditioning system, such as power consumption, indoor temperature, indoor humidity, wind speed, power-on status, and fault alarm information; historical power grid data, such as total grid load, power generation, and backup capacity; and historical user behavior data, such as user air conditioning usage habits, power-on and power-off times, and set temperature changes. The system also collects data on external environmental factors related to power demand, including but not limited to: meteorological data, such as temperature, humidity, wind speed, and rainfall; holiday data, such as statutory holidays, weekends, and other special dates; and special event data, such as promotional events at large shopping malls and meeting schedules in office buildings.

[0082] Data preprocessing and feature extraction are performed on the acquired data. Data preprocessing includes but is not limited to data cleaning, data formatting, and data normalization. Feature extraction includes time series features, that is, extracting time series features from historical data, such as daily, weekly, and monthly change trends, seasonal fluctuations, etc.; statistical features, that is, calculating various statistical quantities such as mean, standard deviation, maximum, minimum, etc. to reflect the distribution of data; derived features, that is, calculating new features based on the original data, such as temperature change rate, power consumption growth rate, etc. These features help to more accurately capture the changing patterns of electricity demand.

[0083] Select an appropriate forecasting model based on the characteristics of the data and the timeframe of the forecast target. Common forecasting models include time series analysis models, machine learning algorithms, and deep learning models. For example, for short-term forecasts (hours to a day), models that can capture time dependencies, such as LSTM, can be used. For long-term forecasts (days to weeks), methods more suitable for processing long time series, such as ARI MA or GBDT, can be used. When the forecast results indicate that electricity demand will increase significantly in the future, the system will determine that demand is increasing; when the forecast results indicate that electricity demand will remain relatively stable in the future, the system will determine that demand is stable; when the forecast results indicate that electricity demand will decrease significantly in the future, the system will determine that demand is decreasing.

[0084] In some embodiments of the present application, in order to accurately obtain the current grid demand state and adopt corresponding control strategies according to actual conditions to ensure the safe and stable operation of the grid, the current grid demand state is obtained as follows:

[0085] Obtaining current grid load and available power generation capacity to determine grid demand status;

[0086] If the current grid load is much lower than the available power generation capacity, determining that the grid demand is in a low demand state;

[0087] If the current grid load is close to the available power generation capacity, determining that the grid demand is in a medium demand state;

[0088] If the current grid load exceeds the available power generation capacity, it is determined that the grid demand is in a high demand state.

[0089] As mentioned above, the system obtains real-time data on current grid load and available generating capacity through an interface with the power company or smart grid platform. Grid load refers to the total electrical energy consumed by all electrical devices in the current grid; available generating capacity refers to the maximum electrical energy output that can be provided by all generators in the current grid. Based on the acquired grid load and available generating capacity data, the system categorizes grid demand into three scenarios: low demand, where the current grid load is significantly lower than the available generating capacity, indicating a relatively relaxed grid with sufficient spare capacity to meet the increased load; medium demand, where the current grid load is close to the available generating capacity, indicating a relatively tense grid, requiring careful management of increased load to avoid overload; and high demand, where the current grid load exceeds the available generating capacity, indicating that the grid is overloaded and at risk of a power outage, requiring immediate action to reduce the load.

[0090] In some embodiments of the present application, in order to ensure that energy utilization efficiency is optimized to the maximum extent possible while satisfying user comfort and to ensure safe and stable operation of the power grid, the current power grid demand state is obtained, and a corresponding control strategy is generated and executed in combination with the operating state of the air conditioning system and the power demand change trend, specifically:

[0091] Dynamically adjust the corresponding weights based on the specific conditions of the grid demand state, the operating state, and the power demand change trend;

[0092] Determine the current comprehensive score based on the average of the adjusted weights;

[0093] The current risk situation is judged based on the comprehensive score, and a corresponding control strategy is generated and executed according to the risk situation.

[0094] As described above, based on the relationship between grid load and available generation capacity, the grid demand state is determined to be low, medium, or high. Based on the classification results of the machine learning model, the air conditioning system's operating state is determined to be normal operation, slightly overloaded, or severely overloaded. Based on the results of the forecasting model, the future trend of electricity demand is determined to be increasing, stable, or decreasing. The corresponding weights are dynamically adjusted. The system dynamically adjusts the initial weights based on the latest real-time data. For example, if the grid load suddenly increases and approaches the available generation capacity, the system will increase the weight of the grid demand state accordingly. If the air conditioning system transitions from normal operation to slightly overloaded, the system will increase the weight of the air conditioning system's operating state accordingly. A comprehensive score is calculated as the average of the grid demand weight, the air conditioning system status weight, and the power demand trend weight: comprehensive score = (grid demand weight + air conditioning system status weight + power demand trend weight) / 3. By comparing the determined comprehensive score with a preset threshold, the current risk situation is determined and a corresponding control strategy is generated to ensure maximum energy efficiency while maintaining user comfort and ensuring safe and stable grid operation.

[0095] In some embodiments of the present application, in order to ensure the scientificity and effectiveness of the control strategy, maximize the optimization of energy utilization efficiency, and ensure the safe and stable operation of the power grid, the corresponding weights are dynamically adjusted based on the specific conditions of the power grid demand state, the operating state, and the power demand change trend, specifically:

[0096] When the grid demand states are respectively the low demand state, the medium demand state, and the high demand state, dynamically adjusting the corresponding weights to be the first weight, the second weight, and the third weight;

[0097] When the operating states are respectively the normal operation, the mild overload, and the severe overload, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight;

[0098] When the power demand change trend is the demand increase, the demand stability, and the demand decrease, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight;

[0099] The first weight is smaller than the second weight, and the second weight is smaller than the third weight.

[0100] As mentioned above, in a low-demand state, the demand from the power grid has little impact on the control strategy, and the air-conditioning system can be allowed to adjust freely within a certain range to improve user comfort. Therefore, it is assigned the lowest first weight. In a medium-demand state, the demand from the power grid has a moderate impact on the control strategy. The system needs to manage the load of the air-conditioning system more carefully and take appropriate control measures. Therefore, it is assigned a medium second weight. In a high-demand state, the demand from the power grid has the greatest impact on the control strategy. The system must take emergency measures to quickly reduce the burden on the power grid and avoid the risk of power outages. Therefore, it is assigned the highest third weight.

[0101] During normal operation, the operation of the air-conditioning system has little impact on the control strategy, and it can be allowed to adjust freely within a certain range, so it is given the lowest first weight; when it is slightly overloaded, the operation of the air-conditioning system has a moderate impact on the control strategy, and the system needs to take appropriate control measures to prevent further overload, so it is given a medium second weight; when it is severely overloaded, the operation of the air-conditioning system has the greatest impact on the control strategy, and the system must take emergency measures to quickly reduce the load, so it is given the highest third weight.

[0102] When demand increases, changes in electricity demand have the greatest impact on the control strategy, and the system needs to take measures in advance to cope with the upcoming high-load period, so it is given the highest third weight; when demand is stable, changes in electricity demand have a moderate impact on the control strategy, and the system can make appropriate adjustments based on the current power grid and air-conditioning system status, so it is given a medium second weight; when demand decreases, changes in electricity demand have a smaller impact on the control strategy, and the system can allow the air-conditioning system to adjust freely within a certain range to improve user comfort, so it is given the lowest first weight.

[0103] In some embodiments of the present application, in order to accurately judge the current risk situation based on the comprehensive score, and generate and execute corresponding control strategies accordingly, to ensure that energy utilization efficiency is optimized to the maximum extent while satisfying user comfort, and to ensure the safe and stable operation of the power grid, the current risk situation is judged based on the comprehensive score, specifically:

[0104] If the comprehensive score is less than a preset first threshold, it is determined that the current risk situation is low;

[0105] If the comprehensive score is greater than or equal to a preset first threshold and less than a preset second threshold, it is determined that the current risk situation is medium;

[0106] If the comprehensive score is greater than or equal to a preset second threshold, it is determined that the current situation is high-risk.

[0107] As described above, the system pre-sets first and second thresholds to distinguish different risk levels. These thresholds can be adjusted based on historical data, grid capacity, air conditioning system characteristics, and other factors to ensure they are appropriate for specific usage scenarios. If the combined score is less than the first threshold, the risk is determined to be low. In this case, the grid load is low, the air conditioning system is operating normally, and the power demand trend is stable or decreasing. Overall, the system's operating pressure is low, allowing the air conditioning system to adjust freely within a certain range to improve user comfort. In this case, the control strategy can be more relaxed, minimizing the impact on users. If the combined score is greater than or equal to the first threshold and less than the second threshold, the risk is determined to be medium. In this case, the grid load is close to the available generating capacity, the air conditioning system may be slightly overloaded, and the power demand trend may be increasing or remaining stable. Overall, the system's operating pressure is moderate, requiring appropriate control measures to avoid further load increases. In this case, the control strategy needs to be more cautious to balance user comfort and energy efficiency. If the comprehensive score is greater than or equal to the second threshold, it is judged to be a high-risk situation. At this time, the grid load exceeds the available power generation capacity, the air conditioning system may be in a serious overload state, and the trend of changes in power demand is expected to increase. Overall, the system's operating pressure is high and there is a risk of power outages. Emergency measures must be taken immediately to reduce the load. In this case, the control strategy needs to be very strict. In order to ensure the safe and stable operation of the power grid, it may have a certain impact on user comfort.

[0108] In some embodiments of the present application, in order to generate and execute corresponding control strategies based on the current risk situation, to ensure that energy utilization efficiency is optimized to the maximum extent while satisfying user comfort and to ensure the safe and stable operation of the power grid, the generation and execution of corresponding control strategies based on the risk situation are specifically as follows:

[0109] When the risk situation is low, a first control strategy is generated and executed, that is, slightly increasing the set temperature, maintaining the current wind speed, and not enabling periodic on / off control;

[0110] When the risk situation is a medium risk situation, a second control strategy is generated and executed, that is, a moderate increase in the set temperature, a decrease in the wind speed, and a disabling of periodic on / off control;

[0111] When the risk situation is a high-risk situation, a third control strategy is generated and executed, that is, significantly increasing the set temperature, reducing the current wind speed, and enabling periodic on / off control.

[0112] As described above, when it is determined to be a low-risk situation, a first control strategy is generated and executed, including slightly increasing the set temperature, that is, moderately increasing the set temperature of the air conditioner, for example, increasing it by 1-2°C, which can reduce the power consumption of the air conditioning system without affecting the user's comfort; maintaining the current wind speed, that is, keeping the wind speed setting of the air conditioner unchanged to avoid obvious changes in perception to the user; not enabling periodic switch control, that is, not starting the periodic switch function of the air conditioner, allowing the air conditioning system to operate normally according to actual needs without frequent start and stop.

[0113] When it is determined to be a medium-risk situation, a second control strategy is generated and executed, including moderately increasing the set temperature, that is, further increasing the set temperature of the air conditioner, for example, increasing it by 2-3°C, which can effectively reduce the operating time and power consumption of the air conditioning system while maintaining user comfort as much as possible; reducing the wind speed, that is, appropriately reducing the wind speed setting of the air conditioner, such as from high speed to medium speed or low speed, which can not only reduce power consumption, but also reduce noise and improve user experience; not enabling periodic switching control, that is, still not starting the periodic switching function of the air conditioner, but achieving energy-saving goals by adjusting the temperature and wind speed.

[0114] When it is determined to be a high-risk situation, a third control strategy is generated and executed, including significantly increasing the set temperature, that is, significantly increasing the set temperature of the air conditioner, for example, increasing it by 3-5°C, which can quickly reduce the operating time and power consumption of the air conditioning system and effectively alleviate the pressure on the power grid; reducing the current wind speed, that is, further reducing the wind speed setting of the air conditioner, for example, from high speed to low speed or minimum wind speed, which can not only reduce power consumption, but also extend the service life of the air conditioning system; enabling periodic switching control, that is, starting the periodic switching function of the air conditioner, for example, turning off the air conditioner for 10 minutes every hour, which can significantly reduce power consumption in a short time and avoid power grid overload.

[0115] Compared with the prior art, the embodiment of the present application discloses a method for flexible load regulation of an intelligent flexible control terminal. This method monitors the operating data of the air-conditioning system in real time and uses a machine learning algorithm to classify and identify the data, so as to more accurately understand the operating status of the air-conditioning system, and then take targeted control measures to effectively reduce energy consumption and improve energy efficiency. The present invention can achieve energy-saving goals through intelligent control of the air-conditioning system without affecting user comfort. By comprehensively considering the grid demand status, the operating status of the air-conditioning system and the trend of power demand changes, the control strategy is dynamically adjusted, and the power output of the air-conditioning system can be appropriately limited during peak power consumption periods to avoid grid overload and ensure the safe and stable operation of the grid. Combined with historical data and external environmental factors such as meteorological conditions and holiday arrangements, a trained prediction model is used to determine the future trend of power demand changes, thereby improving the accuracy of future power demand forecasts and making control decisions more scientific and reasonable. The entire control process is highly intelligent and automated, and no human intervention is required to complete the process from data collection, analysis, prediction to the final execution of the control strategy, greatly improving work efficiency and reducing operation and maintenance costs.

[0116] Based on the same inventive concept as the above method, the embodiment of the present application also proposes an intelligent flexible control terminal flexible load adjustment system, such as Figure 2 FIG. 1 is a schematic diagram of a structure of an intelligent flexible control terminal flexible load regulation system, which includes:

[0117] An acquisition module is used to obtain the operating data of the air conditioning system in real time and pre-process the operating data;

[0118] An analysis module, configured to classify and identify the pre-processed data based on a machine learning algorithm to determine an operating status of the air conditioning system;

[0119] The forecasting module is used to combine historical data and external environmental factor data to determine the trend of electricity demand changes in the future;

[0120] The execution module is used to obtain the current grid demand status, combine the operating status of the air-conditioning system and the power demand change trend, generate and execute corresponding control strategies to control the temperature, wind speed and switch status of the air-conditioning system.

[0121] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or a combination of software and hardware embodiments. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0122] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0123] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for adjusting flexible load of an intelligent flexible control terminal, characterized in that: include: Acquire the operating data of the air conditioning system in real time and pre-process the operating data; classifying and identifying the pre-processed data based on a machine learning algorithm to determine the operating status of the air conditioning system; Combine historical data with external environmental factors to determine the trend of electricity demand changes in the future; The current grid demand status is obtained, and in combination with the operating status of the air-conditioning system and the power demand change trend, a corresponding control strategy is generated and executed to control the temperature, wind speed and switch status of the air-conditioning system.

2. The method according to claim 1, wherein The real-time acquisition of the operating data of the air-conditioning system and the pre-processing of the operating data are specifically as follows: The intelligent flexible control unit collects the operating data of the air-conditioning system in real time and sends the operating data to the intelligent flexible control terminal via the communication network. The operating data includes power consumption, indoor temperature, indoor humidity, wind speed, startup status and fault alarm information; The intelligent flexible control terminal pre-processes the received operation data, including data cleaning and data formatting.

3. The method according to claim 1, wherein The classifying and identifying the pre-processed data based on the machine learning algorithm to determine the operating status of the air-conditioning system is specifically as follows: Inputting the real-time collected and pre-processed operating data into a trained machine learning model; The machine learning model is based on a machine learning algorithm and determines and outputs the operating status results of the air-conditioning system according to the input data characteristics. The operating status prediction results include normal operation, mild overload and severe overload.

4. The method according to claim 3, wherein The combination of historical data and external environmental factor data to determine the trend of electricity demand changes in the future period is as follows: Acquire the historical data and external environmental factor data, and perform data preprocessing and feature extraction, the historical data including the historical operation data of the air conditioning system, historical power grid data, and historical user behavior data, and the external environmental factor data including meteorological data, holiday data, and special event data; Through the trained prediction model, based on the extracted data features, the electricity demand change trend in the future period is determined, and the electricity demand change trend includes increasing demand, stable demand and decreasing demand.

5. The method according to claim 4, wherein The acquisition of the current grid demand state is specifically as follows: Obtaining current grid load and available power generation capacity to determine grid demand status; If the current grid load is much lower than the available power generation capacity, determining that the grid demand is in a low demand state; If the current grid load is close to the available power generation capacity, determining that the grid demand is in a medium demand state; If the current grid load exceeds the available power generation capacity, it is determined that the grid demand is in a high demand state.

6. The method according to claim 5, wherein The current grid demand state is obtained, and a corresponding control strategy is generated and executed in combination with the operating state of the air conditioning system and the power demand change trend, specifically: Dynamically adjust the corresponding weights based on the specific conditions of the grid demand state, the operating state, and the power demand change trend; Determine the current comprehensive score based on the average of the adjusted weights; The current risk situation is judged based on the comprehensive score, and a corresponding control strategy is generated and executed according to the risk situation.

7. The method according to claim 6, wherein The dynamically adjusting corresponding weights based on the specific conditions of the grid demand state, the operating state, and the power demand change trend is specifically as follows: When the grid demand states are respectively the low demand state, the medium demand state, and the high demand state, dynamically adjusting the corresponding weights to be the first weight, the second weight, and the third weight; When the operating states are respectively the normal operation, the mild overload, and the severe overload, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight; When the power demand change trend is the demand increase, the demand stability, and the demand decrease, dynamically adjusting the corresponding weights to the first weight, the second weight, and the third weight; The first weight is smaller than the second weight, and the second weight is smaller than the third weight.

8. The method according to claim 6, wherein The current risk situation is judged based on the comprehensive score as follows: If the comprehensive score is less than a preset first threshold, it is determined that the current risk situation is low; If the comprehensive score is greater than or equal to a preset first threshold and less than a preset second threshold, it is determined that the current risk situation is medium; If the comprehensive score is greater than or equal to a preset second threshold, it is determined that the current situation is high-risk.

9. The method according to claim 8, wherein The generation and execution of corresponding control strategies according to the risk situation are specifically as follows: When the risk situation is low, a first control strategy is generated and executed, that is, slightly increasing the set temperature, maintaining the current wind speed, and not enabling periodic on / off control; When the risk situation is a medium risk situation, a second control strategy is generated and executed, that is, a moderate increase in the set temperature, a decrease in the wind speed, and a disabling of periodic on / off control; When the risk situation is a high-risk situation, a third control strategy is generated and executed, that is, significantly increasing the set temperature, reducing the current wind speed, and enabling periodic on / off control.

10. An intelligent flexible control terminal flexible load adjustment system, characterized in that: include: An acquisition module is used to obtain the operating data of the air conditioning system in real time and pre-process the operating data; An analysis module, configured to classify and identify the pre-processed data based on a machine learning algorithm to determine an operating status of the air conditioning system; The forecasting module is used to combine historical data and external environmental factor data to determine the trend of electricity demand changes in the future; The execution module is used to obtain the current grid demand status, combine the operating status of the air-conditioning system and the power demand change trend, generate and execute corresponding control strategies to control the temperature, wind speed and switch status of the air-conditioning system.

Images