Air conditioner gateway control method and device based on edge computing, equipment and medium

Abstract

The application discloses an air conditioner gateway control method and device based on edge computing, equipment and medium. The method comprises the following steps: acquiring the system load value of an air conditioner group, the target load value issued by a cloud platform and meteorological data; acquiring the first historical data and the second historical data stored in the air conditioner gateway; determining the predicted load of the air conditioner group by using a prediction model according to the meteorological data and the second historical data, wherein the prediction model is updated according to the first historical data; determining the load deviation of the air conditioner group according to the real-time deviation between the system load value and the target load value and the predicted deviation between the predicted load and the target load value; when the load deviation meets a preset condition, calculating the temperature set value adjustment amount according to the load deviation and the adaptive parameter, wherein the adaptive parameter is updated according to the first historical data; and generating and issuing a control instruction according to the temperature set value adjustment amount. The application can improve the adaptability of air conditioner load regulation and control to actual operation environment changes.

Description

Technical Field

[0001] This application relates to the field of air conditioning control technology, and in particular to air conditioning gateway control methods, devices, equipment and media based on edge computing. Background Technology

[0002] The existing air conditioning gateway system is an intelligent device integrating advanced technologies. Its core components include a sophisticated data acquisition module, a high-efficiency and stable communication module, a powerful main control chip, a secure and reliable storage module, a power supply module adaptable to various voltage environments, and interfaces supporting multiple communication protocols. These components work together to enable the air conditioning gateway to be widely used in various locations such as supermarkets, hotels, high-end office buildings, and government agencies, meeting the air conditioning management needs in various complex environments. In applications such as supermarkets, the air conditioning gateway, through its data acquisition module, can monitor various operating parameters of the air conditioning unit in real time, including but not limited to key indicators such as humidity, temperature, wind speed, and power consumption. This data is uploaded to the cloud platform in real time via the communication module, providing detailed data support for power supply management departments, helping them to optimize power distribution based on real-time data and achieve the dual goals of energy conservation, emission reduction, and cost control. The air conditioning gateway also has powerful command reception and parsing capabilities. It can receive control commands from the cloud platform and accurately parse the commands through its internal intelligent parsing module, extracting the various parameters required for control. Subsequently, the air conditioning gateway uses its protocol conversion function to convert these control data into a protocol format that the air conditioning system can directly recognize, thereby realizing the control of the air conditioning unit.

[0003] However, in existing air conditioning gateway control methods, the air conditioning gateway is typically only used to collect air conditioning operation data, upload data, and forward control commands issued by the cloud platform. It lacks the ability to update models and adaptively correct parameters based on local operating data. Therefore, existing air conditioning gateway control methods suffer from problems such as difficulty in updating predictive models with local historical data and difficulty in iteratively correcting adaptive parameters based on actual control effects, resulting in insufficient predictive accuracy and adaptability of air conditioning load control. Summary of the Invention

[0004] This application provides an edge computing-based air conditioning gateway control method, apparatus, device, and medium. The aim is to enable the air conditioning gateway to update the load prediction model and adaptive parameters used for temperature regulation based on locally stored historical data, and to determine the load deviation of the air conditioning group by combining the predicted load and the current system load, thereby generating control commands for the air conditioning group and improving the prediction accuracy and adaptability of air conditioning load regulation. To achieve the above objective, a first aspect of this application provides an edge computing-based air conditioning gateway control method applied to an air conditioning gateway, wherein the air conditioning gateway is communicatively connected to both an air conditioning group and a cloud platform. The air conditioning gateway control method includes: Obtain the system load value of the air conditioning unit, the target load value and meteorological data sent by the cloud platform; Obtain first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics for the same period. Based on the meteorological data and the second historical data, the predicted load of the air conditioning unit is determined using the prediction model, wherein the prediction model is updated based on the first historical data; The load deviation of the air conditioning unit is determined based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value. When the load deviation meets the preset conditions, the temperature setpoint adjustment amount is calculated based on the load deviation and the adaptive parameter, wherein the adaptive parameter is updated based on the first historical data; The control command for the air conditioning unit is generated based on the temperature setpoint adjustment and then sent to the air conditioning unit to control it. This application acquires system load value, target load value, meteorological data, and first and second historical data stored on the air conditioning gateway. It uses the historical load characteristics provided by the second historical data in conjunction with meteorological data to determine the predicted load. Simultaneously, it updates the prediction model and adaptive parameters based on the first historical data, enabling the air conditioning gateway to dynamically correct the load prediction results and temperature adjustment parameters according to local actual operating data. Compared to existing technologies where air conditioning gateways mainly rely on cloud platforms to issue fixed control strategies or use fixed parameters for control, this application improves the adaptability of the air conditioning gateway to changes in building environment, meteorological conditions, and load response characteristics, making the temperature setpoint adjustment more closely match the actual load deviation, thereby improving the prediction accuracy and adaptability of air conditioning load control.

[0005] Further, based on the meteorological data and the second historical data, the predicted load of the air conditioning unit is determined using the prediction model, including: Outdoor temperature and outdoor humidity are extracted from the meteorological data, and historical system load values ​​for the same period are extracted from the second historical data. The outdoor temperature, outdoor humidity, and historical system load values ​​for the same period are input into the prediction model to obtain the predicted load of the air conditioning unit.

[0006] This application extracts outdoor temperature and humidity from meteorological data and historical system load values ​​from local historical data. The above data is then input into a prediction model to obtain the predicted load. Compared with the existing technology that passively controls the air conditioning based solely on the current operating status, this method comprehensively considers the impact of meteorological factors and historical load values ​​on the current load change, improves the accuracy of the load prediction results, and enables the air conditioning gateway to predict the load change trend in advance.

[0007] Furthermore, the prediction model is updated based on the first historical data, including: The model is built and the dataset is updated based on the first historical data; When the preset training period is reached, or when the deviation between the predicted load and the system load value exceeds the preset error threshold, the model update dataset is called to train the prediction model and obtain the candidate model parameter set. The candidate model parameter set and the current model parameter set are evaluated in parallel to obtain the evaluation results; Based on the evaluation results, a target model parameter set is determined from the candidate model parameter set and the current model parameter set, and the prediction model is updated based on the target model parameter set.

[0008] This application constructs a historical dataset based on local historical data and trains the prediction model when a preset training period is reached or the prediction error exceeds a preset error threshold. Simultaneously, it evaluates the candidate model parameter set and the current model parameter set in parallel to determine the parameter set used to construct the prediction model. Compared with the existing technology where model parameters are fixed or rely on a cloud platform for unified training and updates, this method enables the prediction model to adaptively update according to the actual local operating environment, improving the prediction model's adaptability to different building environments, different usage periods, and different meteorological conditions, thereby improving the reliability of the predicted load.

[0009] Further, the real-time deviation between the current system load value and the target load value, and the predicted deviation between the predicted load and the target load value, are used to determine the load deviation, including: Calculate the difference between the current system load value and the target load value to obtain the real-time deviation; The difference between the predicted load and the target load value is calculated to obtain the prediction deviation; When the real-time deviation is greater than zero, the real-time deviation is determined as the load deviation; When the real-time deviation is less than or equal to zero and the prediction deviation is greater than a preset threshold, the prediction deviation is determined as the load deviation; when the real-time deviation is less than or equal to zero and the prediction deviation is less than or equal to the preset threshold, the load deviation of the air conditioning unit is determined to be zero.

[0010] This application calculates the real-time deviation between the current system load value and the target load value, as well as the predicted deviation between the predicted load and the target load value. When the real-time deviation exceeds the limit, the real-time deviation is used first. When the current load does not exceed the limit but the predicted deviation exceeds the limit, the predicted deviation is used as the load deviation. Compared with the existing technology that only controls the load based on whether the current load exceeds the target load, this method can respond promptly to the current load exceeding the limit and make flexible adjustments in advance when the predicted load is about to exceed the limit, thereby reducing load control lag and load fluctuation.

[0011] Further, calculating the temperature setpoint adjustment based on the load deviation and the adaptive parameters includes: Obtain multiple system load values ​​and their corresponding target load values ​​within a preset period prior to the current moment from the first historical data; Based on the system load values ​​and their corresponding target load values, multiple historical load deviations within the preset period are obtained; The multiple historical load deviations are integrated to obtain the load deviation integral result; The temperature setpoint adjustment amount is obtained based on the current load deviation, the load deviation integral result, and the adaptive parameter.

[0012] This application obtains multiple system load values ​​and their corresponding target load values ​​from local historical data within a preset period prior to the current moment, calculates multiple historical load deviations and performs integration processing, and then combines the current load deviation and adaptive parameters to calculate the temperature setpoint adjustment. Compared with the existing technology that directly adjusts the temperature setpoint based solely on the current instantaneous load deviation, this application can simultaneously consider the current magnitude and historical accumulation of the load deviation, making the temperature setpoint adjustment smoother and more accurate, and reducing over-adjustment or under-adjustment caused by instantaneous fluctuations.

[0013] Furthermore, the adaptive parameters are updated based on the first historical data, including: Calculate the control performance index within the preset period based on the historical load deviation; When the regulation performance index does not meet the preset regulation performance requirements, the adaptive parameters are iteratively updated based on the historical load deviation with the goal of reducing the regulation performance index.

[0014] This application calculates the control performance index within a preset period based on historical load deviation, and when the control performance index does not meet the preset control performance requirements, iteratively updates the adaptive parameters with the goal of reducing the control performance index. Compared with the existing technology that uses fixed adjustment parameters for air conditioning load control, this application can dynamically optimize the adaptive parameters based on the historical control effect of the air conditioning unit, improve the adaptability of the temperature setpoint adjustment calculation, and enable the system load to converge to the vicinity of the target load more quickly and stably.

[0015] Further, the step of generating control commands for the air conditioning unit based on the temperature setpoint adjustment amount and issuing the control commands to the air conditioning unit includes: Based on the temperature setpoint adjustment amount, determine the adjusted temperature setpoint for each air conditioner in the air conditioning unit; Based on the preset comfort constraints, the adjusted temperature setting value is subjected to amplitude limiting processing to correct the adjustment amount of the temperature setting value; The communication protocol corresponding to each air conditioner in the air conditioning group is determined, the adjustment amount is based on the corrected temperature setpoint, the control command corresponding to each air conditioner is generated according to the communication protocol, and the control command is sent to the corresponding air conditioner through the air conditioning gateway.

[0016] This application determines the adjusted temperature setpoint of each air conditioner based on the temperature setpoint adjustment amount, limits the adjusted temperature setpoint according to preset comfort constraints, and generates control commands according to the communication protocol of each air conditioner and sends them through the air conditioning gateway. Compared with the control methods in the prior art that directly execute cloud platform control commands or lack comfort constraint correction, it can reduce the air conditioning load while avoiding temperature adjustment from exceeding the comfort range, and can adapt to the communication protocols of different air conditioners, improving the executability of air conditioning group control commands and the comfort of the control process.

[0017] The second aspect of this application provides an air conditioning gateway control device based on edge computing, including: a data acquisition module, a historical data acquisition module, a load prediction module, a deviation determination module, an adjustment calculation module, and an instruction control module; The data acquisition module is used to acquire the system load value of the air conditioning unit, the target load value sent by the cloud platform, and meteorological data; The historical data acquisition module is used to acquire first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics. The load forecasting module is used to determine the forecast load of the air conditioning unit based on the meteorological data and the second historical data using the forecasting model, wherein the forecasting model is updated based on the first historical data; The deviation determination module is used to determine the load deviation of the air conditioning unit based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value. The adjustment calculation module is used to calculate the temperature setpoint adjustment amount based on the load deviation and the adaptive parameter when the load deviation meets the preset conditions, wherein the adaptive parameter is updated based on the first historical data; The instruction control module is used to generate control instructions for the air conditioning unit based on the temperature setpoint adjustment amount, and to issue the control instructions to the air conditioning unit to control the air conditioning unit. A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the edge computing-based air conditioning gateway control method described in the above-mentioned embodiments.

[0018] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the edge computing-based air conditioning gateway control method described in the above-described embodiments. Attached Figure Description

[0019] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. The drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an embodiment of an air conditioning gateway control method based on edge computing provided in this application. Figure 2 This is a schematic diagram of the control flow of an air conditioning gateway responding to load adjustment requirements according to an embodiment of this application; Figure 3 This is a schematic diagram of the hardware structure of an air conditioning gateway provided in another embodiment of this application; Figure 4 This is a schematic diagram of a collaborative control architecture for an air conditioning unit based on a cloud platform and edge computing nodes, provided in another embodiment of this application. Figure 5 This is a schematic diagram of the structure of an air conditioning gateway control device based on edge computing, provided in one embodiment of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0023] To address the problems of existing air conditioning gateway control methods, such as over-reliance on centralized decision-making via cloud platforms, lack of edge-side load forecasting capabilities, and difficulty in dynamically adjusting control parameters based on historical load deviations, such as... Figure 1 As shown in the figure, this application provides a detailed flowchart of an air conditioning gateway control method based on edge computing. The air conditioning gateway control method based on edge computing in this embodiment includes steps S101 to S106, which are described in detail below: Step S101, Obtain the system load value of the air conditioning unit, the target load value and meteorological data issued by the cloud platform.

[0024] Preferably, the operating parameters of each air conditioner are continuously acquired, including the set temperature. Return air temperature Real-time power .

[0025] Furthermore, the system load value at the current moment is determined based on the operating parameters of each air conditioner.

[0026] Preferably, the system load value at the current moment is determined based on the operating parameters of each air conditioner, and the specific calculation formula is as follows: in, It is the current total system load, which is the sum of the real-time power of all air conditioners.

[0027] Step S102: Obtain first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics.

[0028] Step S103: Based on the meteorological data and the second historical data, determine the predicted load of the air conditioning unit using the prediction model, wherein the prediction model is updated based on the first historical data.

[0029] Further, step S103 includes: Outdoor temperature and outdoor humidity are extracted from the meteorological data, and historical system load values ​​for the same period are extracted from the second historical data. The outdoor temperature, outdoor humidity, and historical system load values ​​for the same period are input into the prediction model to obtain the predicted load of the air conditioning unit.

[0030] Preferably, real-time outdoor temperature is obtained from a cloud platform. ,humidity And query historical load data from local storage. and week characteristics ; By substituting the above data into the deployed model, a predicted load value can be calculated. This value comprehensively quantifies the combined impact of various factors on the load at the current moment.

[0031] Edge devices will predict load With the target load issued by the cloud platform By making comparisons, we can formulate control strategies more proactively.

[0032] Implementing this preferred solution, compared to the existing technology that passively controls the air conditioning system based solely on its current operating status, comprehensively considers the impact of meteorological factors and historical loads on current load changes, improving the accuracy of load forecasting results and enabling the air conditioning gateway to predict load change trends in advance.

[0033] Furthermore, the prediction model is trained based on historical load data and historical meteorological data, and is updated based on the deviation between the predicted load and the system load value at the current moment.

[0034] Specifically, the update method for the prediction model is as follows: The model is built and the dataset is updated based on the first historical data; When the preset training period is reached, or when the deviation between the predicted load and the system load value exceeds the preset error threshold, the model update dataset is called to train the prediction model and obtain the candidate model parameter set. The candidate model parameter set and the current model parameter set are evaluated in parallel to obtain the evaluation results; Based on the evaluation results, a target model parameter set is determined from the candidate model parameter set and the current model parameter set, and the prediction model is updated based on the target model parameter set.

[0035] Preferably, a multiple linear regression model is established: in, It is the output of the model, i.e., the predicted air conditioning load; coefficient These are the constant terms of the model; coefficients Quantify outdoor temperature Impact on load; coefficient Quantify outdoor humidity Impact on load; coefficient Quantifying historical load The influence; coefficient Quantifying Weekday Characteristics The impact; It is model error.

[0036] The above coefficients The specific values ​​are obtained by training on historical data, and each coefficient can be determined through the following training process: Preferably, the air conditioning gateway will accumulate a large amount of historical datasets, including the actual air conditioning load, outdoor temperature, humidity, and day of the week at each point in time. This dataset can be loaded in batches from the cloud platform during system initialization and continuously updated with locally collected and calculated data during subsequent operation.

[0037] Preferably, the training of the multiple linear regression model is designed to adapt to the limited computing resources of edge devices. The training process employs a triggering mechanism combining periodic and event-specific triggers, rather than continuous execution. During training, the edge computing node accesses stored local historical datasets and uses lightweight, iterative algorithms such as recursive least squares to fit the model parameters, calculating the values ​​that enable the predicted values. Closest to the actual value A set of optimal coefficients: .

[0038] Preferably, the periodicity can be set to monthly, quarterly, or annually, and can be changed according to actual needs; the specific events include situations such as the model prediction error continuously exceeding the standard or changes in building usage patterns, and the preferred solution is to update the model according to the specific events.

[0039] Preferably, the model can be modified to include more dimensions based on the local historical data of the air conditioning gateway, and corresponding parameters can be added to each dimension to improve the prediction accuracy of the prediction model.

[0040] Preferably, to ensure the stability of model updates, an "evaluation mode" is used. The newly trained model parameter set runs in parallel in the background, and its prediction results are compared with real-time data for evaluation. If the new model's accuracy is significantly and consistently better than the currently running model during the continuous evaluation period, the system automatically solidifies the new parameter set and replaces the old parameters, completing a seamless switch. If the better standard is not met, the existing model continues to run, and the new parameters are not enabled.

[0041] Preferably, the trained model coefficients are stored in the storage module of the edge device. At this point, the quantitative impact of each influencing factor on the load is determined. The entire training process is completed autonomously at the edge; the cloud platform does not participate in the computation, but may only serve as an initial source of historical data or receive reports of the final training results.

[0042] Implementing this preferred solution, compared to existing technologies that use fixed model parameters or rely on a cloud platform for unified training and updates, enables the prediction model to adaptively update based on the local operating environment. This improves the prediction model's adaptability to different building environments, different usage periods, and different weather conditions, thereby enhancing the reliability of the predicted load.

[0043] Step S104: Determine the load deviation of the air conditioning unit based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value.

[0044] Further, based on the real-time deviation between the current system load value and the target load value, and the predicted deviation between the predicted load and the target load value, the load deviation of the air conditioning unit is determined, including: Calculate the difference between the current system load value and the target load value to obtain the real-time deviation; The difference between the predicted load and the target load value is calculated to obtain the prediction deviation; When the real-time deviation is greater than zero, the real-time deviation is determined as the load deviation; When the real-time deviation is less than or equal to zero and the prediction deviation is greater than a preset threshold, the prediction deviation is determined as the load deviation of the air conditioning unit. When the real-time deviation is less than or equal to zero, and the prediction deviation is less than or equal to the preset threshold, the load deviation of the air conditioning unit is determined to be zero.

[0045] Preferably, the specific formula for calculating the real-time deviation is as follows: in, This is the current total system load, which is the sum of the real-time power of all air conditioners. The target load value for the current time period is obtained from the cloud platform. It is a load deviation.

[0046] Preferably, the air conditioning gateway will predict the load. With the target load issued by the cloud platform By making comparisons, we can formulate control strategies more proactively.

[0047] Implementing this preferred scheme, compared to the existing technology that only adjusts the load based on whether the current load exceeds the target load, can not only respond promptly to the current load exceeding the standard, but also make flexible adjustments in advance when the load is predicted to exceed the standard, thereby reducing load control lag and load fluctuation.

[0048] Step S105: When the load deviation meets the preset conditions, calculate the temperature setpoint adjustment amount based on the load deviation and the adaptive parameter, wherein the adaptive parameter is updated based on the first historical data.

[0049] Further, based on the load deviation and adaptive parameters, the temperature setpoint adjustment is calculated, including: Obtain multiple system load values ​​and their corresponding target load values ​​within a preset period prior to the current moment from the first historical data; Based on the system load values ​​and their corresponding target load values, multiple historical load deviations within the preset period are obtained; The multiple historical load deviations are integrated to obtain the load deviation integral result; The temperature setpoint adjustment amount is obtained based on the current load deviation, the load deviation integral result, and the adaptive parameter.

[0050] Preferably, when the load deviation At that time, the adjustment amount of the temperature setpoint Determined by the following formula: in, The proportionality coefficient determines the system's response speed to the current load deviation. The larger the value, the greater the initial adjustment range, in order to achieve a rapid response. The integral coefficient is used to accumulate historical deviations. Its function is to eliminate steady-state errors and ensure that the system load can eventually converge accurately and stably to the target. nearby. It represents the integral of the load deviation from the start of regulation to the current time, reflecting the persistence of the deviation.

[0051] It is understood that the proportional coefficient and the integral coefficient are preferred adaptive parameters.

[0052] Furthermore, the adaptive parameters are iteratively updated based on the first historical data, including: Calculate the control performance index within the preset period based on the historical load deviation; When the regulation performance index does not meet the preset regulation performance requirements, the adaptive parameters are iteratively updated based on the historical load deviation with the goal of reducing the regulation performance index.

[0053] Preferably, when the system is first deployed or when a remote command from the cloud platform is triggered, the proportional coefficient and the integral coefficient are set as follows: The edge computing node sends a preset small temperature setpoint step signal to the air conditioning unit, while simultaneously collecting the total system load at high frequency. Real-time response data; Based on step response data, approximate model parameters of the controlled object are estimated using a first-order system model identification method. Subsequently, a set of initial parameters can be automatically calculated using the Ziegler-Nichols method. , parameter.

[0054] Preferably, the first-order system model identification method can be the tangent method, and the approximate model parameters can be the gain K and the time constant T.

[0055] Preferably, the initial parameters are fed into the closed-loop control system and run under typical operating conditions for a period of time; the edge computing nodes calculate the control performance indicators during this period; the lightweight optimization algorithm built into the air conditioning gateway aims to minimize the overall performance indicators. , The parameters are iteratively fine-tuned and evaluated; when the performance indicators meet the preset requirements or the iteration converges, the optimal parameter set is fixed and stored in the local storage module as the benchmark parameters for subsequent load control.

[0056] Preferably, the lightweight optimization algorithm can be the simplex method, and the control performance indicators can be settling time, overshoot, and integral absolute error.

[0057] Implementing this preferred scheme, compared to the existing technology that uses fixed adjustment parameters for air conditioning load control, can dynamically optimize adaptive parameters based on the historical control effects of the air conditioning unit, improve the adaptability of temperature setpoint adjustment calculation, and enable the system load to converge to the vicinity of the target load more quickly and stably.

[0058] Step S106: Generate a control command for the air conditioning unit based on the temperature setpoint adjustment amount, and send the control command to the air conditioning unit to control the air conditioning unit.

[0059] Further, a control command for the air conditioning unit is generated based on the temperature setpoint adjustment amount, and the control command is sent to the air conditioning unit, including: Based on the temperature setpoint adjustment amount, determine the adjusted temperature setpoint for each air conditioner in the air conditioning unit; Based on the preset comfort constraints, the adjusted temperature setting value is subjected to amplitude limiting processing to correct the adjustment amount of the temperature setting value; The communication protocol corresponding to each air conditioner in the air conditioning group is determined, and the control command corresponding to each air conditioner is generated according to the corrected temperature setpoint adjustment amount and the communication protocol.

[0060] Preferably, during long-term operation, the air conditioning gateway continuously monitors and controls the performance. If performance degradation is detected due to seasonal or building usage patterns, recursive parameter fine-tuning based on recent operating data can be triggered, enabling adaptive parameter updates without manual intervention.

[0061] Preferably, the temperature setpoint adjustment amount It must comply with the flexible control mechanism emphasized in this invention, therefore the final temperature adjustment must be constrained within a reasonable range.

[0062] Preferably, in cooling mode, the adjusted temperature is ensured. The temperature should not be lower than the lower limit of human comfort, which can be selected as 23°C; in heating mode, the adjusted temperature should not be higher than the upper limit of human comfort, which can be selected as 26°C. Preferably, the flexible adjustment of the temperature setpoint includes: based on the calculated... It sends instructions to all controlled air conditioners to uniformly and slightly adjust their set temperatures, thereby reducing the overall load by minimizing the temperature difference between indoors and outdoors.

[0063] Preferably, the set temperature can be adjusted uniformly and slightly by increasing it by 0.5℃ to 1℃ under cooling conditions.

[0064] Preferably, the operating mode of each air conditioner in the air conditioning unit can be optimized: based on the acquired outdoor temperature. If the environment is suitable, such as during transitional seasons, the system will automatically switch some air conditioners from "cooling / heating" mode to the more energy-efficient "ventilation" mode.

[0065] Implementing this preferred solution, compared to existing technologies that directly execute cloud platform control commands or lack comfort constraint corrections, can reduce air conditioning load while avoiding temperature adjustments exceeding the comfort range. Furthermore, it can adapt to the communication protocols of different air conditioners, improving the executability of air conditioning control commands and the comfort of the control process.

[0066] For a clearer illustration of this embodiment, see [link to example]. Figure 2 , Figure 2 A schematic diagram of the control flow for the air conditioning gateway to respond to load adjustment requirements, as shown below. Figure 2As shown, after receiving the load adjustment request from the cloud platform, the air conditioning gateway first reads the operating parameters of each air conditioner in the current air conditioning group. Then, combining the historical load curve, air conditioner operating parameters, and load adjustment request, it performs edge computing to determine the temperature adjustment range and temperature level required for this load adjustment. Subsequently, the air conditioning gateway begins to execute flexible load control, converting the temperature adjustment strategy into control commands that each air conditioner in the air conditioning group can recognize and execute. Finally, the air conditioning gateway monitors the current air conditioning control status in real time, waiting for the load to be adjusted to the target range. This process demonstrates the control process of the air conditioning gateway completing parameter reading, strategy calculation, command conversion, and closed-loop monitoring at the edge after receiving the adjustment request from the cloud platform.

[0067] By implementing the embodiments of this application, compared with the prior art where the air conditioning gateway only performs data uploading and command forwarding and mainly relies on the centralized computing and control strategy of the cloud platform, the load forecasting and control decisions can be pushed down to the edge side for execution, reducing the centralized computing pressure and data transmission volume of the cloud platform, and improving the real-time performance and foresight of air conditioning load control.

[0068] See Figure 3 Another embodiment of this application provides a hardware structure for an air conditioning gateway. Figure 3 This is a schematic diagram of the hardware structure of an air conditioning gateway. (For example...) Figure 3 As shown, the air conditioning gateway includes a communication module, a data acquisition module, a main control chip, a storage module, and a power module. The communication module and the data acquisition module are respectively connected to multiple air conditioners in the air conditioning group, which may include air conditioner 1, air conditioner 2, ..., air conditioner n. The data acquisition module is used to collect the operating parameters of each air conditioner, which may include real-time power, set temperature, return air temperature, operating mode, etc. The communication module is used to realize data interaction between the air conditioning gateway and each air conditioner, as well as data communication between the air conditioning gateway and the cloud platform.

[0069] The main control chip is connected to the communication module, the data acquisition module, the storage module, and the power supply module. The main control chip calculates the current system load value based on the collected air conditioner operating parameters, and, in conjunction with the target load value and meteorological data from the cloud platform, performs load prediction, load deviation calculation, temperature setpoint adjustment calculation, and control command generation at the air conditioning gateway. The storage module stores local historical data, prediction model parameters, adaptive parameters, and control records. The power supply module powers the various functional modules in the air conditioning gateway. Therefore, the air conditioning gateway can not only perform data acquisition and protocol forwarding, but also act as an edge computing node to perform load prediction and control decisions locally.

[0070] See Figure 4Another embodiment of this application provides an air conditioning gateway control system based on a cloud platform and edge computing nodes. Figure 4 This is a schematic diagram of a collaborative control architecture for air conditioning units based on a cloud platform and edge computing nodes. Figure 4 As shown in the embodiments of this application, the air conditioning gateway control system may include a cloud platform, multiple air conditioning gateways, and multiple air conditioning units. Each air conditioning gateway acts as an edge computing node, communicating with multiple air conditioners within a corresponding area. For example, air conditioning gateway 1 is connected to its subordinate air conditioners 1, 2, ..., n, and air conditioning gateway n is connected to its subordinate multiple air conditioners. Each air conditioning gateway is communicating with the cloud platform to receive target load values, meteorological data, and load adjustment requirements from the cloud platform, and uploads local operating data, system load values, and control results to the cloud platform.

[0071] In this embodiment, the cloud platform can be used for global load management and to send corresponding target load values ​​and adjustment requirements to each air conditioning gateway. Each air conditioning gateway collects the operating parameters of its corresponding air conditioning unit locally and performs edge-side load prediction and adaptive control based on local historical data and meteorological data. Thus, multiple air conditioning gateways can perform localized adjustments for different regions or different air conditioning units, reducing the centralized computing pressure and communication data volume of the cloud platform, while improving the real-time performance and foresight of air conditioning load control in each region.

[0072] Based on the above method embodiments, this application provides corresponding apparatus embodiments.

[0073] like Figure 5 As shown, one embodiment of this application provides an air conditioning gateway control device based on edge computing, including: a data acquisition module 100, a historical data acquisition module 200, a load prediction module 300, a deviation determination module 400, an adjustment calculation module 500, and an instruction control module 600; The data acquisition module 100 is used to acquire the system load value of the air conditioning unit, the target load value sent by the cloud platform, and meteorological data. The historical data acquisition module 200 is used to acquire first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics. The load forecasting module 300 is used to determine the forecast load of the air conditioning unit based on the meteorological data and the second historical data using the forecasting model, wherein the forecasting model is updated based on the first historical data; Furthermore, the load forecasting module 300 includes: An outdoor data extraction unit is used to extract outdoor temperature and outdoor humidity from the meteorological data, and to extract historical system load values ​​from the local historical data. The predicted load acquisition unit is used to input the outdoor temperature, the outdoor humidity and the historical system load value for the same period into the prediction model to obtain the predicted load.

[0074] The deviation determination module 400 is used to determine the load deviation of the air conditioning unit based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value. Furthermore, the deviation determination module 400 includes: The real-time deviation calculation unit is used to calculate the difference between the current system load value and the target load value to obtain the real-time deviation; The prediction deviation calculation unit is used to calculate the difference between the predicted load and the target load value to obtain the prediction deviation; The first load deviation determination unit is used to determine the real-time deviation as the load deviation when the real-time deviation is greater than zero. The second load deviation determination unit is used to determine the predicted deviation as the load deviation of the air conditioning unit when the real-time deviation is less than or equal to zero and the predicted deviation is greater than a preset threshold. The third load deviation determination unit is used to determine the load deviation of the air conditioning unit as zero when the real-time deviation is less than or equal to zero and the predicted deviation is less than or equal to the preset threshold.

[0075] The adjustment calculation module 500 is used to calculate the temperature setpoint adjustment amount according to the load deviation and the adaptive parameter when the load deviation meets the preset conditions, wherein the adaptive parameter is updated according to the first historical data; Furthermore, the adjustment calculation module 500 includes: The load value acquisition unit is used to acquire multiple system load values ​​and their corresponding target load values ​​within a preset period prior to the current time from the local historical data; The historical load deviation acquisition unit is used to obtain multiple historical load deviations within the preset period based on each system load value and its corresponding target load value. An integral processing unit is used to perform integral processing on the multiple historical load deviations to obtain the load deviation integral result; The adjustment amount acquisition unit is used to obtain the temperature setpoint adjustment amount based on the current load deviation, the load deviation integral result, and the adaptive parameter.

[0076] The instruction control module 600 is used to generate control instructions for the air conditioning unit based on the temperature setpoint adjustment amount, and to send the control instructions to the air conditioning unit to control the air conditioning unit.

[0077] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided in this application, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0078] Based on the above embodiment of the air conditioning gateway control method based on edge computing, another embodiment of this application provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the air conditioning gateway control method based on edge computing of any embodiment of this application.

[0079] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete this application. The one or more module units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0080] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0081] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0082] Based on the above-described method embodiments, another embodiment of this application provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the edge computing-based air conditioning gateway control method described in any of the above-described method embodiments of this application.

[0083] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0084] Based on the above-described method embodiments, another embodiment of this application provides a computer program product, including a computer program or instructions, which, when executed by a communication device, implements an edge computing-based air conditioning gateway control method.

[0085] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, or improvements made by those skilled in the art within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An air conditioning gateway control method based on edge computing, characterized in that, Applied to an air conditioning gateway, the air conditioning gateway is communicatively connected to an air conditioning unit and a cloud platform, and the air conditioning gateway control method includes: Obtain the system load value of the air conditioning unit, the target load value and meteorological data sent by the cloud platform; Obtain first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics for the same period. Based on the meteorological data and the second historical data, the predicted load of the air conditioning unit is determined using the prediction model, wherein the prediction model is updated based on the first historical data; The load deviation of the air conditioning unit is determined based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value. When the load deviation meets the preset conditions, the temperature setpoint adjustment amount is calculated based on the load deviation and the adaptive parameter, wherein the adaptive parameter is updated based on the first historical data; The control command for the air conditioning unit is generated based on the temperature setpoint adjustment amount, and the control command is sent to the air conditioning unit to control the air conditioning unit.

2. The air conditioning gateway control method based on edge computing according to claim 1, characterized in that, The step of determining the predicted load of the air conditioning unit using the prediction model based on the meteorological data and the second historical data includes: Outdoor temperature and outdoor humidity are extracted from the meteorological data, and historical system load values ​​for the same period are extracted from the second historical data. The outdoor temperature, outdoor humidity, and historical system load values ​​for the same period are input into the prediction model to obtain the predicted load of the air conditioning unit.

3. The air conditioning gateway control method based on edge computing according to claim 1, characterized in that, The prediction model is updated based on the first historical data, including: The model is built and the dataset is updated based on the first historical data; When the preset training period is reached, or when the deviation between the predicted load and the system load value exceeds the preset error threshold, the model update dataset is called to train the prediction model and obtain the candidate model parameter set. The candidate model parameter set and the current model parameter set are evaluated in parallel to obtain the evaluation results; Based on the evaluation results, a target model parameter set is determined from the candidate model parameter set and the current model parameter set, and the prediction model is updated based on the target model parameter set.

4. The air conditioning gateway control method based on edge computing according to claim 1, characterized in that, Determining the load deviation of the air conditioning unit based on the real-time deviation between the current system load value and the target load value, and the predicted deviation between the predicted load and the target load value, includes: Calculate the difference between the current system load value and the target load value to obtain the real-time deviation; The difference between the predicted load and the target load value is calculated to obtain the prediction deviation; When the real-time deviation is greater than zero, the real-time deviation is determined as the load deviation; When the real-time deviation is less than or equal to zero and the prediction deviation is greater than a preset threshold, the prediction deviation is determined as the load deviation of the air conditioning unit. When the real-time deviation is less than or equal to zero, and the prediction deviation is less than or equal to the preset threshold, the load deviation of the air conditioning unit is determined to be zero.

5. The air conditioning gateway control method based on edge computing according to claim 1, characterized in that, The step of calculating the temperature setpoint adjustment based on the load deviation and the adaptive parameters includes: Obtain multiple historical system load values ​​and their corresponding historical target load values ​​within a preset period from the first historical data; Based on each of the historical system load values ​​and their corresponding historical target load values, multiple historical load deviations within the preset period are obtained; The multiple historical load deviations are integrated to obtain the load deviation integral result; The temperature setpoint adjustment amount is obtained based on the current load deviation, the load deviation integral result, and the adaptive parameter.

6. The air conditioning gateway control method based on edge computing according to claim 5, characterized in that, The adaptive parameters are updated based on the first historical data, including: Calculate the control performance index within the preset period based on the historical load deviations described above; When the regulation performance index does not meet the preset regulation performance requirements, the adaptive parameters are iteratively updated based on the historical load deviations, with the goal of reducing the regulation performance index.

7. The air conditioning gateway control method based on edge computing according to claim 1, characterized in that, The step of generating control commands for the air conditioning unit based on the temperature setpoint adjustment amount and issuing the control commands to the air conditioning unit includes: Based on the temperature setpoint adjustment amount, determine the adjusted temperature setpoint for each air conditioner in the air conditioning unit; Based on the preset comfort constraints, the adjusted temperature setting value is subjected to amplitude limiting processing to correct the adjustment amount of the temperature setting value; The communication protocol corresponding to each air conditioner in the air conditioning group is determined, and the control command corresponding to each air conditioner is generated according to the corrected temperature setpoint adjustment amount and the communication protocol.

8. An air conditioning gateway control device based on edge computing, characterized in that, include: The system includes a data acquisition module, a historical data acquisition module, a load forecasting module, a deviation determination module, an adjustment calculation module, and an instruction control module. The data acquisition module is used to acquire the system load value of the air conditioning unit, the target load value sent by the cloud platform, and meteorological data; The historical data acquisition module is used to acquire first historical data and second historical data stored in the air conditioning gateway, wherein the first historical data is used to update the prediction model and adaptive parameters, and the second historical data is used to provide historical load characteristics. The load forecasting module is used to determine the forecast load of the air conditioning unit based on the meteorological data and the second historical data using the forecasting model, wherein the forecasting model is updated based on the first historical data; The deviation determination module is used to determine the load deviation of the air conditioning unit based on the real-time deviation between the system load value and the target load value, and the predicted deviation between the predicted load and the target load value. The adjustment calculation module is used to calculate the temperature setpoint adjustment amount based on the load deviation and the adaptive parameter when the load deviation meets the preset conditions, wherein the adaptive parameter is updated based on the first historical data; The instruction control module is used to generate control instructions for the air conditioning unit based on the temperature setpoint adjustment amount, and to send the control instructions to the air conditioning unit to control the air conditioning unit.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the air conditioning gateway control method based on edge computing as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the edge computing-based air conditioning gateway control method as described in any one of claims 1 to 7.

Images