An air conditioner energy-saving control method and system based on multi-source sensing and feedback correction optimization

Abstract

This application discloses an air conditioning energy-saving control method and system based on multi-source sensing and feedback correction optimization, including: collecting air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body state-related parameters to construct a multi-dimensional data matrix; inputting the multi-dimensional data matrix into a temperature-power consumption coupled prediction model to output indoor temperature prediction data, power consumption prediction data, and comfort evaluation data; generating dynamic optimization target data and target control parameter combinations, and sending them to the air conditioning control system; collecting the execution result data to generate energy-saving control evaluation results and anomaly judgment results; when the preset energy-saving control conditions are not met and there are no abnormal operating conditions, executing the energy-saving processing mode and re-executing the prediction, optimization, sending, and evaluation process; and executing the anomaly processing mode when abnormal operating conditions exist. This method improves the adaptability of control parameters to actual operating conditions, resulting in better energy-saving stability and operational reliability of the air conditioning energy-saving control system during operation.

Description

Technical Field

[0001] This specification relates to the field of power energy conservation technology, and in particular to an air conditioning energy-saving control method and system based on multi-source sensing and feedback correction optimization. Background Technology

[0002] With the continuous acceleration of urbanization and the ongoing upgrading of building intelligence, air conditioning systems have become core energy-consuming equipment in various buildings such as office buildings, shopping malls, and hospitals, accounting for 30%-50% of total building energy consumption. This represents a crucial area requiring breakthroughs in air conditioning energy conservation. Against the backdrop of increasingly urgent "dual-carbon" goals and the need for energy conservation and emission reduction, how to achieve precise control of air conditioning system energy consumption while ensuring indoor occupant comfort and stable operation of equipment in computer rooms has become a core research topic in the field of power energy conservation technology.

[0003] In the existing technology, air conditioning energy-saving solutions based on neural networks and genetic algorithms have been initially applied. For example, the invention patent with publication number CN119989863A proposes a technical approach that uses a recurrent neural network to predict indoor temperature and total active power, and combines it with a genetic algorithm to optimize parameters and output the indoor set temperature and supply air temperature, providing a basic solution for air conditioning energy saving.

[0004] However, the control method in this scheme mainly relies on the temperature and power consumption results from the prediction phase and the resulting optimization parameters. In actual complex building operation scenarios, after the control parameters are issued, indoor occupancy density, outdoor environment, load status, sensor data quality, and equipment operating status will all continuously change. This makes it difficult for the control parameters obtained in the prediction and optimization phase to always match the actual operating conditions. Especially when the operating results deviate from expectations or the operating conditions fluctuate abnormally, simply relying on the existing optimization results to continue control will result in fluctuating energy-saving effects if comfort is maintained, and decreased comfort if energy-saving effects are maintained, leading to poor system stability. Summary of the Invention

[0005] This application addresses the problems existing in the prior art by proposing an air conditioning energy-saving control method and system based on multi-source sensing and feedback correction optimization. This method can improve the energy-saving stability and operational reliability of the air conditioning control system.

[0006] To achieve the above objectives, this application provides the following technical solution: Firstly, an air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization includes the following steps: S1. Collect air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body status related parameters, and construct a multi-dimensional data matrix based on the collection time, collection area, and data type; S2. Input the multidimensional data matrix into the temperature-power consumption coupled prediction model, and output indoor temperature prediction data, power consumption prediction data, and comfort evaluation data; S3. Generate dynamic optimization target data based on indoor temperature prediction data, power consumption prediction data, and comfort evaluation data, and generate target control parameter combinations based on the dynamic optimization target data; S4. Send the target control parameter combination to the air conditioning control system and collect the operating result data of the air conditioning control system after executing the target control parameter combination; S5. Generate energy-saving control evaluation results based on the operation result data to characterize whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and anomaly judgment results to characterize whether there are abnormal operating conditions. S6. When the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition, determine the energy-saving treatment method based on the energy-saving control evaluation result, execute the energy-saving treatment method, and then re-execute steps S2 to S5 until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition; when the anomaly judgment result indicates the existence of an abnormal operating condition, determine the anomaly treatment method based on the anomaly judgment result and execute the anomaly treatment method.

[0007] In some embodiments, the measured indoor temperature data, measured total active power data, measured comfort data, and equipment operating status data are obtained after the target control parameter combination is executed. Based on the measured indoor temperature data, measured total active power data, and measured comfort data, determine whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and generate energy-saving control evaluation results. Based on the measured indoor temperature data, measured total active power data, measured comfort data, and equipment operating status data, determine whether there are any abnormalities in data acquisition, environmental disturbances, load fluctuations, or equipment operation, and generate anomaly judgment results. The preset energy-saving control conditions include at least one of indoor temperature constraints, power consumption constraints, and comfort constraints, and the measured comfort data include at least one of indoor relative humidity, air velocity, carbon dioxide concentration, and personnel density.

[0008] In some embodiments, when the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition, the weight configuration of the corrected dynamic optimization target data is determined as the energy-saving processing method. After the energy-saving processing method is executed, steps S2 to S5 are re-executed until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition. When the anomaly determination result indicates that there is a slight abnormal working condition, the weight configuration of the dynamic optimization target data will be corrected as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 will be executed again. When the anomaly determination result indicates that there is a moderate abnormal operating condition, the model parameters of the temperature-power coupling prediction model and the weight configuration of the dynamically optimized target data are determined as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 are executed again. When the anomaly determination result indicates that there is a severe abnormal operating condition, the air conditioning control mode is switched to the emergency control mode, and a combination of safety control parameters is generated based on the emergency control mode. When the anomaly determination result indicates the existence of extreme abnormal operating conditions, the energy-saving optimization control will be stopped and an anomaly alarm message will be output.

[0009] In some embodiments, the severity of the abnormal condition is determined based on at least one of the following: the degree of deviation between the operating result data and the corresponding predicted data, the duration of the abnormality, the type of abnormality, and the operating status of the equipment. The corresponding prediction data includes at least one of indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0010] In some embodiments, the target data for dynamic optimization includes at least one of temperature deviation data, comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data; The weight configuration of the target data is determined based on the building area type, the outdoor environmental status determined by indoor and outdoor environmental parameters, and the human body status correlation parameters. Building area types include areas for human activity and areas for equipment operation; When the building area type is a human activity area, increase the influence weight of comfort deviation data in the dynamic optimization target data; When the building area type is an equipment operation area, increase the influence weight of energy consumption evaluation data and comfort deviation data used to characterize indoor temperature stability in the dynamic optimization target data; When outdoor environmental conditions or human condition-related parameters change, update the weighting configuration among comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data.

[0011] In some embodiments, the temperature-power coupling prediction model includes a feature weighting layer; The feature weight allocation layer generates the influence weight of each input parameter based on the temporal variation characteristics of each input parameter in the multidimensional data matrix, the correlation characteristics between parameters, and the degree of influence of each input parameter on indoor temperature prediction data, power consumption prediction data, and comfort evaluation data. The temperature-power consumption coupled prediction model weights the multidimensional data matrix according to the influence weights and outputs indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0012] In some embodiments, the target control parameter combination includes at least one of indoor set temperature, supply air temperature and supply air volume; Based on dynamically optimized target data, a combination of target control parameters is generated, including: Determine the candidate value range for each control parameter in the target control parameter combination; Generate multiple candidate combinations of control parameters within the candidate value range; Based on the dynamic optimization target data, evaluate the comfort deviation, energy consumption status and energy efficiency status corresponding to each candidate combination of control parameters. The target control parameter combination is determined from the candidate combinations of control parameters that satisfy comfort constraints, energy consumption constraints, and equipment operation constraints.

[0013] In some embodiments, the air conditioning system operating parameters include at least one of the following: supply air temperature, return air temperature, supply air volume, total active power, and indoor set temperature. Indoor and outdoor environmental parameters include at least one of the following: indoor temperature, relative humidity, air velocity, carbon dioxide concentration, outdoor temperature, solar radiation intensity, and outdoor wind speed. Human body status-related parameters include at least one of indoor personnel density and personnel activity intensity.

[0014] Secondly, this application provides an air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization, used to execute the air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization of the first aspect, including: The multi-source sensing module is used to collect operating parameters of the air conditioning system, indoor and outdoor environmental parameters, and human status-related parameters; The data matrix construction module is used to construct multidimensional data matrices based on the collection time, collection area, and data type. The coupled prediction module is used to input a multi-dimensional data matrix into the temperature-power consumption coupled prediction model and output indoor temperature prediction data, power consumption prediction data, and comfort evaluation data. The dynamic optimization module is used to generate dynamic optimization target data based on indoor temperature prediction data, power consumption prediction data, and comfort evaluation data, and to generate target control parameter combinations based on the dynamic optimization target data; The control sending module is used to send the target control parameter combination to the air conditioning control system; The operation monitoring module is used to collect the operation result data of the air conditioning control system after executing the target control parameter combination; The evaluation and judgment module is used to generate energy-saving control evaluation results and anomaly judgment results based on the operation result data; The hierarchical processing module is used to determine the energy-saving processing method based on the energy-saving control evaluation results when the energy-saving control evaluation results do not meet the preset energy-saving control conditions and the anomaly judgment results do not indicate the existence of an abnormal operating condition, and to generate a new combination of target control parameters based on the energy-saving processing method; it is also used to determine the anomaly processing method based on the anomaly judgment results when the anomaly judgment results indicate the existence of an abnormal operating condition, and to execute the target control parameter regeneration path, emergency control path or anomaly alarm path based on the anomaly processing method.

[0015] In some embodiments, the system further includes: The storage module is used to store historical running data, model parameters, target control parameter combinations, running logs, and abnormal sample data. The communication module is used to enable data transmission between the system and the air conditioning control system; The human-computer interaction module is used to display real-time operating data, prediction results, target control parameter combinations, energy-saving control evaluation results, anomaly judgment results, and system operating status. This application collects operational data of the air conditioning control system after the target control parameter combination is issued, and generates energy-saving control evaluation results and anomaly judgment results based on the operational data. This allows air conditioning energy-saving control to extend from the predictive optimization stage to the actual execution stage. When the execution effect of the control parameters does not meet the preset energy-saving control conditions, the target control parameter combination is regenerated and re-evaluated. In the event of abnormal operating conditions, the corresponding anomaly handling method is determined based on the anomaly judgment results, avoiding the continued use of control parameters that are incompatible with the actual operating state or pose safety hazards, thereby improving the continuous adaptability of the predictive optimization results in the execution stage. Through the above settings, the method and system provided in this application can improve the energy-saving stability, comfort maintenance capability, and operational reliability of the air conditioning control system in complex building operating scenarios. Attached Figure Description

[0016] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The same numbers in the drawings denote the same structures or steps.

[0017] Figure 1 This is a schematic diagram of the overall application scenario of the air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization in the embodiments of this application.

[0018] Figure 2 This is a schematic diagram of the air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization in the embodiments of this application.

[0019] Figure 3This is a schematic diagram of an air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization in the embodiments of this application.

[0020] Figure reference numerals: 100, Air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization; 101, Multi-source sensing module; 102, Data matrix construction module; 103, Coupled prediction module; 104, Dynamic optimization module; 105, Control distribution module; 106, Operation monitoring module; 107, Evaluation and judgment module; 108, Hierarchical processing module; 109, Storage module; 110, Communication module; 111, Human-machine interaction module. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely preferred embodiments of this application and are only used to explain this application. They do not limit the scope of protection 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] The intelligent power distribution system for buildings is a community power distribution terminal system developed for the increasingly widespread use of intelligent residential buildings in cities. It can cooperate with the power distribution network automation master station and power grid automation substation system to achieve functions such as data acquisition, load detection, fault detection, and automatic power supply switching for multiple residences, multiple electrical devices, and multiple power lines. For example... Figure 1 The diagram shows the overall application architecture of an air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization. It intuitively demonstrates the collaborative working logic of each module and the linkage between the system and air conditioning equipment and sensing terminals.

[0023] As one implementation method, such as Figure 2 As shown, this application provides an air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization, including the following steps: S1. Collect air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body status related parameters, and construct a multi-dimensional data matrix based on the collection time, collection area, and data type; Specifically, air conditioning system operating parameters characterize the operating status of the air conditioning equipment at the current time and within the collection area; indoor and outdoor environmental parameters characterize the thermal and humidity conditions and external disturbances of the environment in which the air conditioning system is located; and human body status correlation parameters characterize the distribution and activities of people within the collection area. These parameters are categorized according to collection time, collection area, and data type, establishing a correspondence between different types of parameters within the same time range and spatial area, and are organized into a multi-dimensional data matrix.

[0024] Through the above settings, the scattered equipment operation data, environmental status data, and human status correlation data can be unified into the same data structure, so that parameters from different sources and of different types have clear time correspondence, regional correspondence, and category correspondence, which makes it easier to reflect the correlation between the air conditioning system operation status and the environment and personnel status.

[0025] S2. Input the multidimensional data matrix into the temperature-power consumption coupled prediction model, and output indoor temperature prediction data, power consumption prediction data, and comfort evaluation data; Specifically, the temperature-power consumption coupling prediction model is a prediction model used to characterize the correspondence between the air conditioner's operating state, environmental state, human body state, and changes in indoor temperature and power consumption.

[0026] Furthermore, the temperature-power consumption coupled prediction model is used to process the air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body status correlation parameters in the multidimensional data matrix, and output indoor temperature prediction data, power consumption prediction data, and comfort evaluation data corresponding to the current collection time and collection area.

[0027] It should be noted that the air conditioning system operating parameters reflect the current operating status of the air conditioning control system, the indoor and outdoor environmental parameters reflect the environmental conditions inside and outside the data collection area, and the human body status correlation parameters reflect the status of people within the data collection area. The temperature-power consumption coupled prediction model uses the above parameters as input data at the same time and within the same data collection area to determine the changes in indoor temperature, the changes in air conditioning system power consumption, and the comfort level, and generates indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0028] Furthermore, after inputting the multidimensional data matrix into the temperature-power consumption coupled prediction model, the model reads various parameters from the multidimensional data matrix according to the acquisition time and acquisition area, and processes the parameters related to indoor temperature changes, parameters related to air conditioning power consumption changes, and parameters related to human comfort, respectively, and outputs indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0029] Through the above settings, multiple state parameters in the multidimensional data matrix can be converted into corresponding predicted data, so that indoor temperature, air conditioning power consumption and comfort status can each have clear data output results.

[0030] S3. Generate dynamic optimization target data based on indoor temperature prediction data, power consumption prediction data, and comfort evaluation data, and generate target control parameter combinations based on the dynamic optimization target data; Specifically, indoor temperature prediction data is used to characterize the deviation of indoor temperature in the collection area from the target temperature range, power consumption prediction data is used to characterize the energy consumption level of the air conditioning system, and comfort evaluation data is used to characterize the satisfaction of people's comfort needs in the collection area.

[0031] Furthermore, the target data for dynamic optimization is a data set consisting of temperature deviation data, energy consumption evaluation data, comfort deviation data, and corresponding weight configurations.

[0032] It should be noted that temperature deviation data is determined based on the difference between the predicted indoor temperature data and the target temperature range, and is used to characterize the direction and extent to which the predicted indoor temperature deviates from the target temperature range; energy consumption evaluation data is determined based on the difference between the predicted power consumption data and the preset power consumption constraint or historical power consumption benchmark, and is used to characterize the degree of deviation of the predicted power consumption from the energy consumption control requirements; comfort deviation data is determined based on the difference between the comfort evaluation data and the preset comfort requirements, and is used to characterize the degree of deviation of the predicted comfort state from the comfort requirements. The corresponding weight configuration is determined based on the building area type, outdoor environmental conditions, and human body condition-related parameters, and is used to represent the degree of influence of temperature deviation data, energy consumption evaluation data, and comfort deviation data on the dynamic optimization target data.

[0033] Furthermore, after generating the dynamic optimization target data, multiple candidate combinations of control parameters are formed based on the candidate value range of the air conditioning control parameters. The entire dynamic optimization target data is used as the basis for evaluating the control parameters. A comprehensive evaluation is conducted on each candidate combination of control parameters to determine the target control parameter combination from the candidate combinations of control parameters that meet the requirements of temperature control, energy consumption control, comfort maintenance, and equipment operation constraints.

[0034] For example, when temperature deviation data is used alone as the basis for evaluating control parameters, the evaluation results focus on reducing indoor temperature deviation, but have insufficient constraints on power consumption level and comfort status; when energy consumption evaluation data is used alone as the basis for evaluating control parameters, the evaluation results focus on reducing the energy consumption of the air conditioning system, but have insufficient constraints on indoor temperature deviation and comfort status; when comfort deviation data is used alone as the basis for evaluating control parameters, the evaluation results focus on meeting the comfort needs of personnel, but have insufficient constraints on power consumption level.

[0035] Through the above settings, the degree of indoor temperature deviation, energy consumption level and the degree of satisfaction of human comfort needs can be unified into the same control parameter evaluation process, so that the target control parameter combination is determined by the comprehensive evaluation results of each candidate control parameter combination in terms of temperature control, energy consumption control and comfort maintenance.

[0036] S4. Send the target control parameter combination to the air conditioning control system and collect the operating result data of the air conditioning control system after executing the target control parameter combination; S5. Generate energy-saving control evaluation results based on the operation result data to characterize whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and anomaly judgment results to characterize whether there are abnormal operating conditions. Specifically, after the target control parameter combination is sent to the air conditioning control system, the air conditioning control system adjusts the corresponding controlled object according to the target control parameter combination. The controlled object includes at least one of the indoor set temperature, supply air temperature, and supply air volume. During the execution of the target control parameter combination, the operating result data of the air conditioning control system is collected. The operating result data includes measured indoor temperature data, measured total active power data, measured comfort level correlation data, and equipment operating status data.

[0037] When generating energy-saving control evaluation results, the measured indoor temperature data is compared with the indoor temperature constraint, the measured total active power data is compared with the power consumption constraint, and the comfort status within the collection area is determined based on the measured comfort correlation data to determine whether the comfort constraint is met. Based on the above comparison results, it is determined whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, thus forming the energy-saving control evaluation results.

[0038] When generating anomaly determination results, the system assesses whether the air conditioning control system exhibits data acquisition anomalies, environmental disturbance anomalies, load fluctuation anomalies, or equipment operation anomalies based on the data integrity, continuity, variation magnitude, and equipment operating status of the operational results data. Energy-saving control evaluation results characterize the control effect of the target control parameter combination, while anomaly determination results characterize the operating conditions during the execution of the target control parameter combination.

[0039] Through the above settings, the actual operating conditions after the target control parameter combination is executed can be divided into two levels: control effect evaluation and abnormal operating condition judgment. This provides clear judgment criteria for situations where the control effect fails to meet the preset energy-saving control conditions and where abnormal operating conditions exist.

[0040] S6. When the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition, determine the energy-saving treatment method based on the energy-saving control evaluation result, execute the energy-saving treatment method, and then re-execute steps S2 to S5 until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition; when the anomaly judgment result indicates the existence of an abnormal operating condition, determine the anomaly treatment method based on the anomaly judgment result and execute the anomaly treatment method.

[0041] Specifically, when the energy-saving control evaluation result does not meet the preset energy-saving control conditions, and the anomaly judgment result does not indicate the existence of an abnormal operating condition, it means that the execution effect of the target control parameter combination has not met the preset control requirements, but the operating condition of the air conditioning control system is not in an abnormal state. At this time, the energy-saving processing method is determined based on the energy-saving control evaluation result, the weight configuration of the dynamic optimization target data is adjusted, and the target control parameter combination is regenerated based on the adjusted dynamic optimization target data. After the new target control parameter combination is sent to the air conditioning control system, the system continues to collect operating result data and regenerates the energy-saving control evaluation result and anomaly judgment result until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition.

[0042] For example, if the energy-saving control evaluation results show that the indoor temperature meets the preset indoor temperature constraint, but the total active power exceeds the preset power consumption constraint, and the anomaly judgment result does not indicate the existence of an abnormal operating condition, it is determined that the energy consumption control effect of the current target control parameter combination under normal operating conditions does not meet the requirements. In this case, the weight of the energy consumption evaluation data in the dynamic optimization target data is increased, and the target control parameter combination is regenerated based on the adjusted dynamic optimization target data. After the new target control parameter combination is issued and executed, the operating result data is collected again, and energy-saving control evaluation results and anomaly judgment results are generated until the total active power meets the preset power consumption constraint, and the indoor temperature and comfort status meet the corresponding constraints.

[0043] When generating anomaly determination results, if the result indicates an abnormal operating condition, it means that the execution process of the target control parameter combination is affected by abnormal data acquisition, environmental disturbances, load fluctuations, or equipment malfunctions. In this case, the anomaly handling method is determined and executed based on the anomaly determination result, rather than directly treating the same operating result as a normal energy-saving failure and applying energy-saving measures. Anomaly handling methods include regenerating the target control parameter combination, switching to emergency control mode, or outputting anomaly alarm information. Different outputs are made according to the severity of the abnormal operating condition, so that the air conditioning control system operates according to the corresponding handling path under abnormal conditions.

[0044] For example, when key parameters are continuously missing, parameter changes exceed the normal operating range, equipment operating status is abnormal, or load status changes abruptly in the operation results data, the anomaly judgment result indicates the existence of an abnormal operating condition. In this case, even if the energy-saving control evaluation result does not simultaneously meet the preset energy-saving control conditions, the anomaly handling method is prioritized. For abnormal operating conditions that do not affect the continued adjustment of the air conditioning control system, the target control parameter regeneration path is entered; for abnormal operating conditions that affect the stable operation of the air conditioning control system, the emergency control path is entered; and for abnormal operating conditions that affect the effectiveness of data acquisition or the safety of control execution, the anomaly alarm path is entered.

[0045] The above settings can distinguish between situations where the control effect is not up to standard but the operating condition is normal, and situations where the operating condition is abnormal. For situations where the control effect is not up to standard but the operating condition is normal, the target control parameter combination is continuously corrected through re-prediction, re-optimization, re-issuance, and re-evaluation. For situations where the operating condition is abnormal, the corresponding processing path is entered according to the abnormality judgment result, avoiding the need to continue adjusting the control parameters according to the ordinary situation where energy saving is not up to standard, thereby improving the adaptability of air conditioning energy saving control to changes in actual operating conditions.

[0046] This application collects operational data of the air conditioning control system after the target control parameter combination is issued, and generates energy-saving control evaluation results and anomaly judgment results based on the operational data. This allows air conditioning energy-saving control to extend from the prediction and optimization stage to the actual execution stage. When the execution effect of the control parameters does not meet the preset energy-saving control conditions, the target control parameter combination is regenerated and re-evaluated. In the event of abnormal operating conditions, the corresponding anomaly handling method is determined based on the anomaly judgment results, avoiding the continued use of control parameters that are incompatible with the actual operating state or pose safety hazards, thereby improving the continuous adaptability of the prediction and optimization results in the execution stage.

[0047] As one implementation method, the measured indoor temperature data, measured total active power data, measured comfort data, and equipment operating status data are obtained after the target control parameter combination is executed. Based on the measured indoor temperature data, measured total active power data, and measured comfort data, determine whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and generate energy-saving control evaluation results. Based on the measured indoor temperature data, measured total active power data, measured comfort data, and equipment operating status data, determine whether there are any abnormalities in data acquisition, environmental disturbances, load fluctuations, or equipment operation, and generate anomaly judgment results. The preset energy-saving control conditions include at least one of indoor temperature constraints, power consumption constraints, and comfort constraints, and the measured comfort data include at least one of indoor relative humidity, air velocity, carbon dioxide concentration, and personnel density.

[0048] Specifically, measured indoor temperature data is compared with indoor temperature constraints to determine the temperature control result within the data acquisition area after the target control parameter combination is executed; measured total active power data is compared with power consumption constraints to determine the energy consumption control result after the target control parameter combination is executed; and measured comfort data is compared with comfort constraints to determine the comfort maintenance result after the target control parameter combination is executed. Based on the temperature control result, energy consumption control result, and comfort maintenance result, an energy-saving control evaluation result is generated.

[0049] Furthermore, equipment operating status data is used to characterize the equipment's operating status during the execution of the target control parameter combination by the air conditioning control system. By combining the completeness, continuity, and variation range of measured indoor temperature data, measured total active power data, and related measured comfort data, it is possible to determine whether there are missing, abrupt, or data exceeding the effective range in the operating results data, thereby identifying any data acquisition anomalies. By combining measured indoor temperature data, related measured comfort data, and changes in the external environment, it is possible to determine whether there are any environmental disturbance anomalies. By combining the relationship between changes in personnel density, carbon dioxide concentration, indoor temperature, and total active power, it is possible to determine whether there are any load fluctuation anomalies. By combining equipment operating status data, it is possible to determine whether there are any equipment operating anomalies. Based on the above anomaly determination results, anomaly determination results are generated.

[0050] The above settings enable the same operational result data to be used to generate control effect judgment and operational condition judgment respectively. The energy-saving control evaluation result reflects whether the target control parameter combination has met the control requirements in terms of temperature, power consumption, and comfort. The anomaly judgment result reflects whether the execution result has been affected by data acquisition, environmental disturbances, load fluctuations, or equipment operating status, thus providing a basis for distinguishing between ordinary energy-saving failures and abnormal operating conditions.

[0051] As one implementation method, when the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal working condition, the weight configuration of the dynamic optimization target data is determined as the energy-saving processing method. After the energy-saving processing method is executed, steps S2 to S5 are executed again until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal working condition. When the anomaly determination result indicates that there is a slight abnormal working condition, the weight configuration of the dynamic optimization target data will be corrected as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 will be executed again. When the anomaly determination result indicates that there is a moderate abnormal operating condition, the model parameters of the temperature-power coupling prediction model and the weight configuration of the dynamically optimized target data are determined as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 are executed again. When the anomaly determination result indicates that there is a severe abnormal operating condition, the air conditioning control mode is switched to the emergency control mode, and a combination of safety control parameters is generated based on the emergency control mode. When the anomaly determination result indicates the existence of extreme abnormal operating conditions, the energy-saving optimization control will be stopped and an anomaly alarm message will be output.

[0052] Specifically, the energy-saving treatment method corresponds to the situation where the operational result data is valid and the air conditioning control system is operating normally, but the execution effect of the target control parameter combination does not meet the preset energy-saving control conditions. In this case, the measured indoor temperature data, measured total active power data, and measured comfort data can be used as effective feedback data. By adjusting the weight configuration of the dynamically optimized target data, the emphasis relationship between temperature control, energy consumption control, and comfort maintenance in the control parameter evaluation process is changed, making the regenerated target control parameter combination more in line with the current actual control needs.

[0053] Specifically, the anomaly determination result refers to the result obtained after judging the operating conditions during the execution of the target control parameter combination based on the operating result data. It is used to characterize whether the operating result data is affected by data acquisition anomalies, environmental disturbances, load fluctuations, or equipment malfunctions. The anomaly determination result includes at least one of the following: mild anomaly, moderate anomaly, severe anomaly, and extreme anomaly. Different anomaly determination results are used to determine different anomaly handling methods.

[0054] Among them, a slightly abnormal operating condition refers to an operating condition in which there are short-term fluctuations or local deviations in the operating result data, but the operating result data can still be used for control parameter evaluation, and the air conditioning control system can continue to execute the target control parameter combination. Under this condition, the control parameter evaluation process is adjusted by modifying the weight configuration of the dynamically optimized target data. For example, during the execution of the target control parameter combination, if the measured indoor temperature data is higher than the upper limit of the target temperature range for a short period of time, and then returns to the target temperature range within a preset time, while the equipment operating status data remains normal, this type of operating condition is judged as a slightly abnormal operating condition. After judging the slightly abnormal operating condition, the influence weight of the temperature deviation data in the dynamically optimized target data is increased, and steps S2 to S5 are re-executed.

[0055] Among them, moderately abnormal operating conditions refer to operating conditions where the continuity of operating result data, environmental conditions, or load conditions have affected the input correspondence or stability of the temperature-power consumption coupling prediction model. Under this condition, the prediction process and the control parameter evaluation process are jointly adjusted by correcting the model parameters of the temperature-power consumption coupling prediction model and dynamically optimizing the weight configuration of the target data. For example, during the execution of the target control parameter combination, if the personnel density continues to increase, the carbon dioxide concentration, indoor temperature, and total active power rise simultaneously, and the deviation between the measured indoor temperature data and the predicted indoor temperature data continues to increase, this type of operating condition is judged as a moderately abnormal operating condition. After judging the moderately abnormal operating condition, the model parameters of the temperature-power consumption coupling prediction model are corrected according to the measured data of personnel density, carbon dioxide concentration, indoor temperature, and total active power, and the weight configuration of the dynamically optimized target data is adjusted, and steps S2 to S5 are re-executed.

[0056] Severe abnormal operating conditions refer to situations where equipment operation is abnormal or control execution is restricted, making the target control parameter combination unsuitable for continued execution according to the energy-saving optimization control method. In this condition, the air conditioning control mode is switched to emergency control mode, and a safe control parameter combination is generated based on this mode. For example, if, after the target control parameter combination is issued, the equipment operating status data indicates that the air supply actuator does not operate according to the target air supply volume, resulting in the actual air supply volume consistently being lower than the target volume, this type of condition is classified as a severe abnormal operating condition. Upon determining a severe abnormal operating condition, the issuance of target control parameter combinations generated from dynamically optimized target data is stopped, the air conditioning control mode is switched to emergency control mode, and a safe control parameter combination is generated according to this mode.

[0057] Extreme abnormal operating conditions refer to situations where operational result data loses its effective feedback function, or where the air conditioning control system no longer meets the conditions for continuing to execute energy-saving optimization control. Under such conditions, energy-saving optimization control is stopped, an abnormal alarm message is output, and manual intervention is required for further processing. For example, if communication between the air conditioning control system and the control terminal is continuously interrupted during the execution of the target control parameter combination, preventing the uploading of operational result data and the issuance of new control parameters, this type of condition is identified as an extreme abnormal operating condition. After identifying an extreme abnormal operating condition, energy-saving optimization control is stopped, an abnormal alarm message is output, and manual intervention is required for further processing.

[0058] It should be noted that when the energy-saving control evaluation result does not meet the preset energy-saving control conditions, and the anomaly judgment result does not indicate the existence of abnormal operating conditions, it means that the operating result data is in a valid state, and the air conditioning control system can normally execute the target control parameter combination. The reason for not meeting the standard corresponds to the mismatch between the emphasis relationship between temperature control, energy consumption control, and comfort maintenance in the dynamic optimization target data and the current control requirements. When the anomaly judgment result indicates the existence of a slight abnormal operating condition, it means that there have been short-term fluctuations or local deviations in the operating result data, but these short-term fluctuations or local deviations have not destroyed the validity of the operating result data as the basis for control parameter evaluation, and the air conditioning control system can continue to execute the target control parameter combination. In this case, the anomaly judgment result is used as the processing basis, and based on the state that the operating result data can still be used for control parameter evaluation, the weight configuration of the dynamic optimization target data is corrected, and steps S2 to S5 are re-executed.

[0059] The above settings allow us to first determine whether the operational results data is suitable for energy-saving feedback correction. For cases where the operating conditions are normal but the control effect is not up to standard, the target control parameter combination is regenerated by adjusting the weight configuration of the dynamically optimized target data. For cases with abnormal operating conditions, the corresponding anomaly handling path is entered based on the severity of the abnormality.

[0060] As one implementation method, the severity of abnormal operating conditions is determined based on at least one of the following: the degree of deviation between the operating result data and the corresponding predicted data, the duration of the abnormality, the type of abnormal data, and the equipment operating status. The corresponding prediction data includes at least one of indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0061] Specifically, when determining the severity of abnormal operating conditions, the operational results data are first compared with the corresponding predicted data. Specifically, the measured indoor temperature data corresponds to the predicted indoor temperature data, the measured total active power data corresponds to the predicted power consumption data, and the measured comfort-related data corresponds to the comfort evaluation data. Through these comparisons, the degree of deviation between the operational results data and the corresponding predicted data is determined. The degree of deviation between the operational results data and the corresponding predicted data is used to determine the extent to which the actual operating state deviates from the predicted operating state; the duration of the anomaly is used to determine whether the deviation is a short-term or continuous change; the type of anomaly data is used to distinguish whether the anomaly originates from temperature-related data, power consumption data, comfort-related data, or equipment operating status data; and the equipment operating status is used to determine whether the air conditioning control system can complete the execution according to the target control parameter combination.

[0062] Furthermore, the severity of the abnormal operating condition is determined based on the degree of deviation between the operational result data and the corresponding predicted data, the duration of the abnormality, the type of abnormal data, and the equipment operating status. If the deviation between the operational result data and the corresponding predicted data is small, the duration of the abnormality is short, the type of abnormal data corresponds to a single data point from the measured indoor temperature data, measured total active power data, or related comfort level data, and the equipment operating status is normal, the abnormal operating condition is classified as a mild abnormal operating condition. If the deviation between the operational result data and the corresponding predicted data is large, or the duration of the abnormality is long, the type of abnormal data corresponds to multiple data points from the measured indoor temperature data, measured total active power data, and related comfort level data, and the equipment operating status still indicates that the air conditioning control system can execute the target control parameter combination, the abnormal operating condition is classified as a moderate abnormal operating condition. If the equipment operating status indicates that the air conditioning control system's execution of the target control parameter combination is limited, the abnormal operating condition is classified as a severe abnormal operating condition. If the operational result data is continuously missing, or the target control parameter combination cannot be sent to the air conditioning control system, the abnormal operating condition is classified as an extreme abnormal operating condition.

[0063] With the above settings, the severity of abnormal operating conditions can be determined step by step based on the deviation between the operating result data and the corresponding predicted data, the duration of the abnormality, the source of the abnormal data, and the equipment execution status, thereby providing a clear basis for the selection of abnormality handling methods.

[0064] As one approach, the target data for dynamic optimization includes at least one of temperature deviation data, comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data. The weight configuration of the target data is determined based on the building area type, the outdoor environmental status determined by indoor and outdoor environmental parameters, and the human body status correlation parameters. Building area types include areas for human activity and areas for equipment operation; When the building area type is a human activity area, increase the influence weight of comfort deviation data in the dynamic optimization target data; When the building area type is an equipment operation area, increase the influence weight of energy consumption evaluation data and comfort deviation data used to characterize indoor temperature stability in the dynamic optimization target data; When outdoor environmental conditions or human condition-related parameters change, update the weighting configuration among comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data.

[0065] Specifically, the weighting configuration in the dynamically optimized target data includes weights for comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data. The comfort deviation data weights are used to determine the degree of influence of comfort deviation data in the control parameter evaluation process; the energy consumption evaluation data weights are used to determine the degree of influence of energy consumption evaluation data in the control parameter evaluation process; and the energy efficiency evaluation data weights are used to determine the degree of influence of energy efficiency evaluation data in the control parameter evaluation process.

[0066] Specifically, when determining the weight configuration, the weights for comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data are determined based on the building area type, outdoor environmental conditions, and human body status correlation parameters. The building area type is used to determine the control demand type of the data collection area; the outdoor environmental conditions are used to determine the impact of changes in the external environment on the energy consumption and energy efficiency of the air conditioning system; and the human body status correlation parameters are used to determine the impact of the number of people and their activity status on comfort requirements.

[0067] For example, when the building area type is a human activity area and the human state correlation parameter indicates an increase in human density, the weight of comfort deviation data is increased so that the control parameter evaluation results reflect human comfort needs more. When the building area type is an equipment operation area, the weight of energy consumption evaluation data is increased, as is the weight of comfort deviation data used to characterize indoor temperature stability, so that the control parameter evaluation results reflect the equipment operating environment stability requirements and energy consumption control requirements more. When the outdoor environment status indicates an increase in outdoor temperature or solar radiation intensity, the weight of energy consumption evaluation data and energy efficiency evaluation data is increased so that the control parameter evaluation results reflect the air conditioning operating load after changes in the external environment more.

[0068] Through the above settings, the dynamic optimization of target data can determine the degree of influence of different evaluation data in the evaluation process of control parameters based on the building area type, outdoor environmental conditions and human condition correlation parameters, so that the generation basis of target control parameter combination corresponds to the comfort requirements, energy consumption control requirements and energy efficiency control requirements of the current collection area.

[0069] As one implementation method, the temperature-power coupling prediction model includes a feature weight allocation layer; The feature weight allocation layer generates the influence weight of each input parameter based on the temporal variation characteristics of each input parameter in the multidimensional data matrix, the correlation characteristics between parameters, and the degree of influence of each input parameter on indoor temperature prediction data, power consumption prediction data, and comfort evaluation data. The temperature-power consumption coupled prediction model weights the multidimensional data matrix according to the influence weights and outputs indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0070] Specifically, the feature weight allocation layer is used to assign weights to the input parameters in the multidimensional data matrix when the temperature-power coupled prediction model makes predictions. The feature weight allocation layer reads the air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body status-related parameters according to the data type, reads the changes of the same input parameter in continuous acquisition time according to the acquisition time, and determines the correspondence of different input parameters in the same acquisition area according to the acquisition area.

[0071] Specifically, during the weighting process, the feature weight allocation layer determines the influence weights of the input parameters for the indoor temperature prediction data, power consumption prediction data, and comfort evaluation data, respectively. For the indoor temperature prediction data, the feature weight allocation layer determines the influence weights based on the correspondence between the input parameters and changes in indoor temperature; for the power consumption prediction data, the feature weight allocation layer determines the influence weights based on the correspondence between the input parameters and changes in total active power; and for the comfort evaluation data, the feature weight allocation layer determines the influence weights based on the correspondence between the input parameters and changes in indoor relative humidity, air velocity, carbon dioxide concentration, and occupant density.

[0072] For example, during continuous data collection, when changes in outdoor temperature correspond to changes in indoor temperature and total active power within the collection area, the feature weight allocation layer increases the influence weight of outdoor temperature in indoor temperature and power consumption prediction. Similarly, when changes in population density correspond to changes in carbon dioxide concentration, the feature weight allocation layer increases the influence weight of population density and carbon dioxide concentration in comfort evaluation. The temperature-power consumption coupled prediction model weights the input parameters in the multidimensional data matrix according to the influence weights and outputs indoor temperature prediction data, power consumption prediction data, and comfort evaluation data.

[0073] With the above settings, the temperature-power consumption coupled prediction model can distinguish the degree of influence of different input parameters on different output results during the prediction process, so that indoor temperature prediction, power consumption prediction and comfort evaluation have corresponding input parameter basis.

[0074] As one implementation method, the target control parameter combination includes at least one of the indoor set temperature, supply air temperature, and supply air volume; Based on dynamically optimized target data, a combination of target control parameters is generated, including: Determine the candidate value range for each control parameter in the target control parameter combination; Generate multiple candidate combinations of control parameters within the candidate value range; Based on the dynamic optimization target data, evaluate the comfort deviation, energy consumption status and energy efficiency status corresponding to each candidate combination of control parameters. The target control parameter combination is determined from the candidate combinations of control parameters that satisfy comfort constraints, energy consumption constraints, and equipment operation constraints.

[0075] Specifically, candidate value ranges for each control parameter in the target control parameter combination are determined based on historical operating data. This historical operating data reflects the control parameter values ​​under different operating conditions of the air conditioning system. Determining the candidate value ranges for the indoor set temperature, supply air temperature, and supply air volume based on this historical data ensures that the candidate control parameter combinations remain within the executable range of the air conditioning system. After generating multiple candidate control parameter combinations within the candidate value ranges, each candidate combination is evaluated using dynamic optimization target data to determine the corresponding comfort deviation, energy consumption status, and energy efficiency status. Candidate control parameter combinations that do not meet comfort constraints, energy consumption constraints, or equipment operation constraints are not considered as target control parameter combinations. For candidate control parameter combinations that meet these constraints, the target control parameter combination is determined based on the evaluation results of the comfort deviation, energy consumption status, and energy efficiency status.

[0076] The above settings enable the target control parameter combination to be determined from the candidate combinations of control parameters that can be executed by the air conditioning system, while simultaneously corresponding to comfort constraints, energy consumption constraints, and equipment operation constraints, thus avoiding the generation of control parameter combinations that exceed the equipment's execution capabilities or do not meet control requirements.

[0077] As one implementation method, the operating parameters of the air conditioning system include at least one of the following: supply air temperature, return air temperature, supply air volume, total active power, and indoor set temperature; Indoor and outdoor environmental parameters include at least one of the following: indoor temperature, relative humidity, air velocity, carbon dioxide concentration, outdoor temperature, solar radiation intensity, and outdoor wind speed. Human body status-related parameters include at least one of indoor personnel density and personnel activity intensity.

[0078] Specifically, the air conditioning system operating parameters are used to characterize the operating status of the air conditioning control system itself. Among them, the supply air temperature is used to characterize the temperature of the air supplied by the air conditioning system to the collection area, the return air temperature is used to characterize the temperature of the air returning to the collection area, the supply air volume is used to characterize the flow rate of the air supplied by the air conditioning system to the collection area, the total active power is used to characterize the power consumption status of the air conditioning system, and the indoor set temperature is used to characterize the target temperature currently set by the air conditioning control system.

[0079] Indoor and outdoor environmental parameters are used to characterize the environmental conditions inside and outside the collection area. Among them, indoor temperature, relative humidity and air velocity are used to characterize the thermal and humid environment conditions in the collection area, carbon dioxide concentration is used to characterize the air conditions in the collection area, and outdoor temperature, solar radiation intensity and outdoor wind speed are used to characterize the impact of the external environment of the collection area on the operation of the air conditioning system.

[0080] Human body status correlation parameters are used to characterize the usage status of people in the data collection area. Among them, indoor personnel density is used to characterize the distribution of the number of people in the data collection area, and personnel activity intensity is used to characterize the impact of personnel activities in the data collection area on comfort needs.

[0081] Through the above settings, the operating status of the air conditioning system itself, the status of the external environment within the collection area, and the status of personnel use can be represented by corresponding parameters, making the multi-source sensing data more accurate.

[0082] Based on the same inventive concept, this application also provides an air conditioning energy-saving control system 100 based on multi-source sensing and feedback correction optimization for implementing the air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization mentioned above. The solution provided by the air conditioning energy-saving control system 100 based on multi-source sensing and feedback correction optimization is similar to the solution described in the above method. Therefore, the specific limitations of the air conditioning energy-saving control system 100 based on multi-source sensing and feedback correction optimization provided below can be found in the limitations of the air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization mentioned above, and will not be repeated here.

[0083] like Figure 3 As shown, as one implementation, this application provides an air conditioning energy-saving control system 100 based on multi-source sensing and feedback correction optimization. This air conditioning energy-saving control system 100 includes: Multi-source sensing module 101 is used to collect air conditioning system operating parameters, indoor and outdoor environmental parameters and human status related parameters; The data matrix construction module 102 is used to construct a multi-dimensional data matrix based on the collection time, collection area and data type. The coupled prediction module 103 is used to input a multi-dimensional data matrix into the temperature-power consumption coupled prediction model and output indoor temperature prediction data, power consumption prediction data and comfort evaluation data. The dynamic optimization module 104 is used to generate dynamic optimization target data based on indoor temperature prediction data, power consumption prediction data and comfort evaluation data, and to generate a combination of target control parameters based on the dynamic optimization target data; The control sending module 105 is used to send the target control parameter combination to the air conditioning control system; The operation monitoring module 106 is used to collect the operation result data of the air conditioning control system after executing the target control parameter combination; The evaluation and judgment module 107 is used to generate energy-saving control evaluation results and anomaly judgment results based on the operation result data; The hierarchical processing module 108 is used to determine the energy-saving processing method based on the energy-saving control evaluation result when the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly judgment result does not indicate the existence of an abnormal operating condition, and to generate a new combination of target control parameters based on the energy-saving processing method; it is also used to determine the anomaly processing method based on the anomaly judgment result when the anomaly judgment result indicates the existence of an abnormal operating condition, and to execute the target control parameter regeneration path, emergency control path or anomaly alarm path based on the anomaly processing method.

[0084] As an optional implementation, the system also includes: Storage module 109 is used to store historical running data, model parameters, target control parameter combinations, running logs, and abnormal sample data; Communication module 110 is used to realize data transmission between the system and the air conditioning control system; The human-computer interaction module 111 is used to display real-time operating data, prediction results, target control parameter combinations, energy-saving control evaluation results, anomaly judgment results, and system operating status.

[0085] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0086] The above-described specific embodiments are preferred embodiments of an air conditioning energy-saving control method and system based on multi-source sensing and feedback correction optimization according to this application. They are not intended to limit the specific scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent substitutions or modifications made under the technical concept disclosed in this application should fall within the protection scope of this application.

Claims

1. An air conditioning energy-saving control method based on multi-source sensing and feedback correction optimization, characterized in that, Includes the following steps: S1. Collect air conditioning system operating parameters, indoor and outdoor environmental parameters, and human body status related parameters, and construct a multi-dimensional data matrix based on the collection time, collection area, and data type; S2. Input the multidimensional data matrix into the temperature-power consumption coupled prediction model, and output indoor temperature prediction data, power consumption prediction data and comfort evaluation data. S3. Generate dynamic optimization target data based on the indoor temperature prediction data, the power consumption prediction data, and the comfort evaluation data, and generate a target control parameter combination based on the dynamic optimization target data; S4. Send the target control parameter combination to the air conditioning control system and collect the operation result data of the air conditioning control system after executing the target control parameter combination; S5. Based on the operation result data, generate an energy-saving control evaluation result to characterize whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and an anomaly judgment result to characterize whether there is an abnormal operating condition. S6. When the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly determination result does not indicate the existence of the abnormal working condition, an energy-saving processing method is determined according to the energy-saving control evaluation result. After the energy-saving processing method is executed, steps S2 to S5 are executed again until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly determination result does not indicate the existence of the abnormal working condition. When the anomaly determination result indicates the existence of the abnormal operating condition, an anomaly handling method is determined based on the anomaly determination result, and the anomaly handling method is executed.

2. The method according to claim 1, characterized in that, S5 include: Acquire the measured indoor temperature data, measured total active power data, measured comfort data, and equipment operating status data after the target control parameter combination is executed; Based on the measured indoor temperature data, the measured total active power data, and the measured comfort data, determine whether the execution result of the target control parameter combination meets the preset energy-saving control conditions, and generate the energy-saving control evaluation result. Based on the measured indoor temperature data, the measured total active power data, the measured comfort data, and the equipment operating status data, determine whether there are any abnormalities in data acquisition, environmental disturbances, load fluctuations, or equipment operation, and generate the abnormality judgment result. The preset energy-saving control conditions include at least one of indoor temperature constraints, power consumption constraints, and comfort constraints, and the measured comfort data includes at least one of indoor relative humidity, air velocity, carbon dioxide concentration, and personnel density.

3. The method according to claim 1, characterized in that, S6 include: When the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly determination result does not indicate the existence of the abnormal working condition, the weight configuration of the dynamic optimization target data will be corrected to determine the energy-saving processing method. After the energy-saving processing method is executed, steps S2 to S5 will be executed again until the energy-saving control evaluation result meets the preset energy-saving control conditions and the anomaly determination result does not indicate the existence of the abnormal working condition. When the anomaly determination result indicates that there is a slight abnormal working condition, the weight configuration of the dynamic optimization target data will be corrected as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 will be executed again. When the anomaly determination result indicates that there is a moderate abnormal operating condition, the model parameters of the temperature-power coupling prediction model and the weight configuration of the dynamic optimization target data are corrected and determined as the anomaly handling method. After the anomaly handling method is executed, steps S2 to S5 are executed again. When the anomaly determination result indicates the existence of a severe abnormal operating condition, the air conditioning control mode is switched to the emergency control mode, and a combination of safety control parameters is generated based on the emergency control mode. When the anomaly determination result indicates the existence of extreme abnormal operating conditions, the energy-saving optimization control is stopped and an anomaly alarm message is output.

4. The method according to claim 3, characterized in that, The severity of the abnormal operating condition is determined based on at least one of the following: the degree of deviation between the operating result data and the corresponding predicted data, the duration of the abnormality, the type of abnormal data, and the equipment operating status. The corresponding prediction data includes at least one of the indoor temperature prediction data, the power consumption prediction data, and the comfort evaluation data.

5. The method according to claim 1, characterized in that, The dynamic optimization target data includes at least one of temperature deviation data, comfort deviation data, energy consumption evaluation data, and energy efficiency evaluation data; The weight configuration of the dynamically optimized target data is determined based on the building area type, the outdoor environmental state determined by the indoor and outdoor environmental parameters, and the human body state association parameters. The building area types include personnel activity areas and equipment operation areas; When the building area type is a human activity area, increase the influence weight of the comfort deviation data in the dynamic optimization target data; When the building area type is an equipment operation area, increase the influence weight of the energy consumption evaluation data and the comfort deviation data used to characterize indoor temperature stability in the dynamic optimization target data; When the outdoor environmental conditions or the parameters related to the human body conditions change, the weight configuration among the comfort deviation data, the energy consumption evaluation data, and the energy efficiency evaluation data is updated.

6. The method according to claim 1, characterized in that, The temperature-power coupling prediction model includes a feature weight allocation layer; The feature weight allocation layer generates the influence weight of each input parameter based on the temporal variation characteristics of each input parameter in the multidimensional data matrix, the correlation characteristics between parameters, and the degree of influence of each input parameter on the indoor temperature prediction data, the power consumption prediction data, and the comfort evaluation data. The temperature-power consumption coupled prediction model performs weighted processing on the multidimensional data matrix according to the influence weights, and outputs the indoor temperature prediction data, the power consumption prediction data, and the comfort evaluation data.

7. The method according to claim 1, characterized in that, The target control parameter combination includes at least one of indoor set temperature, supply air temperature and supply air volume; Based on the dynamically optimized target data, a combination of target control parameters is generated, including: Determine the candidate value range for each control parameter in the target control parameter combination; Multiple candidate combinations of control parameters are generated within the range of candidate values; Based on the dynamic optimization target data, evaluate the comfort deviation, energy consumption status and energy efficiency status corresponding to each candidate combination of control parameters; The target control parameter combination is determined from the candidate combinations of control parameters that satisfy comfort constraints, energy consumption constraints, and equipment operation constraints.

8. The method according to claim 1, characterized in that, The operating parameters of the air conditioning system include at least one of the following: supply air temperature, return air temperature, supply air volume, total active power, and indoor set temperature. The indoor and outdoor environmental parameters include at least one of indoor temperature, relative humidity, air velocity, carbon dioxide concentration, outdoor temperature, solar radiation intensity, and outdoor wind speed. The human body status correlation parameters include at least one of indoor personnel density and personnel activity intensity.

9. An air conditioning energy-saving control system based on multi-source sensing and feedback correction optimization, characterized in that, For performing the method according to any one of claims 1 to 8, comprising: A multi-source sensing module is used to collect the operating parameters of the air conditioning system, the indoor and outdoor environmental parameters, and the human body status-related parameters. A data matrix construction module is used to construct the multidimensional data matrix based on the acquisition time, the acquisition area, and the data type. The coupled prediction module is used to input the multidimensional data matrix into the temperature-power consumption coupled prediction model and output the indoor temperature prediction data, the power consumption prediction data and the comfort evaluation data. The dynamic optimization module is used to generate the dynamic optimization target data based on the indoor temperature prediction data, the power consumption prediction data, and the comfort evaluation data, and to generate the target control parameter combination based on the dynamic optimization target data; A control sending module is used to send the target control parameter combination to the air conditioning control system; The operation monitoring module is used to collect the operation result data of the air conditioning control system after executing the target control parameter combination; The evaluation and judgment module is used to generate the energy-saving control evaluation result and the anomaly judgment result based on the operation result data; The hierarchical processing module is used to determine an energy-saving processing method based on the energy-saving control evaluation result and generate a new target control parameter combination based on the energy-saving processing method when the energy-saving control evaluation result does not meet the preset energy-saving control conditions and the anomaly determination result does not indicate the existence of the abnormal operating condition; it is also used to determine an anomaly processing method based on the anomaly determination result when the anomaly determination result indicates the existence of the abnormal operating condition, and execute a target control parameter regeneration path, emergency control path, or anomaly alarm path based on the anomaly processing method.

10. The system according to claim 9, characterized in that, Also includes: The storage module is used to store historical running data, model parameters, target control parameter combinations, running logs, and abnormal sample data. The communication module is used to realize data transmission between the system and the air conditioning control system; The human-computer interaction module is used to display real-time operating data, prediction results, target control parameter combinations, energy-saving control evaluation results, anomaly judgment results, and system operating status.

Images