Direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing

Abstract

The invention provides a direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing, and relates to the technical field of air conditioning control, and the method comprises the steps that temperature and humidity data, intensity data and external environment data of a target area are collected based on a cloud computing platform, and update data are generated; constructing an air conditioning model according to the updating record and the air conditioning record; carrying out multi-region transfer learning on the air conditioning model to obtain a transfer model; and performing air conditioning prediction on the migration model through the updated data, performing air dynamic control according to a prediction result, and generating a control strategy. According to the method and the device, the technical problem that the overall operation efficiency is further influenced due to overlong model training time caused by independent training of each regional air conditioning model and lack of an effective experience sharing mechanism in the prior art can be solved, the technical targets of cross-regional model sharing and rapid regional environment adaptation are achieved, and the method and the device are suitable for popularization and application. The technical effects of improving the air conditioning efficiency and reducing the energy consumption are achieved.

Description

Technical Field

[0001] The present application relates to the field of air conditioning control technology, and in particular to a direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing. Background Art

[0002] With the acceleration of urbanization and the improvement of people's requirements for a comfortable living environment, air conditioning systems are becoming more and more common in various buildings. Among them, direct expansion constant temperature and humidity machines are widely used in public places and production environments such as shopping malls, office buildings, subways, and dust-free workshops due to their high efficiency and good temperature and humidity control performance. However, the existing air conditioning control methods still have some technical defects, which limit their further development in intelligence and efficiency.

[0003] At present, traditional air conditioning methods are mostly based on customized control strategies. However, the existing air conditioning technology of direct expansion constant temperature and humidity machines mainly relies on independent control systems in each area when dealing with complex and dynamic environmental conditions, especially in environments such as dust-free workshops that have extremely high requirements for temperature and humidity control. As a result, the air conditioning model of each area needs to be trained from scratch, which is not only time-consuming, but also consumes a lot of manpower and computing resources. In addition, the environmental characteristics and passenger flow patterns of each area often have similarities, but the existing system cannot effectively share the experience and model parameters brought by these similarities, which limits the flexibility and efficiency of the technology. For example, parameters such as temperature, humidity, and density in different dust-free workshops may show similar trends of change during peak production and non-peak production periods. If each workshop conducts model training independently, it may take days or even weeks to adapt to new environmental conditions, which not only affects the effect of real-time control, but also increases the overall energy consumption of the system.

[0004] In summary, there are technical problems in the prior art due to the independent training of air conditioning models in each area and the lack of an effective experience sharing mechanism, which leads to long model training time and excessive resource consumption, further affecting the system's ability to respond to dynamic environmental changes and overall operating efficiency. Summary of the invention

[0005] The purpose of this application is to provide a direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing, so as to solve the technical problems in the prior art that the air conditioning models of each area are independently trained and there is a lack of an effective experience sharing mechanism, resulting in long model training time and excessive resource consumption, which further affects the system's responsiveness to dynamic environmental changes and overall operating efficiency.

[0006] In view of the above problems, the present application provides a direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing.

[0007] In the first aspect, the present application provides a direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing, which is implemented through a direct expansion constant temperature and humidity machine air conditioning control system based on cloud computing, including: collecting temperature and humidity data, density data and external environment data of the target area based on a cloud computing platform to generate update data; constructing an air conditioning model based on update records and air conditioning records; performing multiple area transfer learning on the air conditioning model to obtain a transfer model; performing air conditioning prediction on the transfer model through the update data, performing air dynamic control according to the prediction results, and generating a control strategy.

[0008] In the second aspect, the present application also provides a direct expansion constant temperature and humidity machine air conditioning control system based on cloud computing, which is used to execute the direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing as described in the first aspect, including: an update data generation module, the update data generation module is used to collect temperature and humidity data, density data and external environment data of the target area based on the cloud computing platform to generate update data; an air conditioning model construction module, the air conditioning model construction module is used to construct an air conditioning model according to update records and air conditioning records; a migration model acquisition module, the migration model acquisition module is used to perform multiple area migration learning on the air conditioning model to obtain a migration model; a control strategy generation module, the control strategy generation module is used to perform air conditioning prediction on the migration model through the update data, perform air dynamic control according to the prediction results, and generate a control strategy.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The temperature and humidity data, density data and external environment data of the target area are collected based on the cloud computing platform to generate update data; an air conditioning model is constructed according to the update records and air conditioning records; the air conditioning model is subjected to multi-region transfer learning to obtain a migration model; air conditioning prediction is performed on the migration model through the update data, and air dynamic control is performed according to the prediction results to generate a control strategy. In other words, by achieving the technical goals of cross-regional model sharing and rapid adaptation to the regional environment, the technical effect of improving air conditioning efficiency and reducing energy consumption is achieved.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 This is a flow chart of a direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing in this application;

[0014] Figure 2 This is a schematic diagram of the structure of the direct expansion constant temperature and humidity air conditioning control system based on cloud computing in this application.

[0015] Description of reference numerals:

[0016] An update data generating module 11 , an air conditioning model building module 12 , a migration model obtaining module 13 , and a control strategy generating module 14 . DETAILED DESCRIPTION

[0017] This application provides a direct expansion constant temperature and humidity machine air conditioning control method and system based on cloud computing, which solves the technical problem in the prior art that the model training time is too long and the resource consumption is too large due to the independent training of the air conditioning model in each region and the lack of an effective experience sharing mechanism, which further affects the system's ability to respond to dynamic environmental changes and overall operating efficiency. The technical goal of cross-regional model sharing and rapid adaptation to the regional environment is achieved, and the technical effect of improving air conditioning efficiency and reducing energy consumption is achieved.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] For example, please refer to the attached Figure 1 The present application provides a direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing, which is applied to a direct expansion constant temperature and humidity machine air conditioning control system based on cloud computing, and specifically includes the following steps:

[0020] Step 1: Collect temperature and humidity data, density data and external environment data of the target area based on the cloud computing platform to generate updated data.

[0021] Specifically, temperature and humidity data are obtained from the target area in real time, where temperature data is the degree of hotness or coldness of the air, and humidity data indicates the moisture content in the air. At the same time, density data indicates the flow of people or the density of objects within a unit area. External environmental data includes parameters such as outdoor temperature, humidity, air pressure, and wind speed to help calculate the difference between indoor and outdoor environments. These data are continuously transmitted to the control center through the cloud computing platform for centralized processing and analysis, thereby generating real-time updated data to provide a basis for subsequent temperature and humidity control strategies.

[0022] Step 2: Construct an air conditioning model based on the update records and air conditioning records.

[0023] Specifically, an air conditioning model is constructed based on the data of update records and air conditioning records. First, the update record is a data record of the changes of various environmental parameters over time. The air conditioning record represents the specific adjustment operation record made in response to environmental changes. By integrating the update record with the corresponding air conditioning record, the control requirements under different environmental conditions are analyzed to construct an air conditioning model. The air conditioning model can predict the control strategy so that the temperature and humidity are always kept within a comfortable range to adapt to different densities and environmental changes.

[0024] Step 3: Perform multiple area transfer learning on the air conditioning model to obtain a transfer model.

[0025] Specifically, by performing transfer learning on the constructed air conditioning model in multiple regions, the control strategy can be applied and optimized in different regions. Transfer learning refers to the process of applying the existing model knowledge in one region to another region. For example, if a model performs well in a high temperature and high humidity region, then through transfer learning, this model can be applied to a new region with similar temperature and humidity. Through migration, a migration model that adapts to the new environment can be quickly obtained, thus avoiding the time-consuming process of retraining from scratch.

[0026] Step 4: Perform air conditioning prediction on the migration model through the updated data, perform air dynamic control according to the prediction result, and generate a control strategy.

[0027] Specifically, the migration model is used to make air conditioning predictions through updated data, and the air is dynamically controlled based on the prediction results. That is, the settings of the direct expansion constant temperature and humidity machine are adjusted according to the real-time temperature and humidity requirements in order to adapt to environmental changes and keep the temperature and humidity within an appropriate range, thereby ensuring air quality.

[0028] The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method is applied to the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system, which can achieve the technical goals of cross-regional model sharing and rapid adaptation to regional environment, and achieve the technical effects of improving air conditioning efficiency and reducing energy consumption.

[0029] Furthermore, the present application also includes:

[0030] The temperature and humidity data, density data and external environment data are used as update features; a rolling window is configured with a fixed time length, and the rolling window is updated according to the fixed time length for the update features to generate the update data.

[0031] Specifically, temperature and humidity data, density data and external environment data are used as update features. For example, temperature can reflect the heating and cooling needs of the space, humidity shows the moisture content of the air, density represents the heat load caused by the flow of people, and the external environment provides data such as outdoor temperature and humidity. These updated features constitute the basic data source that can comprehensively evaluate the current environment.

[0032] The rolling window is configured with a fixed duration to divide the data collection period into a set duration so that data collection can be performed on the update feature within this time range. For example, setting the rolling window to 24 hours means that the selected update feature data is collected and analyzed every 24 hours to ensure that the latest environmental data is focused on. Fixed duration configuration can improve the adaptability of the model because the fixed duration only retains the most recent environmental data to prevent outdated data from affecting decision making.

[0033] Update the update features in a rolling window according to a fixed duration, and periodically update the update features such as temperature, humidity, density, and external environment data according to the set time period. The generated update data is the latest summary of the environmental conditions within the fixed time period. For example, in a 24-hour rolling window, statistics are collected on the average temperature, humidity range, maximum density, and other information for the past 24 hours, and the old data is replaced in the next time period to ensure the timeliness and reliability of the analysis results.

[0034] By using temperature, humidity, density and external environment data as update features and setting a fixed duration for a rolling window, these features are updated regularly to generate the latest update data, ensuring that the model adapts to the current environmental dynamics in real time and improving the accuracy of prediction and adjustment.

[0035] Furthermore, the present application also includes:

[0036] Based on the cloud computing platform, the temperature and humidity data characteristics, density data characteristics and external environment data characteristics of the target area are collected to obtain target characteristics; similar area groups are set for multiple areas according to the target characteristics to obtain similar area groups; the air conditioning model is shared among the similar area groups to obtain the migration model.

[0037] Specifically, multiple features of the target area are collected based on the cloud computing platform, and real-time monitoring of temperature and humidity, crowd density or item density, and external environmental data is performed. For example, the temperature and humidity data features include the temperature of the subway station being 27 degrees Celsius and the humidity being 60%; the density data features include the passenger density being 3 people per square meter; and the external environmental data features include the external temperature being 30 degrees Celsius and the humidity being 55%. These feature data are centrally managed through the cloud computing platform to achieve continuous updating of the target area features, providing a reliable data basis for subsequent analysis.

[0038] Similar area groups are set for multiple areas according to target features, that is, areas with similar environmental conditions are divided into the same group. Similar area groups are set based on the similarity of parameters such as temperature, humidity, density, and external environmental parameters. For example, multiple subway stations with temperatures close to 25 to 27 degrees Celsius, humidity around 60%, and similar peak passenger flow density are selected and set as one group. After classification, similar areas can share air conditioning models.

[0039] Then, the air conditioning model is shared among the similar area groups, that is, the optimized air conditioning model is applied to multiple similar areas in the group to obtain a migration model.

[0040] Sharing models saves retraining time and allows new areas to adapt to environmental requirements more quickly while ensuring that temperature and humidity are within ideal ranges.

[0041] Furthermore, the present application also includes:

[0042] Taking the target feature as a standard, the feature similarity of the multiple regions is measured to obtain similar features; updating is performed according to the rolling windows of the multiple regions, and the updated results are fed back to the similar features to obtain abnormal feedback results; based on the abnormal feedback results, the group conditions of the similar features are updated, and the groups are set according to the updated group conditions to obtain the similar region groups.

[0043] Specifically, the target feature is used as a standard to measure the feature similarity of multiple regions and obtain similar features, which are used to subsequently identify multiple regions with similar environmental conditions.

[0044] Then, the rolling windows of multiple regions are updated, and similar features are updated with data from a fixed time period, that is, a fixed time period is captured with a rolling window, and data is collected within the fixed time period. The rolling window represents data collection that focuses only on the time data within the rolling window, such as collecting environmental data changes from the past 48 hours to the past 24 hours and environmental data changes from the past 24 hours. In this way, real-time updated data can be continuously obtained and fed back to similar features, and slight changes in environmental conditions can be captured in a timely manner to more accurately reflect the dynamic status of each region.

[0045] The group conditions of similar features are updated based on the abnormal feedback results, and by analyzing the feedback, it can be decided whether to adjust the group of the area. For example, if a certain area no longer meets the temperature and humidity standards of the current group due to abnormal feedback, it will be removed from the current group or added to a new group by resetting the group conditions. When an abnormal value different from the usual value appears in the dynamic state, it indicates that there is a special case or environmental change, and the corresponding similar features are eliminated. If the features of the area still belong to similar features later, the group division is performed again. For example, if the temperature drops sharply in the environmental data of an area, the area may be deactivated, and the abnormal feedback result is recorded.

[0046] Dynamic updates of the groups ensure that each group's area is always maintained under similar environmental conditions. The resulting similar area groups have the ability to dynamically adapt to environmental changes, enabling more efficient and accurate air conditioning management of multiple areas.

[0047] Furthermore, the present application also includes:

[0048] The density data is evaluated for switching strength to generate switching strength; a switching strategy is generated according to the switching strength and the prediction result; the operation mode is switched based on the switching strategy to generate a switching result; the migration model is short-term adjusted in response to the switching result and fed back to the migration model to generate a control strategy.

[0049] Specifically, the density data is evaluated for switching intensity, that is, the frequency of air conditioning is evaluated according to the change in the density of people or objects in the area. The density data represents the flow of people or the density of objects per unit area. For example, if the density increases from 3 people per square meter to 5 people per square meter, it may have an impact on the load of air conditioning. The generation process of switching intensity is to determine the intensity or amplitude of adjustment based on the change in density data. If the density increase is large, the switching intensity is high.

[0050] Integrate switching force data with predictions of future environmental conditions to determine the operating mode to switch to in different situations. The prediction results include air conditioning methods based on real-time environmental factors such as temperature and humidity. Based on the combination of these factors, the switching strategy can determine a more appropriate control method, such as increasing fan speed or reducing the amount of fresh air introduced to adapt to the upcoming environmental changes.

[0051] Switching the operating mode based on the switching strategy will adjust the operating mode of the direct expansion constant temperature and humidity machine according to the switching strategy, such as switching from partial fresh air to fresh air mode, or from mechanical ventilation mode to mixed mode. The switching result is the feedback after the actual execution. For example, the temperature is reduced by 2 degrees Celsius and the humidity is reduced by 5% in the fresh air mode, indicating that the switching strategy has taken effect. Mode switching ensures greater flexibility in dealing with dynamic changes in density and environmental conditions.

[0052] The current migration model is adjusted based on the actual feedback data of the switching results. For example, if the temperature and humidity control does not achieve the expected effect after switching to the new mode, the migration model of the corresponding area will be adjusted, such as temporarily increasing the cooling power or lowering the valve opening, and then feedback will be given to the migration model to help the migration model gradually adapt to the specific environmental conditions of the area, and finally generate a control strategy so that it can automatically respond to similar situations in the future.

[0053] By evaluating the switching intensity of density data and then generating a switching strategy based on the prediction results, the operating mode is switched through the execution of the strategy and the switching results are obtained. The migration model is adjusted in a short time based on the feedback of the results to optimize the control. The control strategy finally generated enables the system to have a higher response speed and accuracy, so as to ensure a stable and comfortable air environment under the condition of density changes and environmental fluctuations.

[0054] Furthermore, the present application also includes:

[0055] An internal environment enthalpy value is evaluated by taking the temperature and humidity data and the density data as internal environment evaluation features to generate an internal environment enthalpy value; an external environment enthalpy value is evaluated by taking the external environment data as external environment evaluation features to generate an external environment enthalpy value; an enthalpy value difference between the internal environment enthalpy value and the external environment enthalpy value is obtained, and a preliminary switching strategy is generated based on the enthalpy value difference; the preliminary switching strategy is used as a switching guide and the switching strategy is used as a switching execution, a matching rate between the switching guide and the switching execution is evaluated, a switching result accuracy is generated, and the migration model is adjusted according to the switching result accuracy.

[0056] Specifically, the indoor temperature, humidity and human density data or object density are used as evaluation parameters to calculate the enthalpy value in the air, that is, the total heat contained in the air. For example, in an area, if the temperature is 28 degrees Celsius, the humidity is 65%, and the density is 3 people per square meter, the internal environment enthalpy value is calculated, which represents the heat content of the indoor air.

[0057] The outdoor environmental data is used as input to calculate the enthalpy value of the external environment through parameters such as external temperature, humidity, and wind speed. For example, if the outdoor temperature is 32 degrees Celsius, the humidity is 50%, and the wind speed is 4 meters per second, it is used to evaluate the external environmental enthalpy value. The external environmental enthalpy value reflects the heat content of the outside air and provides a basis for further adjusting the indoor and outdoor air exchange.

[0058] Compare the difference in enthalpy between the indoor and outdoor environments. For example, if the indoor enthalpy is 85 kJ / kg air and the outdoor enthalpy is 70 kJ / kg air, the enthalpy difference is 15 kJ / kg air. Based on the enthalpy difference, the demand for indoor and outdoor air conditioning can be determined, thereby generating a preliminary switching strategy, such as deciding whether to introduce outdoor fresh air or increase cooling under this condition.

[0059] The preliminary switching strategy is used as a switching guide, that is, the preliminary switching strategy is used as a reference for the actual switching operation. At the same time, the switching strategy predicted by the air conditioning model is used as the actual switching execution, and the results are monitored. Afterwards, the matching rate between the switching guide and the switching execution is evaluated, that is, the consistency between the assumed switching strategy and the actual switching operation is checked. If the consistency is high, it means that the switching strategy is more suitable for the current air conditioning scenario, and the lower the consistency, the less suitable the switching strategy is for the scenario.

[0060] By evaluating the matching rate between switching guidance and switching execution, the applicability of the strategy is evaluated, and the control strategy is further optimized.

[0061] Furthermore, the present application also includes:

[0062] Adaptively optimize the migration model whose switching result accuracy does not meet the error threshold to generate an optimized migration model; evaluate the adaptability of the optimized migration model according to the prediction accuracy to generate a adaptability coefficient; and update the model structure of the optimized migration model whose adaptability coefficient is lower than a preset adaptability coefficient.

[0063] Specifically, the migration model whose switching result accuracy meets the error threshold is adaptively optimized, that is, the model whose switching effect does not meet expectations and whose error is outside the acceptable range is further adjusted and optimized, and the adaptive optimization process of the migration model is started. The adaptive optimization process includes adjusting the model parameters to improve the model's adaptability to the current environmental conditions, or using other model architectures to generate another air conditioning model as an optimized migration model to ensure that the model can better cope with environmental changes in subsequent data processing and prediction.

[0064] The adaptability of the optimized migration model is evaluated based on the prediction accuracy. After optimization, the prediction results of the model are comprehensively evaluated to determine its performance under different environmental conditions. The adaptability coefficient is an indicator to measure the adaptability of the optimized migration model. It is generated by comparing the prediction accuracy of the model in actual application with the expected accuracy.

[0065] The model structure of the optimized migration model whose fitness coefficient is lower than the preset fitness coefficient is updated. This means that if the fitness coefficient of the optimized migration model is lower than a given standard, the structure of the optimized migration model needs to be updated, including introducing new data features, redesigning the model architecture, or increasing the complexity of the model, in order to improve the overall performance and adaptability of the model.

[0066] By adaptively optimizing the migration model whose switching result accuracy does not meet the standard, an optimized migration model is generated, and the adaptability of the optimized model is evaluated according to the prediction accuracy to obtain the adaptability coefficient. If the coefficient is lower than the preset standard, the model structure is further updated to ensure that the model maintains efficient prediction capabilities in the face of environmental changes, ultimately achieving a more accurate air conditioning effect.

[0067] In summary, the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method provided in this application has the following technical effects:

[0068] The temperature and humidity data, density data and external environment data of the target area are collected based on the cloud computing platform to generate update data; an air conditioning model is constructed according to the update records and air conditioning records; the air conditioning model is subjected to multi-region transfer learning to obtain a migration model; air conditioning prediction is performed on the migration model through the update data, and air dynamic control is performed according to the prediction results to generate a control strategy. In other words, by achieving the technical goals of cross-regional model sharing and rapid adaptation to the regional environment, the technical effect of improving air conditioning efficiency and reducing energy consumption is achieved.

[0069] Embodiment 2: Based on the direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing in the previous embodiment, the present application also provides a direct expansion constant temperature and humidity machine air conditioning control system based on cloud computing, please refer to the attached Figure 2 ,include:

[0070] An update data generation module 11, the update data generation module 11 is used to collect temperature and humidity data, density data and external environment data of the target area based on the cloud computing platform to generate update data; an air conditioning model construction module 12, the air conditioning model construction module 12 is used to construct an air conditioning model according to update records and air conditioning records; a migration model acquisition module 13, the migration model acquisition module 13 is used to perform multiple area migration learning on the air conditioning model to obtain a migration model; a control strategy generation module 14, the control strategy generation module 14 is used to perform air conditioning prediction on the migration model through the update data, perform air dynamic control according to the prediction results, and generate a control strategy.

[0071] Furthermore, the update data generation module 11 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0072] The temperature and humidity data, density data and external environment data are used as update features; a rolling window is configured with a fixed time length, and the rolling window is updated according to the fixed time length for the update features to generate the update data.

[0073] Furthermore, the migration model obtaining module 13 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0074] Based on the cloud computing platform, the temperature and humidity data characteristics, density data characteristics and external environment data characteristics of the target area are collected to obtain target characteristics; similar area groups are set for multiple areas according to the target characteristics to obtain similar area groups; the air conditioning model is shared among the similar area groups to obtain the migration model.

[0075] Furthermore, the migration model obtaining module 13 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0076] Taking the target feature as a standard, the feature similarity of the multiple regions is measured to obtain similar features; updating is performed according to the rolling windows of the multiple regions, and the updated results are fed back to the similar features to obtain abnormal feedback results; based on the abnormal feedback results, the group conditions of the similar features are updated, and the groups are set according to the updated group conditions to obtain the similar region groups.

[0077] Furthermore, the control strategy generation module 14 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0078] The density data is evaluated for switching strength to generate switching strength; a switching strategy is generated according to the switching strength and the prediction result; the operation mode is switched based on the switching strategy to generate a switching result; the migration model is short-term adjusted in response to the switching result and fed back to the migration model to generate a control strategy.

[0079] Furthermore, the control strategy generation module 14 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0080] An internal environment enthalpy value is evaluated by taking the temperature and humidity data and the density data as internal environment evaluation features to generate an internal environment enthalpy value; an external environment enthalpy value is evaluated by taking the external environment data as external environment evaluation features to generate an external environment enthalpy value; an enthalpy value difference between the internal environment enthalpy value and the external environment enthalpy value is obtained, and a preliminary switching strategy is generated based on the enthalpy value difference; the preliminary switching strategy is used as a switching guide and the switching strategy is used as a switching execution, a matching rate between the switching guide and the switching execution is evaluated, a switching result accuracy is generated, and the migration model is adjusted according to the switching result accuracy.

[0081] Furthermore, the control strategy generation module 14 in the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system is also used for:

[0082] Adaptively optimize the migration model whose switching result accuracy does not meet the error threshold to generate an optimized migration model; evaluate the adaptability of the optimized migration model according to the prediction accuracy to generate a adaptability coefficient; and update the model structure of the optimized migration model whose adaptability coefficient is lower than a preset adaptability coefficient.

[0083] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method and specific examples in the aforementioned embodiment one are also applicable to the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system in this embodiment. Through the aforementioned detailed description of the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method, those skilled in the art can clearly understand the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0084] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0085] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A direct expansion constant temperature and humidity machine air conditioning control method based on cloud computing, characterized in that: include: Based on the cloud computing platform, the temperature and humidity data, density data and external environment data of the target area are collected to generate updated data; constructing an air conditioning model based on the update records and the air conditioning records; Performing multiple area transfer learning on the air conditioning model to obtain a transfer model; The migration model is used to perform air conditioning prediction through the updated data, and air dynamic control is performed according to the prediction result to generate a control strategy.

2. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 1, characterized in that: The collecting of temperature and humidity data, density data and external environment data of the target area to generate update data includes: The temperature and humidity data, density data and external environment data are used as update features; The rolling window is configured with a fixed time length, and the rolling window is updated according to the fixed time length for the update feature to generate the update data.

3. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 1, characterized in that: The performing multiple region transfer learning on the air conditioning model to obtain a transfer model includes: Based on the cloud computing platform, the temperature and humidity data characteristics, density data characteristics and external environment data characteristics of the target area are collected to obtain target characteristics; Setting similar region groups for the multiple regions according to the target features to obtain similar region groups; The air conditioning model is shared among the similar area groups to obtain the migration model.

4. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 3, characterized in that: The step of setting similar region groups for the plurality of regions according to the target features to obtain similar region groups includes: Taking the target feature as a standard, measuring feature similarity of the multiple regions to obtain similar features; Performing updates according to the rolling windows of the multiple regions, and feeding back the updated results to the similar features to obtain abnormal feedback results; The group condition of the similar features is updated based on the abnormal feedback result, and the group is set according to the updated group condition to obtain the similar area group.

5. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 1, characterized in that: The method of performing air dynamic control according to the prediction results and generating a control strategy includes: Performing a switching intensity evaluation on the density data to generate a switching intensity; generating a switching strategy according to the switching intensity and the prediction result; Switching the operation mode based on the switching strategy and generating a switching result; In response to the switching result, the migration model is short-term adjusted and fed back to the migration model to generate a control strategy.

6. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 5, characterized in that: The short-term adjustment of the migration model in response to the switching result includes: Performing an internal environment enthalpy value assessment using the temperature and humidity data and the density data as internal environment assessment features to generate an internal environment enthalpy value; Using the external environment data as external environment assessment features to perform external environment enthalpy value assessment, and generate an external environment enthalpy value; Acquire an enthalpy difference between the internal environment enthalpy value and the external environment enthalpy value, and generate a preliminary switching strategy based on the enthalpy difference; The preliminary switching strategy is used as a switching guide, the switching strategy is used as a switching execution, a matching rate between the switching guide and the switching execution is evaluated, a switching result accuracy rate is generated, and the migration model is adjusted according to the switching result accuracy rate.

7. The cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to claim 6, characterized in that: Also includes: Adaptively optimizing the migration model whose handover result accuracy does not meet the error threshold to generate an optimized migration model; Evaluating the adaptability of the optimized migration model according to the prediction accuracy and generating an adaptability coefficient; The model structure of the optimized migration model whose adaptability coefficient is lower than the preset adaptability coefficient is updated.

8. A cloud computing-based direct expansion constant temperature and humidity machine air conditioning control system, characterized in that: The steps for implementing the cloud computing-based direct expansion constant temperature and humidity machine air conditioning control method according to any one of claims 1 to 7 include: An update data generation module, wherein the update data generation module is used to collect temperature and humidity data, density data and external environment data of the target area based on the cloud computing platform to generate update data; An air conditioning model building module, the air conditioning model building module is used to build an air conditioning model according to the update record and the air conditioning record; A migration model obtaining module, wherein the migration model obtaining module is used to perform multiple area migration learning on the air conditioning model to obtain a migration model; A control strategy generation module is used to perform air conditioning prediction on the migration model through the updated data, perform air dynamic control according to the prediction result, and generate a control strategy.

Images