Heating and ventilation control equipment modeling method and device, heating and ventilation control equipment and medium

Abstract

The invention relates to the technical field of intelligent household electrical appliances, and discloses a heating and ventilation control equipment modeling method, which comprises the following steps: pre-training all model parameters of a corresponding general model according to an equipment full-condition model data table, the equipment full-condition model data table being established according to a mechanism model or full-condition model data of heating and ventilation control equipment; according to a field initial equipment full-working-condition model data table, part of model parameters of the general model are updated, a global high-precision model is obtained, and the field initial equipment full-working-condition model data table is generated by updating field small sample initial data of the heating and ventilation control equipment into an equipment full-working-condition data model table; partial model parameters of the global high-precision model are updated according to the on-site real-time equipment full-working-condition model data table, a real-time global high-precision model is obtained, and the on-site real-time equipment full-working-condition model data table is obtained by updating on-site real-time operation data of the heating and ventilation control equipment into the on-site initial equipment full-working-condition model data table.

Description

Technical Field

[0001] The present application relates to the field of smart home appliance technology, for example, to a method and device for modeling HVAC control equipment, HVAC control equipment, and storage medium. Background Art

[0003] Related art discloses a HVAC control device, including a chiller, a cooling tower, a cooling pump, a refrigeration pump and other devices. Global optimization control of the HVAC control device requires the mapping relationship between the control feature and the target feature. Due to the complexity of the HVAC control device, the control feature and the target feature are the coupling of multiple device features. In order to make the control more accurate, it is very important to accurately model each device.

[0004] In the process of implementing the embodiments of the present disclosure, it is found that there are at least the following problems in the related art:

[0005] In the process of global optimization control modeling of HVAC control equipment, traditional mechanism modeling methods, machine learning black box modeling methods, or polynomial parameter identification semi-mechanism gray box models cannot take into account high generalization performance and high accuracy in the full range of working conditions. And with the long-term operation of HVAC control equipment, equipment aging, replacement, sensor offset, etc. will affect the accuracy of the model, and the model needs to be updated to keep the model high accuracy and high generalization ability.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical components or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0008] The embodiments of the present disclosure provide a method and apparatus for modeling HVAC control equipment, HVAC control equipment, and storage medium, so as to take into account high generalization performance and high precision in the entire operating range during the global optimization control modeling of the HVAC control equipment, reasonably update the model to maintain high precision and high generalization ability, reduce the amount of data during model initialization, and have rapid deployment capabilities.

[0009] In some embodiments, the method includes: pre-training all model parameters of the corresponding general model according to the equipment full-condition model data table, the equipment full-condition model data table is established according to the mechanism model or full-condition model data of the HVAC control equipment; according to the on-site initial equipment full-condition model data table, updating some model parameters of the general model to obtain a global high-precision model, the on-site initial equipment full-condition model data table is generated by updating the on-site small sample initial data of the HVAC control equipment to the equipment full-condition data model table; according to the on-site real-time equipment full-condition model data table, updating some model parameters of the global high-precision model to obtain a real-time global high-precision model, the on-site real-time equipment full-condition model data table is updated according to the on-site real-time operation data of the HVAC control equipment to the on-site initial equipment full-condition model data table.

[0010] In some embodiments, the device includes: a pre-training module, configured to pre-train all model parameters of the corresponding general model according to the equipment full-condition model data table, the equipment full-condition model data table is established according to the mechanism model or full-condition model data of the HVAC control equipment; a global high-precision model generation module, configured to update some model parameters of the general model according to the on-site initial equipment full-condition model data table to obtain a global high-precision model, the on-site initial equipment full-condition model data table is generated by updating the on-site small sample initial data of the HVAC control equipment to the equipment full-condition data model table; a real-time global high-precision model generation module, configured to update some model parameters of the global high-precision model according to the on-site real-time equipment full-condition model data table to obtain a real-time global high-precision model, the on-site real-time equipment full-condition model data table is updated according to the on-site real-time operation data of the HVAC control equipment to the on-site initial equipment full-condition model data table.

[0011] In some embodiments, the apparatus comprises a processor and a memory storing program instructions, and the processor is configured to execute the method for HVAC control equipment modeling when running the program instructions.

[0012] In some embodiments, the HVAC control device includes: a HVAC control device body; and the device for modeling the HVAC control device is installed on the HVAC control device body.

[0013] In some embodiments, the storage medium stores program instructions, and when the program instructions are run, the method for modeling HVAC control equipment is executed.

[0014] The method and device for HVAC control device modeling, HVAC control device, and storage medium provided in the embodiments of the present disclosure can achieve the following technical effects:

[0015] According to the full-condition model data table of the equipment established based on the mechanism model or full-condition model data of the HVAC control equipment, all model parameters of the corresponding general model are pre-trained, which can retain the correlation law between the full-condition model data, pave the way for the generalization of the model, and make it easier to judge the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained from the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. Under the premise of retaining the correlation law between the full-condition model data, the model can perform with high accuracy in the actual field, which can improve the generalization and accuracy of the model. In addition, since the amount of data required for model initialization is small, the model has the ability to be deployed quickly. According to the on-site real-time equipment full-condition model data table updated based on the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model, which is conducive to coping with equipment environment changes and performance degradation. Real-time updating of the model allows the model to maintain high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole operating range, the model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, so as to have the ability of rapid deployment.

[0016] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] One or more embodiments are exemplarily described by corresponding drawings, which do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements, and the drawings do not constitute a scale limitation, and wherein:

[0018] Figure 1 is a schematic diagram of a method for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0019] Figure 2 is a schematic diagram of another method for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0020] Figure 3 is a schematic diagram of another method for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0021] Figure 4 is a schematic diagram of another method for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0022] Figure 5 is a schematic diagram of a device for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0023] Figure 6 is a schematic diagram of another device for modeling HVAC control equipment provided by an embodiment of the present disclosure;

[0024] Figure 7 It is a schematic diagram of a HVAC control device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In order to be able to understand the features and technical contents of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The attached drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.

[0026] The terms "first", "second", etc. in the specification and claims of the embodiments of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so that the embodiments of the embodiments of the present disclosure described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.

[0027] Unless otherwise stated, the term "plurality" means two or more.

[0028] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B indicates: A or B.

[0029] The term "and / or" is a description of the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0030] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

[0031] Combination Figure 1 As shown, the embodiment of the present disclosure provides a method for modeling a HVAC control device, comprising:

[0032] S101, the HVAC control device pre-trains all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, where the equipment full-operating-condition model data table is established based on the mechanism model or full-operating-condition model data of the HVAC control device.

[0033] S102, the HVAC control device updates some model parameters of the general model according to the on-site initial equipment full-operating condition model data table to obtain a global high-precision model. The on-site initial equipment full-operating condition model data table is generated by updating the on-site small sample initial data of the HVAC control device to the equipment full-operating condition data model table.

[0034] S103, the HVAC control device updates some model parameters of the global high-precision model according to the on-site real-time equipment full-operating condition model data table to obtain a real-time global high-precision model. The on-site real-time equipment full-operating condition model data table is obtained by updating the on-site initial equipment full-operating condition model data table according to the on-site real-time operation data of the HVAC control device.

[0035] According to the method for modeling HVAC control equipment provided by the embodiment of the present disclosure, all model parameters of the corresponding general model are pre-trained according to the equipment full-condition model data table established according to the mechanism model of the HVAC control equipment or the full-condition model data, which can retain the correlation law between the full-condition model data, pave the way for the generalization of the model, and more conveniently determine the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained by the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. Under the premise of retaining the correlation law between the full-condition model data, the model is made to have high performance accuracy in the actual field, which can improve the generalization and accuracy of the model, and because the amount of data required for model initialization is small, the model has the ability to be quickly deployed. According to the on-site real-time equipment full-condition model data table updated by the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model, which is conducive to coping with equipment environment changes and performance degradation, and real-time updating of the model allows the model to maintain high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole operating range, the model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, so as to have the ability of rapid deployment.

[0036] Optionally, the HVAC control device establishes a data table of the full-operating-condition model of the equipment according to the mechanism model or full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a data table of the full-operating-condition model of the equipment according to the full-operating-condition model data of the HVAC control device. Alternatively, the HVAC control device establishes a data table of the full-operating-condition model of the equipment according to the mechanism model or full-operating-condition model data of similar equipment. Among them, the similarity of the equipment parameters of the similar equipment and the HVAC control device is within a similar interval. In this way, establishing a data table of the full-operating-condition model of the equipment according to the HVAC control device and similar equipment whose similarity to its equipment parameters is within a similar interval is conducive to expanding the sample data of the data table of the full-operating-condition model of the equipment, and can retain the correlation law between the full-operating-condition model data, pave the way for the generalization of the model, and make it more convenient to judge the scope of application of the model.

[0037] Optionally, the equipment parameters include: when the type of similar equipment is a chiller, the equipment parameters are the brand model and nominal cooling capacity of the unit. Alternatively, when the type of similar equipment is a cooling tower, the equipment parameters are the rated water flow rate and the rated air volume. When the type of similar equipment is a water pump, the equipment parameters are the rated flow rate and the rated head. In this way, it is helpful to better determine the chillers, cooling towers, and water pumps whose similarity to the equipment parameters of the HVAC control equipment is within a similar range. This is conducive to expanding the sample data of the equipment full-condition model data table, and the correlation law between the full-condition model data can be retained, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0038] Optionally, each device full-condition model data table of the HVAC control device can store n condition ranges as input feature data, and each condition range can store k output feature data. Among them, n is the number of condition ranges, and k is the number of output feature data that can be stored in each condition range. In this way, each device full-condition model data table can store n×k groups of input and output data, which is conducive to expanding the sample data of the device full-condition model data table, and can retain the correlation law between the full-condition model data, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0039] Optionally, the full-operating model data of the HVAC control equipment includes part or all of the following: chiller model data, cooling tower water supply model data, cooling tower fan power model data, cooling pump power model data, cooling pump flow model data, freezing pump power model data, and freezing pump flow model data. This is conducive to expanding the data volume of the full-operating model data of the HVAC control equipment, thereby facilitating the expansion of the sample data of the full-operating model data table of the equipment, and can retain the correlation law between the full-operating model data, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0040] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a chiller model data table according to the chiller model data. Specifically, the chiller model data includes the chilled water supply temperature, the cooling water supply temperature, the chiller load rate and the chiller power. The HVAC control device establishes a chiller model data table according to the chiller model data, including: the HVAC control device establishes a chiller model data table, so that the input characteristic data of the chiller model data table includes n operating ranges consisting of the chilled water supply temperature, the cooling water supply temperature and the chiller load rate, and each operating range can store k output characteristic data, and each output characteristic data includes the chiller power. More specifically, the value range of the chilled water supply temperature is [5°C, 18°C], with a temperature data point every 0.5°C. The value range of the cooling water supply temperature is [20°C, 35°C], with a temperature data point every 0.5°C. The range of the chiller load rate is [10%, 100%], with a load rate data point every 5%. In this way, the chiller model data table can store n×k groups of input and output data consisting of chilled water supply temperature, cooling water supply temperature, chiller load rate and chiller power, which is conducive to expanding the data volume of the chiller model data table, thereby facilitating the expansion of the sample data of the equipment full-condition model data table, and can retain the correlation law between the full-condition model data, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0041] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a cooling tower water supply model data table according to the cooling tower water supply model data. Specifically, the cooling tower water supply model data includes: part or all of the outdoor wet-bulb temperature, cooling water return temperature, cooling water flow, cooling tower fan frequency and cooling water supply temperature. The HVAC control device establishes a cooling tower water supply model data table according to the cooling tower water supply model data, including: the HVAC control device establishes a cooling tower water supply model data table, the input characteristic data of the cooling tower water supply model data table includes n operating ranges composed of outdoor wet-bulb temperature, cooling water return temperature, cooling water flow and cooling tower fan frequency, and each operating range can store k output characteristic data, each output characteristic data includes cooling water supply temperature. More specifically, the value range of the outdoor wet-bulb temperature is [18°C, 35°C], with a temperature data point every 0.5°C. The range of cooling water return temperature is [25℃, 42℃], with a temperature data point every 0.5℃. The range of cooling water flow is [30%, 100℃], with a flow data point every 5%. In this way, the cooling tower water supply model data table can store n×k groups of input and output data consisting of outdoor wet bulb temperature, cooling water return temperature, cooling water flow, cooling tower fan frequency and cooling water supply temperature, which is conducive to expanding the data volume of the cooling tower water supply model data table, thereby helping to expand the sample data of the equipment full-condition model data table, and can retain the correlation law between the full-condition model data, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0042] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a cooling tower fan power model data table according to the cooling tower fan power model data. Specifically, the cooling tower water supply model data includes: cooling tower frequency and cooling tower fan power. The HVAC control device establishes a cooling tower fan power model data table according to the cooling tower fan power model data, including: the HVAC control device establishes a cooling tower fan power model data table, the input characteristic data of the cooling tower fan power model data table includes n operating ranges composed of cooling tower frequencies, and each operating range can store k output characteristic data, and each output characteristic data includes cooling tower fan power. More specifically, the value range of the cooling tower frequency is [20Hz, 50Hz], with a frequency data point every 2Hz. In this way, the cooling tower fan power model data table can store n×k groups of input and output data consisting of cooling tower frequency and cooling tower fan power, which is conducive to expanding the data volume of the cooling tower fan power model data table, thereby facilitating the expansion of the sample data of the equipment full-condition model data table, and can retain the correlation law between the full-condition model data, thus paving the way for the generalization of the model and making it more convenient to judge the scope of application of the model.

[0043] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a cooling pump power model data table according to the cooling pump power model data. Specifically, the cooling pump power model data includes cooling pump frequency and cooling pump power. The HVAC control device establishes a cooling pump power model data table according to the cooling pump power model data, including: the HVAC control device establishes a cooling pump power model data table, the input characteristic data of the cooling pump power model data table includes n operating ranges composed of cooling pump frequencies, and each operating range can store k output characteristic data, and each output characteristic data includes cooling pump power. More specifically, the value range of the cooling pump frequency is [20Hz, 50Hz], with a frequency data point every 2Hz. In this way, the cooling pump power model data table can store n×k groups of input and output data consisting of cooling pump frequency and cooling pump power, which is conducive to expanding the data volume of the cooling pump power model data table, thereby facilitating the expansion of the sample data of the equipment full-condition model data table, and can retain the correlation law between the full-condition model data, thus paving the way for the generalization of the model and making it more convenient to judge the scope of application of the model.

[0044] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a cooling pump flow model data table according to the cooling pump flow model data. Specifically, the cooling pump flow model data includes the cooling pump frequency and the cooling water flow. The HVAC control device establishes a cooling pump flow model data table according to the cooling pump flow model data, including: the HVAC control device establishes a cooling pump flow model data table, the input characteristic data of the cooling pump flow model data table includes n operating ranges composed of cooling pump frequencies, and each operating range can store k output characteristic data, and each output characteristic data includes cooling water flow. More specifically, the value range of the cooling pump frequency is [20Hz, 50Hz], with a frequency data point every 2Hz. In this way, the cooling pump flow model data table can store n×k groups of input and output data consisting of cooling pump frequency and cooling water flow, which is conducive to expanding the data volume of the cooling pump flow model data table, thereby helping to expand the sample data of the equipment full-condition model data table, and can retain the correlation rules between the full-condition model data, paving the way for the generalization of the model and making it more convenient to judge the scope of application of the model.

[0045] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a refrigeration pump power model data table according to the refrigeration pump power model data. Specifically, the refrigeration pump power model data includes the refrigeration pump frequency and the refrigeration pump power. The HVAC control device establishes a refrigeration pump power model data table according to the refrigeration pump power model data, including: the HVAC control device establishes a refrigeration pump power model data table, the input characteristic data of the refrigeration pump power model data table includes n operating ranges composed of refrigeration pump frequencies, and each operating range can store k output characteristic data, and each output characteristic data includes refrigeration pump power. More specifically, optionally, the value range of the refrigeration pump frequency is [20Hz, 50Hz], with a frequency data point every 2Hz. In this way, the refrigeration pump power model data table can store n×k groups of input and output data consisting of refrigeration pump frequency and refrigeration pump power, which is conducive to expanding the data volume of the refrigeration pump power model data table, thereby facilitating the expansion of the sample data of the equipment full-condition model data table, and can retain the correlation law between the full-condition model data, pave the way for the generalization of the model, and make it more convenient to judge the scope of application of the model.

[0046] Optionally, the HVAC control device establishes a full-operating-condition model data table of the device according to the full-operating-condition model data of the HVAC control device, including: the HVAC control device establishes a freezing pump flow model data table according to the freezing pump flow model data. Specifically, the freezing pump flow model data includes freezing pump frequency and freezing water flow. The HVAC control device establishes a freezing pump flow model data table according to the freezing pump flow model data, including: the HVAC control device establishes a freezing pump flow model data table, the input characteristic data of the freezing pump flow model data table includes n operating ranges composed of freezing pump frequency, and each operating range can store k output characteristic data, and each output characteristic data includes freezing water flow. In this way, the freezing pump flow model data table can store n×k groups of input and output data composed of freezing pump frequency and freezing water flow, which is conducive to expanding the data volume of the freezing pump flow model data table, thereby facilitating the expansion of the sample data of the full-operating-condition model data table of the device, and can retain the correlation law between the full-operating-condition model data, paving the way for the generalization of the model, and making it more convenient to judge the scope of application of the model.

[0047] Optionally, the mechanism model includes a proportional similarity law. The HVAC control device establishes a full-condition model data table for the device according to the mechanism model or full-condition model data of similar devices, including: the HVAC control device generates a water pump power model data table, a flow model data table, and a fan power model data table according to the proportional similarity law. Among them, the water pump power model data table includes a cooling pump power model data table and / or a freezing pump power model data table. The flow model includes a cooling pump flow model data table and / or a freezing pump flow model data table. Alternatively, the HVAC control device fills the full-condition model data into the condition grid of the full-condition model data table of the device to which the full-condition model data belongs. Optionally, the mechanism model also includes refrigeration cycle calculation, or ε-NTU (efficiency-heat transfer unit number method) heat exchange calculation. In this way, it is conducive to better establish a full-condition model data table for the device according to the mechanism model of similar devices. Thereby, it is conducive to expanding the sample data of the full-condition model data table of the equipment, and the correlation law between the full-condition model data can be retained, which paves the way for the generalization of the model and makes it more convenient to judge the scope of application of the model.

[0048] Optionally, the HVAC control device generates a water pump power model data table, a flow model data table, and a fan power model data table according to the proportional similarity law, including: a =P d ×(F a / F d ) 3 .Q a =Q d ×(F a / F d ). Among them, F a is the actual frequency of the pump or fan, F d is the rated frequency of the pump or fan; P a The operating frequency of the pump or fan is F a The actual power corresponding to the time. d The operating frequency of the pump or fan is F d The corresponding rated power. a The operating frequency of the water pump is F a The actual flow rate corresponding to the pump is F d The water pump includes a cooling pump or a refrigeration pump, and the fan includes a cooling tower fan.

[0049] In this way, it is helpful to better establish the equipment full-condition model data table according to the mechanism model or full-condition model data of similar equipment. This is conducive to expanding the sample data of the equipment full-condition model data table, training the general model according to the equipment full-condition model data table, and improving the accuracy and generalization performance of the model. In this way, it is conducive to more accurately establish the equipment full-condition model data table according to the mechanism model of similar equipment. This is conducive to expanding the sample data of the equipment full-condition model data table, and the correlation law between the full-condition model data can be retained, which paves the way for the generalization of the model and makes it easier to judge the scope of application of the model.

[0050] Optionally, the general model includes CNN (Convolutional Neural Network) and / or MLP (Multilayer Perceptron). The HVAC control device pre-trains all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, including: when the type of the equipment full-operating-condition model data table is a chiller model data table or a cooling tower water supply model data table, the HVAC control device uses the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output to pre-train all model parameters of CNN and MLP. When the type of the equipment full-operating-condition model data table is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a freezing pump power model data table or a freezing pump flow model data table, the HVAC control device uses the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output to pre-train all model parameters of MLP. In this way, when the type of the equipment full-operating model data table is a chiller model data table or a cooling tower water supply model data table, the total amount of data is large, and all model parameters of CNN and MLP are pre-trained. When the type of the equipment full-operating model data table is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a freezing pump power model data table, or a freezing pump flow model data table, the total amount of data is small, and only all model parameters of MLP are pre-trained. This is conducive to retaining the correlation law between the full-operating model data, paving the way for the generalization of the model, and making it easier to judge the scope of application of the model.

[0051] Optionally, the HVAC control device updates some model parameters of the general model according to the full-operating model data table of the initial equipment on site to obtain a global high-precision model, including: when the type of the global high-precision model is CNN and MLP, the HVAC control device inputs the actual input feature data of the full-operating model data table of the initial equipment on site into the general model to obtain a fuzzy output value. The HVAC control device compares and calculates the fuzzy output value with the actual output feature data of the full-operating model data table of the initial equipment on site to obtain a loss function. The HVAC control device updates the model parameters of the MLP layer using the gradient descent back propagation of the loss function, keeps the model parameters of the CNN layer unchanged, and obtains a global high-precision model. In this way, when the type of the full-operating model data table of the equipment is a chiller model data table or a cooling tower water supply model data table, after pre-training all model parameters of CNN and MLP, the model parameters of the MLP layer are updated by the full-operating model data table of the initial equipment on site updated with the initial data of the small sample on site, and the model parameters of the CNN layer are kept unchanged, and the general model is updated to obtain a global high-precision model. The fixed CNN layer parameters preserve the correlation between the full-condition model data of the HVAC control equipment. Under the premise of preserving the correlation between the full-condition model data, the model can be highly accurate in actual field performance, which can improve the generalization and accuracy of the model. In addition, since the amount of data required for model initialization is small, the model has the ability to be deployed quickly. This is conducive to improving the accuracy and generalization performance of the model.

[0052] Optionally, the HVAC control device updates some model parameters of the general model according to the full-operating model data table of the initial equipment on site to obtain a global high-precision model, including: when the type of the global high-precision model is MLP, the HVAC control device inputs the actual input feature data of the full-operating model data table of the initial equipment on site into the general model to obtain a fuzzy output value. The HVAC control device compares and calculates the fuzzy output value with the actual output feature data of the full-operating model data table of the initial equipment on site to obtain a loss function. The HVAC control device updates the model parameters of the last layer of the MLP using the gradient descent back propagation of the loss function, keeps the model parameters of other layers of the MLP unchanged, and obtains a global high-precision model. In this way, when the type of the full-operating model data table of the equipment is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a freezing pump power model data table, or a freezing pump flow model data table, after pre-training all the model parameters of the MLP, only the model parameters of the last layer of the MLP are updated, the model parameters of other layers of the MLP are kept unchanged, and the general model is updated to obtain a global high-precision model. The fixed MLP layer parameters except the last layer preserve the correlation law between the full-condition model data of the HVAC control equipment. Under the premise of retaining the correlation law between the full-condition model data, the model can perform with high accuracy in the actual field, which can improve the generalization and accuracy of the model. In addition, since the amount of data required for model initialization is small, the model has the ability to be deployed quickly. This is conducive to improving the accuracy and generalization performance of the model.

[0053] Optionally, the HVAC control device updates the full-condition model data table of the on-site initial equipment according to the on-site real-time operation data of the HVAC control device to obtain the full-condition model data table of the on-site real-time equipment, including: when the HVAC control device is running and reaches the model update time, the HVAC control device determines the number of on-site real-time operation data. The HVAC control device updates the full-condition model data table of the on-site initial equipment according to the number of on-site real-time operation data to obtain the full-condition model data table of the on-site real-time equipment. In this way, when the system runs for a period of time and reaches the model update time, according to the on-site real-time equipment full-condition model data table updated with the on-site real-time operation data of the HVAC control device, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. It is conducive to coping with equipment environment changes and performance attenuation, and the real-time update of the model enables the model to maintain high precision and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high precision are taken into account in the full range of working conditions, and the model is reasonably updated to maintain high precision and high generalization ability.

[0054] Optionally, the HVAC control device updates the full-condition model data table of the on-site initial equipment according to the number of on-site real-time operation data to obtain the full-condition model data table of the on-site real-time equipment, including: when the on-site real-time operation data is less than k, the HVAC control device updates the on-site real-time operation data to the full-condition model data table of the on-site initial equipment to obtain the full-condition model data table of the on-site real-time equipment. When the number of on-site real-time operation data is greater than or equal to k, the HVAC control device updates the screened on-site real-time operation data to the full-condition model data table of the on-site initial equipment to obtain the full-condition model data table of the on-site real-time equipment. Among them, the screened on-site real-time operation data is obtained based on the predicted value of the global high-precision model of the on-site real-time operation data and the deviation of the on-site real-time operation data. In this way, if the number of on-site real-time operation data is less than k, all the on-site real-time operation data are updated to the full-condition model data table of the on-site initial equipment. If the number of on-site real-time operation data is greater than or equal to k, the screened on-site real-time operation data is first determined based on the predicted value of the global high-precision model of the on-site real-time operation data and the deviation of the on-site real-time operation data, and then the screened on-site real-time operation data is updated to the full-condition model data table of the on-site initial equipment. When the model is updated, only the data in the working condition range that has changed is used to update the model. The model also uses fine-tuning to update the model while adapting to the attenuation changes of the system, while maintaining the high accuracy and generalization of the model. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole working condition range, and the model is reasonably updated to maintain high accuracy and generalization ability. It is conducive to coping with changes in equipment environment and performance attenuation, and updating the model in real time allows the model to maintain high accuracy and generalization performance.

[0055] Optionally, the HVAC control device obtains the screened on-site real-time operation data according to the predicted value of the global high-precision model of the on-site real-time operation data and the deviation of the on-site real-time operation data, including: the HVAC control device determines the predicted value of the global high-precision model of the data point of the on-site real-time operation data and the absolute value of the deviation of the data point of the on-site real-time operation data. The HVAC control device removes the data point with the largest absolute value of the deviation in the on-site real-time operation data to obtain the screened on-site real-time operation data. In this way, if the number of on-site real-time operation data is greater than or equal to k, it is necessary to first remove the data that performs poorly in the global high-precision model, that is, the data point with the largest absolute value of the deviation between the predicted value of the global high-precision model of the data point of the on-site real-time operation data and the data point of the on-site real-time operation data, and then add the new data to the full-condition model data table of the on-site initial equipment to obtain the full-condition model data table of the on-site real-time equipment. When the model is updated, only the data in the working condition range where the data has changed within the working condition range is used to update the model. The model also adopts a fine-tuning method so that the model is updated in adapting to the attenuation changes of the system while maintaining the high precision and high generalization of the model. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole working range, and the model is reasonably updated to maintain high accuracy and high generalization ability. It is conducive to coping with equipment environment changes and performance degradation, and the model is updated in real time to maintain high accuracy and high generalization performance.

[0056] Optionally, the HVAC control device updates some model parameters of the global high-precision model according to the full-operating-condition model data table of the on-site real-time equipment to obtain a real-time global high-precision model, including: when the type of the global high-precision model is CNN and MLP, the HVAC control device inputs the actual input feature data of the full-operating-condition model data table of the on-site real-time equipment into the global high-precision model to obtain a fuzzy output value. The HVAC control device compares and calculates the fuzzy output value with the actual output feature data of the full-operating-condition model data table of the on-site real-time equipment to obtain a loss function. The HVAC control device updates the model parameters of the MLP layer by back-propagation of the gradient descent of the loss function, keeps the model parameters of the CNN layer unchanged, and obtains a real-time global high-precision model. When the type of the global high-precision model is MLP, the HVAC control device inputs the actual input feature data of the full-operating-condition model data table of the on-site real-time equipment into the global high-precision model to obtain a fuzzy output value. The HVAC control device compares and calculates the fuzzy output value with the actual output feature data of the full-operating-condition model data table of the on-site real-time equipment to obtain a loss function. The HVAC control equipment uses the gradient descent back propagation of the loss function to update the model parameters of the last layer of the MLP, keeping the model parameters of other layers of the MLP unchanged, and obtaining a real-time global high-precision model. In this way, after updating the full-condition model data table of the initial equipment on site according to the number of real-time operation data on site and obtaining the full-condition model data table of the real-time equipment on site, some model parameters of the corresponding global high-precision model are updated according to the full-condition model data table of the real-time equipment on site. The fixed model parameters of the CNN layer or the model parameters of the MLP layer except the last layer are conducive to coping with equipment environment changes and performance degradation, and the model is updated in real time so that the model maintains high precision and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high precision are taken into account in the full range of working conditions, the model is reasonably updated to maintain high precision and high generalization ability, and the amount of data during model initialization is reduced, with rapid deployment capabilities.

[0057] Combination Figure 2 As shown, the embodiment of the present disclosure provides another method for modeling a HVAC control device, comprising:

[0058] S201, the HVAC control device determines the type of the equipment full-operating-condition model data table.

[0059] S202, when the type of the equipment full-operating-condition model data table is a chiller model data table or a cooling tower water supply model data table, the HVAC control device uses the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output to pre-train all model parameters of CNN and MLP.

[0060] S203, when the type of the equipment full-operating-condition model data table is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a refrigeration pump power model data table or a refrigeration pump flow model data table, the HVAC control device uses the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output to pre-train all model parameters of the MLP.

[0061] S204, the HVAC control device updates some model parameters of the general model according to the on-site initial equipment full-operating condition model data table to obtain a global high-precision model. The on-site initial equipment full-operating condition model data table is generated by updating the on-site small sample initial data of the HVAC control device to the equipment full-operating condition data model table.

[0062] S205, the HVAC control device updates some model parameters of the global high-precision model according to the on-site real-time equipment full-operating condition model data table to obtain a real-time global high-precision model. The on-site real-time equipment full-operating condition model data table is obtained by updating the on-site initial equipment full-operating condition model data table according to the on-site real-time operation data of the HVAC control device.

[0063] The equipment full-operating-condition model data table is established based on the mechanism model or full-operating-condition model data of the HVAC control equipment. The general model includes CNN and / or MLP.

[0064] Using the method for modeling HVAC control equipment provided by the embodiment of the present disclosure, all model parameters of the corresponding general model are pre-trained according to the full-operating-condition model data table of the equipment established based on the mechanism model of the HVAC control equipment or the full-operating-condition model data. When the type of the full-operating-condition model data table of the equipment is a chiller model data table or a cooling tower water supply model data table, the total amount of data is large at this time, and all model parameters of CNN and MLP are pre-trained. When the type of the full-operating-condition model data table of the equipment is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a freezing pump power model data table, or a freezing pump flow model data table, the total amount of data is small at this time, and only all model parameters of MLP are pre-trained. The correlation law between the full-operating-condition model data can be retained, paving the way for the generalization of the model, and it is more convenient to judge the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained from the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. Under the premise of retaining the correlation law between the full-condition model data, the model can be highly accurate in the actual field, which can improve the generalization and accuracy of the model. In addition, since the amount of data required for model initialization is small, the model has the ability to be deployed quickly. According to the on-site real-time equipment full-condition model data table obtained from the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model, which is conducive to coping with equipment environment changes and performance degradation. The model is updated in real time so that the model maintains high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the full range of operating conditions. The model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, which has the ability to be deployed quickly.

[0065] Combination Figure 3 As shown, the embodiment of the present disclosure provides another method for modeling a HVAC control device, comprising:

[0066] S301, the HVAC control device pre-trains all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, where the equipment full-operating-condition model data table is established based on the mechanism model or full-operating-condition model data of the HVAC control device.

[0067] S302, when the type of the global high-precision model is CNN and MLP, the HVAC control device inputs the actual input feature data of the full-condition model data table of the on-site initial equipment into the general model to obtain a fuzzy output value.

[0068] S303, the HVAC control device compares and calculates the fuzzy output value with the actual output characteristic data of the full-operating-condition model data table of the on-site initial equipment to obtain a loss function.

[0069] S304, the HVAC control device updates the model parameters of the MLP layer using the gradient descent back propagation of the loss function, keeps the model parameters of the CNN layer unchanged, and obtains a global high-precision model.

[0070] S305, when the type of the global high-precision model is MLP, the HVAC control device inputs the actual input feature data of the full-condition model data table of the on-site initial equipment into the general model to obtain a fuzzy output value.

[0071] S306, the HVAC control device compares and calculates the fuzzy output value with the actual output characteristic data of the full-operating condition model data table of the on-site initial equipment to obtain a loss function.

[0072] S307, the HVAC control device updates the model parameters of the last layer of the MLP using the gradient descent back propagation of the loss function, keeping the model parameters of other layers of the MLP unchanged, and obtaining a global high-precision model.

[0073] S308, the HVAC control device updates some model parameters of the global high-precision model according to the on-site real-time equipment full-operating condition model data table to obtain a real-time global high-precision model. The on-site real-time equipment full-operating condition model data table is obtained by updating the on-site initial equipment full-operating condition model data table according to the on-site real-time operation data of the HVAC control device.

[0074] Among them, common models include CNN and / or MLP.

[0075] By adopting the method for modeling HVAC control equipment provided by the embodiment of the present disclosure, all model parameters of the corresponding general model are pre-trained according to the equipment full-condition model data table established based on the mechanism model of the HVAC control equipment or the full-condition model data, which can retain the correlation law between the full-condition model data, pave the way for the generalization of the model, and make it easier to judge the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained from the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. In the case where the type of the equipment full-condition model data table is a chiller model data table or a cooling tower water supply model data table, after pre-training all model parameters of CNN and MLP, the model parameters of the MLP layer are updated by the on-site initial equipment full-condition model data table updated with the on-site small sample initial data, and the model parameters of the CNN layer are kept unchanged, and the general model is updated to obtain a real-time global high-precision model. When the type of the equipment full-condition model data table is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a freezing pump power model data table or a freezing pump flow model data table, after pre-training all the model parameters of the MLP, only the model parameters of the last layer of the MLP are updated, and the model parameters of other layers of the MLP are kept unchanged, and the general model is updated to obtain a real-time global high-precision model. The fixed CNN layer parameters or the MLP layer parameters except the last layer enable the correlation law between the full-condition model data of the HVAC control equipment to be preserved. Under the premise of retaining the correlation law between the full-condition model data, the model can be made to have high accuracy in the actual field, which can improve the generalization and accuracy of the model, and because the amount of data required for model initialization is small, the model has the ability to be deployed quickly. According to the field real-time equipment full-condition model data table updated according to the field real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model, which is conducive to coping with equipment environment changes and performance degradation. The real-time update of the model enables the model to maintain high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole working range, the model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, so as to have the ability of rapid deployment. Figure 4 As shown, the embodiment of the present disclosure provides another method for modeling a HVAC control device, comprising:

[0076] S401, the HVAC control device pre-trains all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, where the equipment full-operating-condition model data table is established based on the mechanism model or full-operating-condition model data of the HVAC control device.

[0077] S402, the HVAC control device updates some model parameters of the general model according to the on-site initial equipment full-operating condition model data table to obtain a global high-precision model. The on-site initial equipment full-operating condition model data table is generated by updating the on-site small sample initial data of the HVAC control device to the equipment full-operating condition data model table.

[0078] S403: When the HVAC control device reaches the model update time after operation, the HVAC control device determines the amount of on-site real-time operation data.

[0079] S404, the HVAC control device updates the on-site initial equipment full-condition model data table according to the amount of on-site real-time operation data to obtain the on-site real-time equipment full-condition model data table.

[0080] S405, the HVAC control device updates some model parameters of the general model according to the full-operating condition model data table of the initial equipment on site to obtain a global high-precision model.

[0081] According to the method for modeling HVAC control equipment provided by the embodiment of the present disclosure, all model parameters of the corresponding general model are pre-trained according to the full-condition model data table of the equipment established according to the mechanism model of the HVAC control equipment or the full-condition model data, so that the correlation law between the full-condition model data can be retained, which paves the way for the generalization of the model and makes it easier to judge the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained by the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. Under the premise of retaining the correlation law between the full-condition model data, the model is made to have high performance accuracy in the actual field, which can improve the generalization and accuracy of the model, and because the amount of data required for model initialization is small, the model has the ability to be deployed quickly. When the system runs for a period of time and reaches the model update time, according to the on-site real-time equipment full-condition model data table updated by the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. It is beneficial to cope with changes in equipment environment and performance degradation, and updating the model in real time allows the model to maintain high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole operating range, the model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, so as to have the ability of rapid deployment.

[0082] Combination Figure 5As shown, the embodiment of the present disclosure provides a device 200 for modeling HVAC control equipment, including a pre-training module 501, a global high-precision model generation module 502 and a real-time global high-precision model generation module 503. The pre-training module 501 is configured to pre-train all model parameters of the corresponding general model according to the equipment full-condition model data table, and the equipment full-condition model data table is established according to the mechanism model or full-condition model data of the HVAC control equipment. The global high-precision model generation module 502 is configured to update some model parameters of the general model according to the on-site initial equipment full-condition model data table to obtain a global high-precision model. The on-site initial equipment full-condition model data table is generated by updating the on-site small sample initial data of the HVAC control equipment to the equipment full-condition data model table. The real-time global high-precision model generation module 503 is configured to update some model parameters of the global high-precision model according to the on-site real-time equipment full-condition model data table to obtain a real-time global high-precision model. The on-site real-time equipment full-condition model data table is obtained by updating the on-site real-time operation data of the HVAC control equipment to the on-site initial equipment full-condition model data table.

[0083] The device for modeling HVAC control equipment provided by the embodiment of the present disclosure is used. According to the full-condition model data table of the equipment established based on the mechanism model of the HVAC control equipment or the full-condition model data, all model parameters of the corresponding general model are pre-trained, which can retain the correlation law between the full-condition model data, pave the way for the generalization of the model, and make it more convenient to judge the scope of application of the model. According to the on-site real-time equipment full-condition model data table obtained from the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. Under the premise of retaining the correlation law between the full-condition model data, the model is made to have high performance accuracy in the actual field, which can improve the generalization and accuracy of the model, and because the amount of data required for model initialization is small, the model has the ability to be quickly deployed. According to the on-site real-time equipment full-condition model data table updated based on the on-site real-time operation data of the HVAC control equipment, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model, which is conducive to coping with equipment environment changes and performance degradation, and real-time updating of the model allows the model to maintain high accuracy and high generalization performance. In this way, in the process of global optimization control modeling of HVAC control equipment, good generalization performance and high accuracy are taken into account in the whole operating range, the model is reasonably updated to maintain high accuracy and high generalization ability, and the amount of data during model initialization is reduced, so as to have the ability of rapid deployment.

[0084] Combination Figure 6As shown, the embodiment of the present disclosure provides another device 300 for modeling HVAC control equipment, including a processor (processor) 600 and a memory (memory) 601. Optionally, the device may also include a communication interface (Communication Interface) 602 and a bus 603. Among them, the processor 600, the communication interface 602, and the memory 601 can communicate with each other through the bus 603. The communication interface 602 can be used for information transmission. The processor 600 can call the logic instructions in the memory 601 to execute the method for modeling HVAC control equipment in the above embodiment.

[0085] In addition, the logic instructions in the memory 601 described above can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0086] The memory 601 is a computer-readable storage medium that can be used to store software programs and computer executable programs, such as program instructions / modules corresponding to the method in the embodiment of the present disclosure. The processor 600 executes the function application and data processing by running the program instructions / modules stored in the memory 601, that is, the method for modeling HVAC control equipment in the above embodiment is implemented.

[0087] The memory 601 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 601 may include a high-speed random access memory and may also include a non-volatile memory.

[0088] Combination Figure 7 As shown, an embodiment of the present disclosure provides a HVAC control device 100, including: a HVAC control device body, and the above-mentioned device 200 (300) for modeling a HVAC control device. The device 200 (300) for modeling a HVAC control device is installed in the HVAC control device body. The installation relationship described here is not limited to placement inside the HVAC control device, but also includes installation connections with other components of the HVAC control device, including but not limited to physical connections, electrical connections, or signal transmission connections. It can be understood by those skilled in the art that the device 200 (300) for modeling a HVAC control device can be adapted to a feasible HVAC control device body, thereby realizing other feasible embodiments.

[0089] An embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned method for modeling HVAC control equipment.

[0090] The computer-readable storage medium mentioned above may be a transient computer-readable storage medium or a non-transitory computer-readable storage medium.

[0091] The technical solution of the embodiment of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiment of the present disclosure. The aforementioned storage medium may be a non-transient storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes, or a transient storage medium.

[0092] The above description and the accompanying drawings fully illustrate the embodiments of the present disclosure so that those skilled in the art can practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible changes. Unless explicitly required, separate components and functions are optional, and the order of operation may vary. The parts and features of some embodiments may be included in or replace the parts and features of other embodiments. Moreover, the words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and the claims, unless the context clearly indicates, the singular forms of "a", "an" and "the" are intended to include plural forms as well. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of listings containing one or more associated ones. In addition, when used in the present application, the term "comprise" and its variants "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof. In the absence of further restrictions, the elements defined by the sentence "comprising a ..." do not exclude the presence of other identical elements in the process, method or device comprising the elements. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments may refer to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can refer to the description of the method part.

[0093] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods for each specific application to implement the described functions, but such implementations should not be considered to exceed the scope of the embodiments of the present disclosure. The technicians may clearly understand that, for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above may refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0094] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units can be only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to implement this embodiment. In addition, each functional unit in the embodiment of the present disclosure may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit.

[0095] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for modeling HVAC control equipment, characterized in that: include: Pre-train all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, which is established according to the mechanism model or full-operating-condition model data of the HVAC control equipment; According to the on-site initial equipment full-operating condition model data table, some model parameters of the general model are updated to obtain a global high-precision model. The on-site initial equipment full-operating condition model data table is generated by updating the on-site small sample initial data of the HVAC control equipment to the equipment full-operating condition data model table; According to the on-site real-time equipment full-operating condition model data table, some model parameters of the global high-precision model are updated to obtain a real-time global high-precision model. The on-site real-time equipment full-operating condition model data table is updated according to the on-site real-time operation data of the HVAC control equipment to the on-site initial equipment full-operating condition model data table.

2. The method according to claim 1, characterized in that According to the mechanism model or full-condition model data of the HVAC control equipment, establish the equipment full-condition model data table, including: Establish a full-operating-condition model data table for the equipment based on the full-operating-condition model data of the HVAC control equipment; or, A full-operating-condition model data table of the equipment is established according to the mechanism model or full-operating-condition model data of similar equipment; wherein the similarity of equipment parameters of similar equipment and HVAC control equipment is within a similar range.

3. The method according to claim 2, characterized in that Each device full operating condition model data table of the HVAC control equipment can store n operating condition ranges as input characteristic data, and each operating condition range can store k output characteristic data; where n is the number of operating condition ranges, and k is the number of output characteristic data that can be stored in each operating condition range.

4. The method according to any one of claims 1 to 3, characterized in that: The general model includes a convolutional neural network (CNN) and / or a multi-layer perceptron (MLP). All model parameters of the corresponding general model are pre-trained according to the equipment full-condition model data table, including: When the type of the equipment full-operating-condition model data table is a chiller model data table or a cooling tower water supply model data table, all model parameters of CNN and MLP are pre-trained using the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output; When the type of the equipment full-operating-condition model data table is a cooling tower fan power model data table, a cooling pump power model data table, a cooling pump flow model data table, a refrigeration pump power model data table, or a refrigeration pump flow model data table, all model parameters of the MLP are pre-trained using the input feature data of the equipment full-operating-condition model data table as input and the output feature data of the equipment full-operating-condition model data table as output.

5. The method according to any one of claims 1 to 3, characterized in that: According to the full-condition model data table of the initial equipment on site, some model parameters of the general model are updated to obtain a global high-precision model, including: When the type of the global high-precision model is CNN and MLP, the actual input feature data of the full-condition model data table of the on-site initial equipment is input into the general model to obtain the fuzzy output value; Compare and calculate the fuzzy output value with the actual output characteristic data of the full-operating condition model data table of the initial equipment on site to obtain the loss function; The model parameters of the MLP layer are updated by back-propagation of the gradient descent of the loss function, while the model parameters of the CNN layer are kept unchanged to obtain a global high-precision model.

6. The method according to any one of claims 1 to 3, characterized in that: According to the full-condition model data table of the initial equipment on site, some model parameters of the general model are updated to obtain a global high-precision model, including: When the type of the global high-precision model is MLP, the actual input characteristic data of the full-operating condition model data table of the on-site initial equipment is input into the general model to obtain the fuzzy output value; Compare the fuzzy output value with the actual output characteristic data of the full-operating condition model data table of the initial equipment on site to calculate the loss function; The model parameters of the last layer of MLP are updated by back-propagation of the gradient descent of the loss function, while the model parameters of other layers of MLP are kept unchanged to obtain a global high-precision model.

7. The method according to any one of claims 1 to 3, characterized in that: According to the on-site real-time operation data of the HVAC control equipment, the on-site initial equipment full-condition model data table is updated to obtain the on-site real-time equipment full-condition model data table, including: When the HVAC control equipment is running and the model update time is reached, determine the amount of real-time operation data on site; The on-site initial equipment full-operating condition model data table is updated according to the quantity of on-site real-time operation data to obtain the on-site real-time equipment full-operating condition model data table.

8. A device for modeling HVAC control equipment, characterized in that: include: A pre-training module is configured to pre-train all model parameters of the corresponding general model according to the equipment full-operating-condition model data table, where the equipment full-operating-condition model data table is established according to the mechanism model or full-operating-condition model data of the HVAC control equipment; A global high-precision model generation module is configured to update some model parameters of the general model according to the on-site initial equipment full-operating condition model data table to obtain a global high-precision model. The on-site initial equipment full-operating condition model data table is generated by updating the on-site small sample initial data of the HVAC control equipment into the equipment full-operating condition data model table; The real-time global high-precision model generation module is configured to update some model parameters of the global high-precision model according to the on-site real-time equipment full-operating condition model data table to obtain a real-time global high-precision model. The on-site real-time equipment full-operating condition model data table is obtained by updating the on-site initial equipment full-operating condition model data table according to the on-site real-time operation data of the HVAC control equipment.

9. A device for modeling HVAC control equipment, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the method for HVAC control device modeling according to any one of claims 1 to 7 when running the program instructions.

10. A HVAC control device, characterized in that: include: HVAC control equipment body; The device for modeling HVAC control equipment as described in claim 8 or 9 is installed on the HVAC control equipment body.

11. A storage medium storing program instructions, characterized in that: When the program instructions are executed, the method for modeling HVAC control equipment according to any one of claims 1 to 7 is executed.

Images