Central air conditioner terminal energy consumption processing method based on large language model

Abstract

The invention relates to the technical field of air conditioner energy consumption, in particular to a central air conditioner terminal energy consumption processing method based on a large language model, which comprises the following steps of: S1, storing air conditioner science data including air conditioner principle data, air conditioner fault data, a heat transfer science document and a refrigeration science document and related data of an air conditioner over the years in a data lake; normalizing the numerical data to obtain a data set; s2, inputting air-conditioning data into a large language model for learning, inputting the training set into a regression model for training to obtain a training model, and inputting the prediction set into the training model to obtain a coil regulation coefficient; and S3, the energy consumption amount of the central air conditioner is calculated according to the coil pipe regulation and control coefficient and the real-time data, and an energy consumption statistical report is generated. According to the method, the air-conditioning data is learned through the large language model, the coil regulation coefficient is obtained according to the historical data, and the energy consumption of the central air-conditioning room is accurately calculated according to the coil regulation coefficient and the real-time data, so that a scientific basis is provided for energy-saving optimization, and the diagnosis and maintenance efficiency of the air-conditioning failure rate is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning energy consumption, and in particular to a central air conditioning terminal energy consumption processing method, device and storage medium based on a large language model. Background Art

[0002] Central air conditioning systems have been widely used because they can provide comfortable cooling and heating environments for large areas. Central air conditioning systems are usually composed of outdoor units, indoor units, and complex pipes and control systems. Among them, the indoor unit is the key part that directly exchanges heat with the indoor environment, and its energy consumption occupies an important position in the energy management of the entire central air conditioning system. The energy consumption accounting of the indoor unit is of great significance to improving energy utilization efficiency and achieving energy conservation and emission reduction.

[0003] The traditional method of calculating the energy consumption of the indoor unit of central air conditioners has many limitations. On the one hand, the method of manually recording energy consumption data is not only inefficient, but also prone to human errors. The recorder may cause incomplete or inaccurate data due to negligence, unreasonable recording time intervals, etc. On the other hand, simple automatic calculations are often based on relatively rough models, which only consider a few fixed parameters such as operating time and set temperature, but fail to fully integrate the actual operating conditions of the indoor unit. For example, complex factors such as the fan speed of the indoor unit, the actual heat exchange efficiency, and the impact of changes in ambient temperature and humidity in different seasons and time periods on energy consumption are not fully included in the calculation scope. This makes it difficult to accurately reflect the actual energy consumption of the indoor unit. , And it is not conducive to the formulation of accurate and effective energy-saving strategies and energy management measures.

[0004] At the same time, central air conditioners will inevitably encounter various faults during long-term operation. When a fault occurs, maintenance personnel usually need to go to the site for investigation and diagnosis. The traditional fault diagnosis method mainly relies on the past experience of maintenance personnel. Relying on the memory and understanding of common fault phenomena, maintenance personnel gradually check the components and system links that may have problems. However, this empirical diagnosis method has obvious shortcomings. First, it takes a long time to accumulate experience. For some new or relatively inexperienced maintenance personnel, it is difficult to accurately judge the fault. Secondly, even experienced maintenance personnel may misdiagnose or take too long to diagnose some more complex, rare or multiple faults. For example, some faults may show similar symptoms, but the actual root causes of the faults are completely different. It is difficult to quickly and accurately determine the real fault point based on experience alone, which not only leads to low maintenance efficiency, but also may cause additional energy waste due to maintenance delays, increasing the user's use cost and inconvenience. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and provide a central air-conditioning terminal energy consumption processing method based on a large language model, comprising the following steps: S1: Store air conditioning theory data, air conditioning fault data, heat transfer documents, refrigeration documents, fluid mechanics documents, HVAC data and historical air conditioning related data into the data lake, and normalize the numerical data of the historical air conditioning related data to obtain a data set; S2: input the air conditioning data into a large language model for learning, select the gradient boosting tree GBM as a regression model, input the training set of the data set into the regression model for training to obtain a training model, and input the prediction set of the data set into the training model to obtain a coil control coefficient; S3: Calculate the energy consumption of the central air conditioner according to the coil control coefficient and real-time data and generate an energy consumption statistical report.

[0006] Preferably, the real-time data includes: Obtain the physical space location and relationship of the central air-conditioning outdoor unit, coil, main electric meter, and sub-meter electric meter; Get the brand of the room air conditioner, brand rated power, room area, and current electricity price; Collect air-conditioning related data on air-conditioning sensors including air-conditioning mode, wind speed, set temperature, sweeping mode, indoor temperature, outdoor temperature, air-conditioning running time, and related sub-meter readings, and clean the air-conditioning related data to ensure data validity.

[0007] Preferably, in step S3, calculating the energy consumption of the central air conditioner according to the coil control coefficient and the real-time data further includes: S31: Obtain the number of floors of the coil and the number of floors of the sub-meter according to the association relationship, calculate the proportion of the coil electricity consumption according to the number of floors of the coil and the number of floors of the sub-meter of the meter, and calculate the proportion of the coil electricity consumption The calculation formula is as follows: in, is the number of floors of the coil, The number of floors of the electric meter sub-meter; S32: Calculate the power consumption coefficient of a single coil according to the coil control coefficient and the coil operation time. The calculation formula is as follows: in, is the coil control coefficient, is the coil operation time; S33: Calculate the total power consumption of all coils according to the coil power consumption proportion and the power consumption of the sub-meter. The calculation formula is as follows: in, The power consumption of the floor terminal sub-meter within the command time; S34: Calculate the power consumption of each coil per second according to the total power consumption of all coils and the coil control coefficient, that is, the total power consumption of all coils per unit time divided by the operation time of each coil per unit time multiplied by the sum of the coil control coefficients is equal to the power consumption of each coil per second. The calculation formula is as follows: S35: Calculating the power consumption of the fan according to the rated power of the fan of the brand and the coil control coefficient; S36: Calculate the energy consumption of each coil per second according to the coil power consumption apportionment ratio, the power consumption of each coil per second, and the power consumption of the fan; S37: Calculate the power consumption of the room according to the power consumption of the coil per second and the startup operation time, and calculate the room electricity fee according to the unit price of electricity fee and the total power consumption of the room.

[0008] Preferably, in step S35, the power consumption of the fan is calculated according to the rated power of the fan of the brand and the coil control coefficient, further comprising: in, is the rated power of the fan of the brand mentioned, Power is used for the fan.

[0009] Preferably, in step S36, the energy consumption of each coil per second is calculated according to the coil power consumption apportionment ratio, the power consumption of each coil per second and the power consumption of the fan, further comprising: in, The energy consumption per coil per second.

[0010] Preferably, in step S37, the room power consumption is calculated according to the power consumption per second of the coil and the startup operation time, and the room electricity fee is calculated according to the unit price of electricity fee and the total power consumption of the room, further comprising: The room power consumption is calculated based on the power consumption of the coil per second and the startup operation time. The calculation formula is as follows: in, Indicates the boot-up running time per unit time; Calculate the room electricity cost per unit time based on the unit price of electricity and the total electricity consumption of the room. The calculation formula is as follows: in, The unit price of electricity.

[0011] Preferably, an energy consumption statistical report including daily energy consumption, monthly energy consumption and annual energy consumption is generated according to the coil electricity consumption sharing ratio, the single coil electricity consumption coefficient, the total electricity consumption of all coils, the electricity consumption of each coil per second, the fan electricity consumption, the energy consumption of each coil per second, the room electricity consumption and the room electricity fee, the energy consumption statistical report is displayed through a visualization tool, and energy-saving suggestions are provided based on the energy consumption statistical report.

[0012] Preferably, when air-conditioning fault information is monitored, the air-conditioning fault information is passed to the large language model, the large prediction model obtains real-time data of the indoor unit, and obtains diagnostic information based on the air-conditioning data, proposes a diagnostic plan based on the diagnostic information, and pushes the diagnostic information to stakeholders in the form of mobile phone APP or mini-program text messages, and the stakeholders handle it according to the diagnostic plan.

[0013] Based on the same concept, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the central air-conditioning terminal energy consumption processing method based on a large language model as described in any one of the embodiments.

[0014] Based on the same concept, the present invention also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the central air-conditioning terminal energy consumption processing method based on a large language model as described in any one of the embodiments.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention stores air-conditioning data including air-conditioning principle data, air-conditioning fault data, heat transfer documents, refrigeration documents, fluid mechanics documents, HVAC and air-conditioning related data over the years in the data lake, which makes it easy to manage data and information, and facilitates maintenance and updating. By normalizing the numerical data of air-conditioning related data over the years to obtain a data set, data of different magnitudes and scales can be unified within a standard range, eliminating the impact of differences in data dimensions, etc., ensuring the consistency of data in subsequent large language models, fault diagnosis and other processes, and improving the accuracy and reliability of the model.

[0016] The present invention inputs air-conditioning data into a large language model for learning, mines the associations between different knowledge points, forms a complete knowledge network, and facilitates the updating and learning of new air-conditioning data. This enables it to better adapt to and use new knowledge to assist in analysis and judgment when facing the ever-changing actual application scenarios of air-conditioning; by selecting the gradient boosting tree GBM as the regression model, it has robustness and noise resistance, so that the prediction results output by the model can more accurately approach the actual coil control coefficient.

[0017] The present invention calculates the energy consumption of the central air conditioner through the coil control coefficient and real-time data and generates an energy consumption statistical report, which effectively solves the shortcomings of the traditional energy consumption calculation method, especially when the air conditioner provides scientific and reasonable inferences for the situation that the real-time data of the air conditioner is lost and cannot be calculated. It is of great significance for improving energy utilization efficiency and reducing operating costs. At the same time, it combines the large language model to give a more scientific diagnosis plan for air conditioning failure, improves the efficiency of air conditioning failure repair and reduces maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the following detailed description of the preferred embodiment.The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting the invention.

[0019] Figure 1 It is a flow chart of the central air-conditioning terminal energy consumption processing method based on the large language model of the present invention; Figure 2 Another flow chart of the central air-conditioning terminal energy consumption processing method based on the large language model of the present invention; Figure 3 It is a schematic diagram of the relationship between the six floors of the room and the electric meter of the central air-conditioning terminal energy consumption processing method based on the large language model of the present invention; Figure 4 It is a structural diagram of the central air-conditioning terminal energy consumption processing system based on the large language model of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0022] Example 1 See also Figure 1 and Figure 2 As shown, this embodiment combines the large prediction model with the energy consumption calculation and maintenance of the central air conditioner to more accurately calculate the energy consumption and fault diagnosis of the central air conditioner terminal, and provides a central air conditioner terminal energy consumption processing method based on the large language model, including the following steps: S1: Store air-conditioning data including air-conditioning principle data, air-conditioning fault data, heat transfer documents, refrigeration documents, fluid mechanics documents, HVAC and historical air-conditioning related data in the data lake, and normalize the numerical data of historical air-conditioning related data to obtain a data set. Specifically, in this embodiment, the air-conditioning related data for the past 10 years is stored in the data lake. The data lake is a storage system used to centrally store data information, which can accommodate any structured and unstructured data.

[0023] Preferably, the collected document materials in various formats including PDF, Word, HTML, etc. are uniformly converted into a format suitable for processing and storage, such as a plain text format or a structured database format (such as extracting tabular data into a relational database table).

[0024] Clean the relevant data of air conditioners over the years to remove duplicate data records, erroneous data values ​​(such as temperature or pressure values ​​that are obviously beyond the reasonable range) and incomplete data rows or columns. For missing data, data filling techniques (such as mean filling, model-based filling, etc.) can be used to process it.

[0025] Since the data related to air conditioners over the years have the characteristics of relatively uniform data distribution and known data range, the minimum-maximum normalization formula is used for conversion, as shown below: in, is the normalized value, is the original data value, and are the minimum and maximum values ​​of the variable respectively. This method can map the data to the [0,1] interval.

[0026] S2: Input air conditioning data into the large language model for learning, select the gradient boosting tree GBM as the regression model, input the training set of the data set into the regression model for training to obtain the training model, and input the prediction set of the data set into the training model to obtain the coil control coefficient. Specifically, in this embodiment, 80% of the data in the data set is used as the training set, and 20% of the data is used as the prediction set to supervise and fine-tune the large language model LLMs. The large language model is a pre-trained language model used to pre-train large-scale data, which can imitate humans to learn and think.

[0027] The gradient boosting tree GBM captures this complex nonlinear feature very well. During the training process, the weights of each decision tree are continuously adjusted so that the prediction results output by the model can be more accurately close to the actual coil control coefficient. Compared with simple models such as linear regression, it has a significant accuracy advantage in processing such complex relationships. When actually collecting air-conditioning related data, there will inevitably be some noise data, such as individual abnormal temperature values ​​and flow values ​​caused by sensor measurement errors. The GBM model has a certain tolerance for such noise data. During the training process, it will not suffer from serious overfitting due to a small amount of abnormal data. It can still learn the main rules in the data relatively stably, ensuring the reliability of the final training model in predicting the coil control coefficient and reducing the prediction bias caused by data quality issues.

[0028] Preferably, numerical variables such as indoor and outdoor temperature, humidity, supply and return air temperature of the air conditioning system, refrigerant temperature and pressure at the coil inlet and outlet, fan speed, and non-numerical variables that may affect regulation (such as whether the air conditioner is operating in cooling or heating mode, etc.) are input into the large language model.

[0029] The large language model analyzes the potential correlation between each variable in the data and the coil control coefficient, and removes those variables with low correlation, redundancy or no significant impact on the coil control coefficient through data analysis methods (such as calculating the correlation coefficient between variables, evaluating the importance of features, etc.), and selects the characteristic variables that have a significant impact on the coil control coefficient. For example, by calculating the correlation coefficient between variables (such as the Pearson correlation coefficient, etc.), variables that are highly correlated with the control coefficient, such as the coil inlet and outlet temperature difference, the indoor and outdoor temperature difference, and the fan air volume, are determined as key features.

[0030] Import key features into the corresponding machine learning library, select the gradient boosting tree GBM related classes or functions to initialize the regression model, and preliminarily set the key parameters of the gradient boosting tree GBM according to the data characteristics and problem requirements. For example, set the number of decision trees (n_estimators parameter) to a moderate value (such as 100) to start, set the maximum depth of the tree (max_depth parameter) to a certain number of layers (such as 5 layers), and set the learning rate (learning_rate parameter) to a smaller value (such as 0.1). These parameters will be further adjusted and optimized according to the model performance during the training process.

[0031] The training set data is input into the initialized GBM regression model and iterative training begins. In each iteration, GBM builds a new decision tree based on the residual between the current prediction result and the true control coefficient target value to continuously optimize the model's prediction ability, thereby reducing this residual and obtaining a trained GBM regression model.

[0032] The preprocessed prediction set data is input into the trained GBM regression model. The model will predict the coil control coefficient corresponding to each data in the prediction set based on the relationship between the learned features and the coil control coefficient, and output the corresponding prediction results.

[0033] S3: Calculate the energy consumption of the central air conditioner based on the coil control coefficient and real-time data and generate an energy consumption statistical report. Specifically, in this embodiment, the coil control coefficient is a billing coefficient value calculated by the large language model. The real-time data related to the air conditioner is collected through the collection system. The default control coefficient of the untrained initial coil is 1.

[0034] Preferably, the real-time data includes: Obtain the physical space location and relationship of the central air-conditioning outdoor unit, coil, main electric meter, and sub-meter electric meter; Get the brand of the room air conditioner, brand rated power, room area, and current electricity price; Air conditioning related data on the air conditioning sensor including air conditioning mode, wind speed, set temperature, sweeping mode, indoor temperature, outdoor temperature, air conditioning running time, and related sub-meter readings are collected, and the air conditioning related data are cleaned to ensure data validity. Specifically, in this embodiment, the collected real-time data is cleaned, denoised, and standardized.

[0035] Preferably, in step S3, calculating the energy consumption of the central air conditioner according to the coil control coefficient and the real-time data further includes: S31: The number of floors of the coil and the number of floors of the sub-meters are obtained according to the association relationship, and the proportion of the electricity consumption of the coil is calculated according to the number of floors of the coil and the number of floors of the sub-meters. The calculation formula is as follows: in, is the number of coil floors, is the number of floors of the electric meter sub-meter. Specifically, in this embodiment, please refer to Figure 3 As shown in the figure, suppose a building has 10 floors, 9 floors are equipped with sub-meters, and 8 floors are equipped with coils. , , so that the floors without coils and sub-meters installed can be excluded; S32: Calculate the power consumption coefficient of a single coil based on the coil control coefficient and coil operation time. The calculation formula is as follows: in, is the coil control coefficient, is the coil operation time; S33: Calculate the total power consumption of all coils based on the proportion of coil power consumption and the power consumption of the sub-meter. The calculation formula is as follows: in, The power consumption of the floor terminal sub-meter within the command time; S34: Calculate the power consumption of each coil per second based on the total power consumption of all coils and the coil control coefficient, that is, the total power consumption of all coils per unit time divided by the running time of each coil per unit time multiplied by the sum of the coil control coefficients is equal to the power consumption of each coil per second. The calculation formula is as follows: S35: Calculate the fan power consumption based on the brand fan rated power and coil control coefficient; S36: Calculate the energy consumption of each coil per second according to the coil power consumption proportion, the power consumption of each coil per second, and the power consumption of the fan; S37: Calculate the room power consumption according to the power consumption per second of the coil and the startup operation time, and calculate the room electricity fee according to the unit price of electricity fee and the total power consumption of the room. Specifically, in this embodiment.

[0036] Preferably, in step S35, the power consumption of the fan is calculated according to the rated power of the brand fan and the coil control coefficient, further comprising: in, is the rated power of the brand fan, Power for the fan.

[0037] Preferably, in step S36, the energy consumption of each coil per second is calculated according to the coil power consumption apportionment ratio, the power consumption of each coil per second and the power consumption of the fan, further comprising: in, The energy consumption per coil per second.

[0038] Preferably, in step S37, the room power consumption is calculated according to the power consumption per second of the coil and the startup operation time, and the room electricity fee is calculated according to the unit price of electricity fee and the total power consumption of the room, further comprising: The room power consumption is calculated based on the coil power consumption per second and the startup time. The calculation formula is as follows: in, Indicates the boot-up running time per unit time; Calculate the room electricity cost per unit time based on the unit price of electricity and the total electricity consumption of the room. The calculation formula is as follows: in, The unit price of electricity.

[0039] Preferably, an energy consumption statistical report including daily energy consumption, monthly energy consumption and annual energy consumption is generated based on the coil electricity consumption apportionment ratio, the single coil electricity consumption coefficient, the total electricity consumption of all coils, the electricity consumption of each coil per second, the fan electricity consumption, the energy consumption of each coil per second, the room electricity consumption and the room electricity fee. The energy consumption statistical report is displayed through a visualization tool, and energy-saving suggestions are provided based on the energy consumption statistical report.

[0040] Preferably, when air-conditioning fault information is monitored, the air-conditioning fault information is passed to the large language model, the large prediction model obtains real-time data of the indoor unit, and obtains diagnostic information based on air-conditioning data, proposes a diagnostic plan based on the diagnostic information, and pushes the diagnostic information to stakeholders in the form of mobile phone APP or mini-program text messages. The stakeholders process according to the diagnostic plan. Specifically, in this embodiment, if there is any arrears, the stakeholders can pay the bill, and if it is a maintenance personnel, they can quickly process it according to the diagnostic plan, thereby saving time for troubleshooting the air-conditioning fault and improving the work efficiency of the maintenance personnel.

[0041] Example 2 See also Figure 4As shown, this embodiment also provides a central air-conditioning terminal energy consumption processing system based on a large language model, including: Data lake module: the storage module includes data storage units such as heat transfer unit, refrigeration unit, fluid mechanics unit, air conditioning failure and maintenance unit, air conditioning composition principle unit and air conditioning collection database unit, which are used to collect air conditioning data including air conditioning principle data, air conditioning failure data, heat transfer documents, refrigeration documents, fluid mechanics documents, HVAC and other air conditioning data and related data over the years, and store them in the data storage unit; The large language model processing module obtains air conditioning data from the data storage unit for learning, selects the gradient boosting tree GBM as the regression model, normalizes the numerical data of air conditioning related data over the years to obtain a data set, inputs the training set of the data set into the regression model for training to obtain a training model, inputs the prediction set of the data set into the training model to obtain the coil control coefficient, and sends the coil control coefficient to the intelligent air conditioning management module; Data middle platform module: The data middle platform module is used to obtain real-time data and coil control coefficients from the intelligent air-conditioning management module through the acquisition unit, store the real-time data in the distributed database unit, calculate the energy consumption of the central air-conditioning according to the coil control coefficient and real-time data, and generate an energy consumption statistical report; Intelligent air conditioning management module, the intelligent air conditioning management module includes a building management unit, an air conditioning management unit, a billing unit, a notification unit, a fault unit and an early warning unit, which are used to collect real-time air conditioning data and process central air conditioning energy consumption and energy consumption statistical reports. The building management unit is used to obtain the physical space location and association relationship of the central air conditioning outdoor unit, coil, total electric meter, sub-meter electric meter, etc. The air conditioning management unit is used to obtain the air conditioning mode, wind speed, set temperature, sweeping mode, etc. The billing unit is used to calculate the cost according to the central air conditioning energy consumption. The early warning unit is used to obtain air conditioning fault information according to the energy consumption statistical report. The fault unit is used to generate diagnostic information and propose a diagnostic plan based on the air conditioning fault information. The notification unit is used to send the diagnostic information and diagnostic plan to the stakeholders.

[0042] Example 3 In some embodiments of the present application, a computer device is also provided, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the central air-conditioning terminal energy consumption processing method based on a large language model in Example 1 of the present invention.

[0043] The present invention also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the central air-conditioning terminal energy consumption processing method based on a large language model as described in any one of Example 1.

[0044] It is understandable that, for the aforementioned central air-conditioning terminal energy consumption processing method based on a large language model, if it is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0045] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0046] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for processing energy consumption of central air conditioner terminals based on a large language model, characterized in that: The following steps are involved: S1: Store air conditioning theory data, air conditioning fault data, heat transfer documents, refrigeration documents, fluid mechanics documents, HVAC data and historical air conditioning related data into the data lake, and normalize the numerical data of the historical air conditioning related data to obtain a data set; S2: input the air conditioning data into a large language model for learning, select the gradient boosting tree GBM as a regression model, input the training set of the data set into the regression model for training to obtain a training model, and input the prediction set of the data set into the training model to obtain a coil control coefficient; S3: Calculate the energy consumption of the central air conditioner according to the coil control coefficient and real-time data and generate an energy consumption statistical report.

2. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 1 is characterized in that: Real-time data includes: Obtain the physical space location and relationship of the central air-conditioning outdoor unit, coil, main electric meter, and sub-meter electric meter; Get the brand of the room air conditioner, brand rated power, room area, and current electricity price; Collect air-conditioning related data on air-conditioning sensors including air-conditioning mode, wind speed, set temperature, sweeping mode, indoor temperature, outdoor temperature, air-conditioning running time, and related sub-meter readings, and clean the air-conditioning related data to ensure data validity.

3. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 2 is characterized in that: In step S3, calculating the energy consumption of the central air conditioner according to the coil control coefficient and the real-time data further includes: S31: Obtain the number of floors of the coil and the number of floors of the sub-meter according to the association relationship, calculate the proportion of the coil electricity consumption according to the number of floors of the coil and the number of floors of the sub-meter of the meter, and calculate the proportion of the coil electricity consumption The calculation formula is as follows: in, is the number of floors of the coil, The number of floors of the electric meter sub-meter; S32: Calculate the power consumption coefficient of a single coil according to the coil control coefficient and the coil operation time. The calculation formula is as follows: in, is the coil control coefficient, is the coil operation time; S33: Calculate the total power consumption of all coils according to the coil power consumption proportion and the power consumption of the sub-meter. The calculation formula is as follows: in, The power consumption of the floor terminal sub-meter within the command time; S34: Calculate the power consumption of each coil per second according to the total power consumption of all coils and the coil control coefficient, that is, the total power consumption of all coils per unit time divided by the operation time of each coil per unit time multiplied by the sum of the coil control coefficients is equal to the power consumption of each coil per second. The calculation formula is as follows: S35: Calculating the power consumption of the fan according to the rated power of the fan of the brand and the coil control coefficient; S36: Calculate the energy consumption of each coil per second according to the coil power consumption apportionment ratio, the power consumption of each coil per second, and the power consumption of the fan; S37: Calculate the power consumption of the room according to the power consumption of the coil per second and the startup operation time, and calculate the room electricity fee according to the unit price of electricity fee and the total power consumption of the room.

4. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 3 is characterized in that: In step S35, the power consumption of the fan is calculated according to the rated power of the fan of the brand and the coil control coefficient, further comprising: in, is the rated power of the fan of the brand mentioned, Power is used for the fan.

5. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 4 is characterized in that: In step S36, the energy consumption of each coil per second is calculated according to the coil power consumption apportionment ratio, the power consumption of each coil per second and the power consumption of the fan, further comprising: in, The energy consumption per coil per second.

6. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 5 is characterized in that: In step S37, the room power consumption is calculated according to the power consumption per second of the coil and the startup operation time, and the room electricity fee is calculated according to the unit price of electricity fee and the total power consumption of the room, further comprising: The room power consumption is calculated based on the power consumption of the coil per second and the startup operation time. The calculation formula is as follows: in, Indicates the boot-up running time per unit time; Calculate the room electricity cost per unit time based on the unit price of electricity and the total electricity consumption of the room. The calculation formula is as follows: in, The unit price of electricity.

7. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 6 is characterized in that: An energy consumption statistical report including daily energy consumption, monthly energy consumption and annual energy consumption is generated based on the coil electricity consumption apportionment ratio, the single coil electricity consumption coefficient, the total electricity consumption of all coils, the electricity consumption of each coil per second, the fan electricity consumption, the energy consumption of each coil per second, the room electricity consumption and the room electricity fee. The energy consumption statistical report is displayed through a visualization tool, and energy-saving suggestions are provided based on the energy consumption statistical report.

8. The central air-conditioning terminal energy consumption processing method based on a large language model according to claim 7 is characterized in that: When air conditioning fault information is monitored, the air conditioning fault information is passed to the large language model, the large prediction model obtains real-time data of the indoor unit, and obtains diagnostic information based on the air conditioning data, proposes a diagnostic plan based on the diagnostic information, and pushes the diagnostic information to stakeholders in the form of mobile phone APP or mini-program text messages, and the stakeholders handle it according to the diagnostic plan.

9. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the central air-conditioning terminal energy consumption processing method based on a large language model as described in any one of claims 1 to 8.

10. A storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the central air-conditioning terminal energy consumption processing method based on a large language model as described in any one of claims 1 to 8.

Images