Central air conditioner supply effect AI intelligent evaluation method and system
By monitoring air conditioning operation data in real time and evaluating it using AI models, functions for air conditioning temperature change, speed setting, and temperature effect are constructed, solving the problem of passive evaluation of central air conditioning systems and realizing intelligent air conditioning management and energy-saving control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHENGZHOU CHUNQUAN ENERGY SAVING
- Filing Date
- 2025-12-30
- Publication Date
- 2026-05-12
AI Technical Summary
In existing technologies, the evaluation of the performance of central air conditioning systems mainly relies on user feedback, leaving operators in a passive position. There is a lack of automated evaluation methods, which cannot effectively assist in air conditioning management and energy-saving control.
By monitoring air conditioner operation data and environmental data in real time, and using an AI intelligent evaluation model for comprehensive evaluation, the system constructs functions for air conditioner temperature change, setting, and temperature effect. The model is trained using linear regression or random forest algorithms to achieve intelligent evaluation of the air conditioner's supply effect.
It enables intelligent prediction and dynamic adjustment of air conditioning performance, supports intelligent control of air conditioning, improves management level, and ensures user comfort and energy-saving operation.
Smart Images
Figure CN122020602A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of air conditioning system usage monitoring technology, and in particular to an AI-powered intelligent evaluation method and system for the supply effect of central air conditioning. Background Technology
[0002] Central air conditioning systems mainly consist of refrigeration compressors, refrigerant circulation systems, fan coil units, and cooling towers. These devices work together to regulate indoor temperature. With the emergence of new economies, new industries, and new demands, the application scenarios for central air conditioning are constantly expanding, and the breadth of application industries is continuously increasing. For example, sectors covered by the "new infrastructure" industry, such as 5G infrastructure, intercity high-speed railways and urban rail transit, and big data centers, are all key service areas for central air conditioning. Through a centralized monitoring platform, users can centrally monitor and manage the split air conditioners in multiple rooms. They can remotely control the fan speed, temperature, and air conditioning mode of all split air conditioners, and also achieve intelligent energy-saving control of the air conditioners, such as adjusting fan speed, adjusting mode, turning the air conditioner on and off, and adjusting room temperature. Given the centralized supply and decentralized use of central air conditioning, how to evaluate the air conditioning effect to achieve user comfort and satisfaction has currently remained at the user feedback stage, with operators in a passive position. Therefore, there is an urgent need for a means of evaluating air conditioning performance that can assist operators in managing air conditioning through automatic analysis, especially for buildings with management platforms such as Building Automation (BA) and Energy Management System (EMS). The analysis results of air conditioning performance can provide useful support for automatic control and energy-saving management. Summary of the Invention
[0003] To address the current issues in evaluating the effectiveness of central air conditioning supply, this invention provides an AI-powered intelligent evaluation method and system for central air conditioning supply effectiveness. By monitoring real-time air conditioning operation data and environmental data in the target air-conditioned area and utilizing an AI model to intelligently evaluate the current air conditioning operation effectiveness, the system can predict the next steps in air conditioning operation, facilitating maintenance and management by operators.
[0004] According to the design scheme provided by the present invention, on the one hand, an AI-based intelligent evaluation method for the supply effect of central air conditioning is provided, comprising:
[0005] Real-time collection of indoor air conditioning operation data and target area environmental data. The air conditioning operation data includes: operating level, operating time and set temperature. The environmental data includes: indoor target area room temperature.
[0006] The preprocessed air conditioning operation data and environmental data are input into the AI intelligent evaluation model. The AI intelligent evaluation model is used to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using labeled air conditioning effect observation data samples.
[0007] As an AI-powered intelligent evaluation method for the central air conditioning supply effect of the present invention, the AI intelligent evaluation model is further expressed as: η=a*f1(δT,t)+b*f2(δt2,D,t)+c*f3(δt3,T0,T1,t), where a, b, and c are weighting coefficients, f1(δT,t) is the air conditioning temperature change effect function, f2(δt2,D,t) is the air conditioning speed effect function, f3(δt3,T0,T1,t) is the air conditioning temperature effect function, δT is the change in room temperature per unit time, t is the air conditioning running time, δt2 is the proportion of time the air conditioning speed is activated during air conditioning operation, D is the air conditioning operating speed, and δt3 is the proportion of time during air conditioning operation where the absolute value of the difference between room temperature T0 and the air conditioning set temperature T1 is less than the temperature difference threshold.
[0008] As a further embodiment of the AI-powered intelligent evaluation method for the central air conditioning supply effect of the present invention, the AI-powered intelligent evaluation model adopts a linear regression model to achieve a comprehensive evaluation of the air conditioning supply effect using the air conditioning temperature change effect function, the air conditioning speed effect function, and the air conditioning temperature effect function.
[0009] As part of the AI-powered intelligent evaluation method for the central air conditioning supply effect of this invention, the AI intelligent evaluation model training process further includes:
[0010] Multiple sets of historical observation data on air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data included air conditioning operation data at historical moments and corresponding environmental data.
[0011] The sample data is preprocessed and divided into training set and test set according to a preset ratio. The preprocessing includes missing value handling and data standardization.
[0012] The AI intelligent evaluation model is trained using a training set and evaluated using a test set to obtain an AI intelligent evaluation model that meets expectations and can be deployed in the central air conditioning control platform.
[0013] As part of the AI-powered intelligent evaluation method for central air conditioning supply effectiveness of this invention, further comprising evaluating the air conditioning supply effectiveness of the current indoor area using an AI-powered intelligent evaluation model, and including:
[0014] If multiple indoor areas are equipped with temperature sensors to collect the room temperature of the corresponding target area, the air conditioning temperature change effect function is the weighted arithmetic mean of the air conditioning temperature change effect functions of each area, and the air conditioning temperature effect function is the weighted arithmetic mean of the air conditioning temperature effect functions of each area.
[0015] If the air conditioner has multiple speed settings, the air conditioner speed setting effect function is the ratio of the weighted arithmetic sum of the running time of each air conditioner speed setting to the time length t.
[0016] As a further embodiment of the AI-powered intelligent evaluation method for the central air conditioning supply effect of the present invention, the AI intelligent evaluation model adopts the random forest algorithm to achieve a comprehensive evaluation of the air conditioning supply effect using the air conditioning temperature change effect function, the air conditioning speed effect function, and the air conditioning temperature effect function.
[0017] As part of the AI-powered intelligent evaluation method for the central air conditioning supply effect of this invention, the AI intelligent evaluation model training process further includes:
[0018] Multiple sets of historical observation data features of air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data features include the air conditioning operation data features at historical moments and the corresponding environmental data features.
[0019] The sample data is preprocessed, including missing value handling and data standardization.
[0020] Define the number of trees, the depth of each tree, and the number of features used for each node in the tree;
[0021] Training data is extracted with replacement from the preprocessed sample data to construct a training subset for each tree; for each node of a tree, a subset of features is randomly selected from the node's feature set.
[0022] For each training subset, the best feature is selected for splitting based on the splitting criteria of each feature in the feature subset, forming child nodes. The corresponding decision tree is constructed by recursively generating each child node until the tree depth is satisfied. The specified number of decision trees are generated repeatedly to form a random forest. The final prediction output of the random forest is obtained by averaging the prediction results of each decision tree.
[0023] Furthermore, this invention also provides an AI-powered intelligent evaluation system for the supply effect of central air conditioning, comprising: a data acquisition module and an effect evaluation module, wherein,
[0024] The data acquisition module is used to collect real-time indoor air conditioning operation data and target area environmental data. The air conditioning operation data includes: operating level, operating time and set temperature, and the environmental data includes: indoor target area room temperature.
[0025] The effect evaluation module is used to input the preprocessed air conditioning operation data and environmental data into the AI intelligent evaluation model, and use the AI intelligent evaluation model to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using labeled air conditioning effect observation data samples.
[0026] The beneficial effects of this invention are:
[0027] 1. This invention monitors and analyzes the room temperature T0, set temperature T1, operating level D, and operating time t of the air-conditioned area in real time. Based on real-time data and combined with historical data analysis records, the AI module evaluates the current air conditioning effect and provides support for predicting and executing the next step of air conditioning operation. It can intelligently evaluate and determine the usage effect of air conditioning terminals, supports AI intelligent control of central air conditioning, solves the problem of air conditioning terminal effect analysis, and is suitable for the judgment and adjustment control of air conditioning effect in various air conditioning usage scenarios. It can be applied to air conditioning controllers, air conditioning intelligent control systems, etc.
[0028] 2. This invention addresses the evaluation problem of achieving reliable and effective air conditioning system performance by balancing three aspects: water flow, heat supply and demand, and quality. In this invention, a low η value indicates a problem with terminal control or supply balance, providing feedback and analysis for management and operations personnel. For a single air conditioning terminal with a low η value, the first step is to check if the control setting is too low, the operating time is too short, the room is poorly insulated, or doors and windows are not properly closed. Secondary checks could involve valve malfunctions, pipe blockages, design flaws, or insufficient power configuration of the terminal air conditioner. If multiple areas show low η values, a system supply problem should be considered, including main unit malfunction, substandard air conditioning water temperature, or insufficient water circulation flow. The η value provides crucial support for the stable and effective operation of central air conditioning systems. The analysis is scientific and reasonable, and the calculation results are of practical significance and value for air conditioning operation, helping to improve air conditioning management, ensure user experience, support energy-saving operation, and contribute to building energy conservation and carbon reduction. Attached image description:
[0029] Figure 1 This is a schematic diagram of the AI-powered intelligent evaluation process for the central air conditioning supply effect in the example.
[0030] Figure 2 This is a schematic diagram of the principle architecture of the air conditioning supply effect analysis algorithm in the embodiment. Detailed implementation method:
[0031] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will be further described in detail below with reference to the accompanying drawings and technical solutions.
[0032] Regarding the issue of evaluating the effectiveness of central air conditioning supply, see the embodiments of the present invention. Figure 1 As shown, an AI-powered intelligent evaluation method for central air conditioning supply effectiveness is provided, comprising:
[0033] S101. Real-time collection of current indoor air conditioning operation data and target area environmental data, wherein the air conditioning operation data includes: operating level, operating time and set temperature, and the environmental data includes: indoor target area room temperature;
[0034] S102. Input the preprocessed air conditioning operation data and environmental data into the AI intelligent evaluation model, and use the AI intelligent evaluation model to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using the labeled air conditioning effect observation data samples.
[0035] Specifically, the AI intelligent evaluation model can be expressed as: η=a*f1(δT,t)+b*f2(δt2,D,t)+c*f3(δt3,T0,T1,t), where a, b, and c are weighting coefficients, f1(δT,t) is the air conditioner temperature change effect function, f2(δt2,D,t) is the air conditioner speed effect function, f3(δt3,T0,T1,t) is the air conditioner temperature effect function, δT is the change in room temperature per unit time, t is the air conditioner running time, δt2 is the proportion of time the air conditioner speed is activated during air conditioner operation, D is the air conditioner operating speed, and δt3 is the proportion of time during air conditioner operation where the absolute value of the difference between room temperature T0 and the air conditioner set temperature T1 is less than the temperature difference threshold.
[0036] The output of η can be used to adjust the operating parameters of the air conditioning supply and is applicable to the control of management software such as BA automatic control system and EMS smart energy management system.
[0037] The air conditioning temperature change effect function f1(δT, t) calculates the change (increase / decrease) of room temperature δT per unit time. A higher δT value results in a higher air conditioning temperature change effect function value; a lower δT value results in a lower air conditioning temperature change effect function value. For air conditioning settings, there are different settings such as high, medium, and low, each with varying impact on the effect. f2(δt2, D, t) is the air conditioning setting effect function. During air conditioning operation, the operating setting D is monitored and statistically analyzed to calculate the percentage of time δt2 spent at that setting. δt2 is equal to the ratio of the weighted arithmetic sum of the operating times of each setting to the time length t. A higher δt2 results in a lower air conditioning setting effect function value, and vice versa. f3(δt3, T0, T1, t) is the air conditioning temperature effect function. During air conditioning operation, the absolute value of the difference between room temperature T0 and the set temperature T1 is calculated, along with the percentage of time δt3 where the absolute value is less than the temperature difference threshold n (n≥0). A higher δt3 results in a higher air conditioning temperature effect function value, and vice versa.
[0038] During the transition from shutdown to startup phase of the central air conditioning system, this embodiment utilizes AI-powered intelligent analysis of the supply effect. Based on a "data input-output processing-result output" architecture, it acquires room temperature data (T0) through sensors. The time variation of T0 is processed as secondary data (δT). Anomaly processing is performed on the secondary data δT, including cleaning up noisy data and filling in missing time points, ensuring data quality meets algorithm requirements. The AI employs linear regression learning algorithms and / or decision trees. On various storage media systems, the algorithm is trained using GPUs, TPUs, and other computing power hardware to achieve data computation. The algorithm's calculated results are then output and applied to fields such as Building Automation (BA) and Energy Management Systems (EMS). Analysis of the air conditioning temperature variation effect provides support for rationally adjusting operating parameters and improving the supply plan.
[0039] If multiple indoor areas are equipped with temperature sensors to collect the room temperature of the corresponding target area, the air conditioning temperature change effect function can be the weighted arithmetic mean of the air conditioning temperature change effect functions of each area, and the air conditioning temperature effect function can be the weighted arithmetic mean of the air conditioning temperature effect functions of each area.
[0040] To improve the effectiveness of AI intelligent analysis, the number of sensors is increased to increase the number of input data samples. Room temperature data is collected from multiple locations, and the air conditioning temperature change effect function value can be calculated independently for each location. The weighted arithmetic mean of the effect values from each location is the final effective value δT, which is then entered into the AI calculation stage.
[0041] When the central air conditioning supply effect is low, AI analysis can provide decision-making and remedial measures, including control settings, room enclosure, fan valve malfunctions, and pipe air blockages. If the supply effect remains low, then the air conditioning supply volume and water temperature should be checked and adjusted. For example, if the effect is poor in office buildings in the morning, the start-up time, set temperature, and number of main units should be verified.
[0042] During the stable operation phase of the air conditioner, the air conditioner speed effect function is calculated by monitoring and statistically analyzing the operating speed D, and determining the percentage of time the speed is used for starting operation, δt2. δt2 is equal to the ratio of the weighted arithmetic sum of the operating times for each speed to the time length t. The higher δt2, the lower the air conditioner speed effect function value; conversely, the lower δt2, the higher the air conditioner speed effect function value. The absolute value of the difference between room temperature T0 and set temperature T1 is calculated, and the percentage of time δt3 where the absolute value is less than the temperature difference threshold n (n≥0) is calculated. Room temperature data is collected from multiple locations, and the air conditioner temperature effect function value is calculated independently for each location. All values are then weighted and the arithmetic mean is taken as the final effective value of the air conditioner effect function.
[0043] During the stable operation phase of the central air conditioning system, this project implemented AI-powered intelligent analysis of supply effectiveness. Based on a "data input-output processing-result output" architecture, it acquired room temperature T0 and air level D from multiple locations via sensors. This data was first processed into secondary data δt2 (air level effect value) and δt3 (temperature effect value). Anomaly processing was then performed on the secondary data δt2 and δt3, including cleaning up noise data and filling in missing time points, ensuring data quality met the algorithm's requirements. The AI employed linear regression learning algorithms and decision trees, and was trained on various storage media systems using GPUs, TPUs, and other computing power hardware to perform data calculations. The algorithm's calculated results were then output and applied to areas such as Building Automation (BA) and Energy Management Systems (EMS). The analysis of air conditioning temperature variations provided support for rationally adjusting operating parameters and improving supply solutions.
[0044] When the central air conditioning supply effect is low, AI analysis can be used to provide inspection and handling measures, including the duration of each setting, temperature settings, fan coil power, duct air blockage, and air volume. Common issues across multiple areas can be inspected and addressed based on supply volume, air conditioning water temperature, and water circulation flow rate.
[0045] The AI intelligent evaluation model employs a linear regression model to comprehensively evaluate the air conditioning supply effect using functions related to air conditioning temperature variation, air conditioning speed setting, and air conditioning temperature. Specifically, the training process for the AI intelligent evaluation model can be designed to include:
[0046] Multiple sets of historical observation data on air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data included air conditioning operation data at historical moments and corresponding environmental data.
[0047] The sample data is preprocessed and divided into training set and test set according to a preset ratio. The preprocessing includes missing value handling and data standardization.
[0048] The AI intelligent evaluation model is trained using a training set and evaluated using a test set to obtain an AI intelligent evaluation model that meets expectations and can be deployed in the central air conditioning control platform.
[0049] In linear regression, the weight vector is solved using the least squares method to fit the training samples. The goal is to find a weight vector that minimizes the difference between the predicted and actual values. Using the trained model, the value of η is predicted based on real-time air conditioning operation data and environmental data. The magnitude of the η value is then used to evaluate the current air conditioning supply effectiveness.
[0050] The AI intelligent evaluation model can also employ the random forest algorithm to comprehensively evaluate the air conditioning supply effect using the air conditioning temperature change effect function, air conditioning speed effect function, and air conditioning temperature effect function. Specifically, the training process of the AI intelligent evaluation model can be designed to include:
[0051] Multiple sets of historical observation data features of air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data features include the air conditioning operation data features at historical moments and the corresponding environmental data features.
[0052] The sample data is preprocessed, including missing value handling and data standardization.
[0053] Define the number of trees, the depth of each tree, and the number of features used for each node in the tree;
[0054] Training data is extracted with replacement from the preprocessed sample data to construct a training subset for each tree; for each node of a tree, a subset of features is randomly selected from the node's feature set.
[0055] For each training subset, the best feature is selected for splitting based on the splitting criteria of each feature in the feature subset, forming child nodes. The corresponding decision tree is constructed by recursively generating each child node until the tree depth is satisfied. The specified number of decision trees are generated repeatedly to form a random forest. The final prediction output of the random forest is obtained by averaging the prediction results of each decision tree.
[0056] We constructed feature data related to the three effect functions and the corresponding target variable η by collecting historical air conditioning operation data. We improved the model's generalization ability by constructing multiple decision trees and averaging the results.
[0057] In this solution, linear regression and / or random forest are selected as appropriate models based on air conditioning operation data and environmental parameter characteristics to intelligently evaluate and determine the usage effect of air conditioning terminals, thus solving the problem of air conditioning terminal effect analysis. This solution is applicable to the judgment and control of air conditioning effects in various air conditioning usage scenarios.
[0058] Furthermore, based on the above method, this embodiment of the invention also provides an AI-powered intelligent evaluation system for central air conditioning supply effectiveness, comprising: a data acquisition module and an effectiveness evaluation module, wherein,
[0059] The data acquisition module is used to collect real-time indoor air conditioning operation data and target area environmental data. The air conditioning operation data includes: operating level, operating time and set temperature, and the environmental data includes: indoor target area room temperature.
[0060] The effect evaluation module is used to input the preprocessed air conditioning operation data and environmental data into the AI intelligent evaluation model, and use the AI intelligent evaluation model to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using labeled air conditioning effect observation data samples.
[0061] Unless otherwise specifically stated, the relative steps, numerical expressions, and values of the components and steps described in these embodiments do not limit the scope of the invention.
[0062] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0063] The units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations are not considered to be beyond the scope of this invention.
[0064] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in hardware or as a software functional module. This invention is not limited to any particular combination of hardware and software.
[0065] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for intelligently evaluating the supply effect of central air conditioning systems using AI, characterized in that, Include: Real-time collection of indoor air conditioning operation data and target area environmental data. The air conditioning operation data includes: operating level, operating time and set temperature. The environmental data includes: indoor target area room temperature. The preprocessed air conditioning operation data and environmental data are input into the AI intelligent evaluation model. The AI intelligent evaluation model is used to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using labeled air conditioning effect observation data samples.
2. The AI-powered intelligent evaluation method for central air conditioning supply effectiveness according to claim 1, characterized in that, The AI intelligent evaluation model is expressed as: η=a*f1(δT,t)+b*f2(δt2,D,t)+c*f3(δt3,T0,T1,t), where a, b, and c are weighting coefficients, f1(δT,t) is the air conditioner temperature change effect function, f2(δt2,D,t) is the air conditioner speed effect function, f3(δt3,T0,T1,t) is the air conditioner temperature effect function, δT is the change in room temperature per unit time, t is the air conditioner running time, δt2 is the proportion of time the air conditioner speed is activated during air conditioner operation, D is the air conditioner operating speed, and δt3 is the proportion of time during air conditioner operation where the absolute value of the difference between room temperature T0 and the air conditioner set temperature T1 is less than the temperature difference threshold.
3. The AI-powered intelligent evaluation method for central air conditioning supply effectiveness according to claim 1, characterized in that, The evaluation of the air conditioning supply effect in the current indoor area using an AI-powered intelligent assessment model also includes: If multiple indoor areas are equipped with temperature sensors to collect the room temperature of the corresponding target area, the air conditioning temperature change effect function is the weighted arithmetic mean of the air conditioning temperature change effect functions of each area, and the air conditioning temperature effect function is the weighted arithmetic mean of the air conditioning temperature effect functions of each area. If the air conditioner has multiple operating speeds, the air conditioner speed effect function is the ratio of the weighted arithmetic sum of the operating time of each air conditioner speed to the time length t.
4. The AI-powered intelligent evaluation method for central air conditioning supply effectiveness according to claim 1, characterized in that, The AI intelligent evaluation model uses a linear regression model to comprehensively evaluate the air conditioning supply effect using the air conditioning temperature change effect function, the air conditioning speed effect function, and the air conditioning temperature effect function.
5. The AI-powered intelligent evaluation method for central air conditioning supply effectiveness according to claim 4, characterized in that, The training process of the AI intelligent evaluation model includes: Multiple sets of historical observation data on air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data included air conditioning operation data at historical moments and corresponding environmental data. The sample data is preprocessed and divided into training set and test set according to a preset ratio. The preprocessing includes missing value handling and data standardization. The AI intelligent evaluation model is trained using a training set and evaluated using a test set to obtain an AI intelligent evaluation model that meets expectations and can be deployed in the central air conditioning control platform.
6. The AI-powered intelligent evaluation method for central air conditioning supply effect according to claim 1, characterized in that, The AI intelligent evaluation model uses the random forest algorithm to comprehensively evaluate the air conditioning supply effect using the air conditioning temperature change effect function, the air conditioning speed effect function, and the air conditioning temperature effect function.
7. The AI-powered intelligent evaluation method for central air conditioning supply effectiveness according to claim 6, characterized in that, The training process of the AI intelligent evaluation model includes: Multiple sets of historical observation data features of air conditioning operation effects were collected as samples, and the air conditioning operation effects corresponding to each set of observation data were manually labeled. The historical observation data features include the air conditioning operation data features at historical moments and the corresponding environmental data features. The sample data is preprocessed, including missing value handling and data standardization. Define the number of trees, the depth of each tree, and the number of features used for each node in the tree; Training data is extracted with replacement from the preprocessed sample data to construct a training subset for each tree; for each node of a tree, a subset of features is randomly selected from the node's feature set. For each training subset, the best feature is selected for splitting based on the splitting criteria of each feature in the feature subset, forming child nodes. The corresponding decision tree is constructed by recursively generating each child node until the tree depth is satisfied. The specified number of decision trees are generated repeatedly to form a random forest. The final prediction output of the random forest is obtained by averaging the prediction results of each decision tree.
8. A method for intelligently evaluating the supply effect of central air conditioning systems using AI, characterized in that, It includes: a data acquisition module and an effect evaluation module, wherein, The data acquisition module is used to collect real-time indoor air conditioning operation data and target area environmental data. The air conditioning operation data includes: operating level, operating time and set temperature, and the environmental data includes: indoor target area room temperature. The effect evaluation module is used to input the preprocessed air conditioning operation data and environmental data into the AI intelligent evaluation model, and use the AI intelligent evaluation model to evaluate the air conditioning supply effect in the current indoor area. The AI intelligent evaluation model is constructed by a weighted combination of the air conditioning temperature change effect function, the air conditioning speed effect function and the air conditioning temperature effect function, and is trained using labeled air conditioning effect observation data samples.
9. An electronic device, characterized in that, include: At least one processor, and a memory coupled to said at least one processor; The memory stores a computer program that can be executed by the at least one processor to implement the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, enables the implementation of the method as described in any one of claims 1 to 7.