Air conditioner energy demand control method and system

Abstract

The invention discloses an air conditioner energy demand control method and system, and the method comprises the steps: obtaining a start-stop signal, carrying out the initialization processing of a compressor based on the start-stop signal, and obtaining the nominal performance of an indoor unit and the single rotation speed of the compressor; the real-time environment temperature is obtained, and a first indoor unit function demand value is determined based on the real-time environment temperature, indoor unit nominal performance, the compressor single rotating speed and a preset mapping rule; obtaining the running state of the compressor, and obtaining a second indoor unit energy demand value based on a preset energy demand prediction model, the real-time environment temperature and the running state of the compressor; the energy demand prediction model is constructed based on a machine learning algorithm and historical operation data; performing weighted fusion on the first indoor function demand value and the second indoor function demand value to determine a target indoor function demand value; and the target rotating speed is calculated based on the target indoor unit energy demand value, a control instruction is generated based on the target rotating speed, and air conditioner energy demand control is completed based on the control instruction, so that the air conditioner system dynamically responds to the indoor thermal load and accurately adjusts the indoor thermal load.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to a method and system for controlling the functions of an air conditioner. Background Technology

[0002] Currently, with global warming and frequent occurrences of abnormally hot summers, air conditioning has become an indispensable household appliance. An AHU (Air Handling Unit) is an independent air handling device that draws in indoor air, filters, cools or heats, humidifies or dehumidifies it, and then sends it back into the room by a fan. In some areas, it is often presented in a form where the indoor and outdoor units are separate and can be freely combined, to improve indoor air quality and the flexibility of temperature control.

[0003] Due to significant design differences among manufacturers, stable communication between indoor and outdoor units can easily fail, forcing the system to rely on passive monitoring of pipe pressure and localized temperature to determine operating conditions. This control method results in slow response, hindering rapid cooling or heating; it also leads to large temperature fluctuations, high energy consumption, and low system efficiency. Furthermore, the inconsistent control algorithms across brands result in noticeable differences in user experience when switching between different units, making it difficult to balance comfort and energy efficiency. Summary of the Invention

[0004] This invention provides an air conditioner energy consumption control method and system to uniformly control indoor and outdoor units from different manufacturers, thereby reducing air conditioner energy consumption.

[0005] To address the aforementioned technical problems, embodiments of the present invention provide an air conditioning function demand control method, comprising:

[0006] Obtain start / stop signals, perform compressor initialization processing based on the start / stop signals, and obtain the indoor unit's nominal performance and the compressor's single speed.

[0007] The real-time ambient temperature is obtained, and the first indoor unit energy requirement value is determined based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule.

[0008] The compressor operating status is obtained, and a second indoor unit energy demand value is obtained based on a preset energy demand prediction model, real-time ambient temperature, and compressor operating status; the energy demand prediction model is constructed based on machine learning algorithms and historical operating data;

[0009] The target internal function requirement value is determined by weighted fusion of the first internal function requirement value and the second internal function requirement value.

[0010] The target speed is calculated based on the target internal function requirement, and a control command is generated based on the target speed. The air conditioning function requirement control is completed based on the control command.

[0011] This invention constructs a dual-path energy demand estimation mechanism that integrates a physical model and a data-driven model, by comprehensively considering multiple key factors such as start / stop signals, real-time ambient temperature, indoor unit nominal performance, and compressor operating status. This improves the accuracy of energy demand prediction and the adaptability of system operation. A weighted fusion strategy is used to determine the target energy demand value, effectively coordinating differences in multi-source information and enhancing the robustness of the control strategy under complex operating conditions. Finally, the compressor speed is derived from the target energy demand to generate control commands, achieving dynamic response and precise adjustment of the air conditioning system to indoor heat load, thus improving system energy efficiency and user comfort.

[0012] Furthermore, the step of acquiring the real-time ambient temperature and determining the first indoor unit energy requirement based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule includes:

[0013] The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule.

[0014] The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

[0015] This invention introduces three operating condition adjustment factors—capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient—and combines them with the nominal performance parameters of the indoor unit to calculate the first indoor unit energy requirement. This allows the physical model to better reflect actual operating conditions, solving the problem of lag or large errors in the response of traditional models when operating conditions change. The aforementioned parameter correction mechanism offers advantages such as strong interpretability and ease of engineering debugging, improving the stability, transparency, and controllability of the prediction process.

[0016] Furthermore, the mapping rules include a capacity mapping table and a pressure correction table; the step of obtaining the real-time ambient temperature and determining the capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules includes:

[0017] The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value.

[0018] The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table.

[0019] The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

[0020] This invention simplifies the calculation process and improves algorithm efficiency by introducing a fixed-value judgment mechanism to determine the ambient temperature coefficient in different temperature ranges. Based on this, it combines a capacity mapping table and a pressure correction table for differentiated parameter adjustments, accurately reflecting energy efficiency levels under different temperature and pressure conditions, thus enhancing the model's adaptability and operational stability. This parameter setting method supports flexible expansion and fine-tuning under various operating conditions, making it suitable for multiple types of air conditioning systems and different geographical and climatic environments.

[0021] Furthermore, the energy demand prediction model is constructed based on machine learning algorithms and historical operating data, including:

[0022] Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity;

[0023] The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data.

[0024] Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated.

[0025] An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

[0026] This invention improves prediction accuracy under non-ideal operating conditions by introducing a random forest energy demand prediction model built based on historical operating data. This model explores the nonlinear correlation between complex factors such as ambient temperature, low-pressure conditions, speed lag, and start-stop signals and energy demand. By extracting temperature and pressure features, speed lag features, and signal pulsation features, the model's ability to perceive dynamic changes in the system is enhanced, making the prediction results more consistent with actual operating trends. The trained model can be embedded in the controller, supporting real-time inference and assisting control decisions, which helps improve the overall system response speed and energy-saving control performance.

[0027] Furthermore, the step of acquiring the start / stop signal, and performing compressor initialization processing based on the start / stop signal to acquire the indoor unit's nominal performance and compressor single speed, includes:

[0028] Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command.

[0029] Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration.

[0030] This invention detects the start / stop signal status and performs compressor initialization based on the start-up command, ensuring that the system only starts under valid start / stop commands, thus helping to avoid malfunctions or system anomalies. Furthermore, by performing fan self-checks and sensor self-calibrations during the start-up process, potential faults or deviations in the initial state can be automatically identified and corrected, ensuring the accuracy and consistency of parameters used in subsequent energy calculations and control strategies. This improves the operational reliability and lifespan of the air conditioning system, while reducing the frequency of manual intervention and maintenance costs.

[0031] Secondly, the present invention provides an air conditioning function control system, including an indoor unit and an outdoor unit, comprising: a control module, wherein the control module is electrically connected to the indoor unit and the outdoor unit respectively;

[0032] The control module is used to acquire start / stop signals, initialize the compressor based on the start / stop signals, and acquire the indoor unit's nominal performance and compressor single speed; acquire the real-time ambient temperature, and determine a first indoor unit energy demand value based on the real-time ambient temperature, indoor unit nominal performance, compressor single speed, and a preset mapping rule; acquire the compressor's operating status, and acquire a second indoor unit energy demand value based on a preset energy demand prediction model, real-time ambient temperature, and compressor operating status; the energy demand prediction model is constructed based on machine learning algorithms and historical operating data; perform weighted fusion of the first and second indoor unit energy demand values ​​to determine a target indoor unit energy demand value; calculate a target speed based on the target indoor unit energy demand value, generate control commands based on the target speed, and complete the air conditioning unit energy demand control based on the control commands.

[0033] Furthermore, the control module is used to acquire the real-time ambient temperature, and determine the first indoor unit energy requirement value based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule, including:

[0034] The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule.

[0035] The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

[0036] Furthermore, the mapping rules include a capacity mapping table and a pressure correction table; the control module is used to acquire the real-time ambient temperature, and based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules, determine the capacity correction coefficient, the ambient temperature coefficient, and the pressure correction coefficient, including:

[0037] The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value.

[0038] The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table.

[0039] The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

[0040] Furthermore, the energy demand prediction model is constructed based on machine learning algorithms and historical operating data, including:

[0041] Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity;

[0042] The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data.

[0043] Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated.

[0044] An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

[0045] Furthermore, the control module is used to acquire start / stop signals, perform compressor initialization processing based on the start / stop signals, and acquire the indoor unit's nominal performance and the compressor's single speed, including:

[0046] Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command.

[0047] Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating an air conditioning function control method provided in an embodiment of the present invention. Detailed Implementation

[0049] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0050] The terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0051] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0052] Example 1

[0053] See Figure 1 , Figure 1 This is a flowchart illustrating an air conditioner function requirement control method according to an embodiment of the present invention. The embodiment of the present invention provides an air conditioner function requirement control method, including steps 101 to 105, as detailed below:

[0054] Step 101: Obtain start / stop signal, perform compressor initialization processing based on the start / stop signal, and obtain the indoor unit's nominal performance and compressor single speed;

[0055] In this embodiment, the step of acquiring the start / stop signal, and performing compressor initialization processing based on the start / stop signal to acquire the indoor unit's nominal performance and the compressor's single speed, includes:

[0056] Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command.

[0057] Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration.

[0058] In this embodiment, the indoor unit refers to an air conditioning assembly installed inside the user's living or working space, responsible for drawing in indoor air and regulating its temperature and supplying it. The outdoor unit, on the other hand, is a power and heat exchange device located in the outdoor environment. After receiving the start-up command from the indoor unit, the outdoor unit first performs its own fan self-test and sensor self-calibration, and then reads the nominal performance Q obtained from the indoor unit. nBased on the compressor's single-speed capability Cq, and combined with real-time collected ambient temperature and pressure signals, the compressor speed control command is dynamically generated through physical mapping and machine learning prediction models to achieve energy demand regulation and efficient operation of the entire air conditioning system.

[0059] In this embodiment, the indoor unit and outdoor unit are connected via a 24V start / stop signal line to transmit start / stop commands. When the user turns on the air conditioner, the indoor unit outputs a high-level voltage to this signal line. Upon detecting the high level, the outdoor unit controller immediately generates a start command. Upon receiving the start command, the outdoor unit controller first initiates the compressor initialization process: on one hand, it starts the fan self-test subroutine, causing the fan to run at its lowest speed to check if the mechanical condition of the fan impeller and drive motor is normal; on the other hand, it sequentially performs self-calibration on the temperature and pressure sensors, using internal reference signals or zero-point calibration values ​​to eliminate sensor deviations. After initialization, the controller automatically reads the nominal cooling / heating performance Q from the indoor unit's nameplate. n The compressor's single-speed capacity Cq, obtained through pre-calibration or online calibration, is used as the basis for subsequent energy demand calculations and speed control. This process ensures that the system enters the energy demand control stage with a reliable and accurate reference state each time it is started, thereby improving the stability and control accuracy of the entire machine.

[0060] In this embodiment, by detecting the start / stop signal status and performing compressor initialization based on the start-up command, it is ensured that the system starts only under valid start / stop commands, which helps to avoid malfunctions or system anomalies. Furthermore, by performing fan self-checks and sensor self-calibrations during the start-up process, potential faults or deviations in the initial state can be automatically identified and corrected, ensuring the accuracy and consistency of the parameters upon which subsequent energy calculations and control strategies are based. This improves the operational reliability and service life of the air conditioning system, while reducing the frequency of manual intervention and maintenance costs.

[0061] Step 102: Obtain the real-time ambient temperature, and determine the first indoor unit energy requirement value based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and the preset mapping rules;

[0062] In this embodiment, obtaining the real-time ambient temperature and determining the first indoor unit energy requirement based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule includes:

[0063] The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule.

[0064] The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

[0065] In this embodiment, the outdoor ambient temperature e is first collected in real time by an ambient temperature sensor. When e is greater than or equal to a preset high temperature threshold (e.g., 8°C), the ambient temperature coefficient fe is set to 1; otherwise, it is set to 0.4. Subsequently, based on the nominal cooling / heating performance Q indicated on the indoor unit's nameplate... n Based on a pre-established capacity mapping table and the indoor / outdoor unit capacity ratio a (indoor unit nominal capacity / outdoor unit nominal capacity), a capacity correction coefficient fc is determined. Then, the current low-pressure P is compared to a preset pressure target range, and the pressure correction coefficient Pc is increased or decreased by ±1% (cumulative range ±5%) according to a pressure correction table. Finally, based on the indoor unit's nominal performance, capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient, the first indoor unit energy requirement value Q1 is calculated, specifically:

[0066] Q1=Qn*fc*fe*Pc (1)

[0067] In this embodiment, by using item mapping and correction, it is possible to quickly respond to changes in ambient temperature and pressure, while ensuring the interpretability and real-time performance of energy demand calculations, thus providing a reliable initial energy demand reference for subsequent target speed calculations and compressor control.

[0068] In this embodiment, by introducing three operating condition adjustment factors—capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient—and combining them with the nominal performance parameters of the indoor unit to calculate the first indoor unit energy requirement, the physical model can better reflect actual operating conditions, solving the problem of lag or large errors in response to changes in operating conditions in traditional models. The aforementioned parameter correction mechanism offers advantages such as strong interpretability and ease of engineering debugging, improving the stability, transparency, and controllability of the prediction process.

[0069] In this embodiment, the mapping rules include a capacity mapping table and a pressure correction table; the step of obtaining the real-time ambient temperature and determining the capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules includes:

[0070] The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value.

[0071] The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table.

[0072] The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

[0073] Please refer to Table 1 and Table 2. Table 1 is a capacity mapping table provided by an embodiment of the present invention, and Table 2 is a pressure correction table provided by an embodiment of the present invention.

[0074] Table 1

[0075] Indoor unit capacity / Outdoor unit capacity (kW / kW) fc DIP switch a＜0.8 0.8 00 0.8≤a≤1.1 1 01 1.1＜a≤1.3 1.2 10 a＞1.3 1.3 11

[0076] Table 2

[0077] Ambient temperature e Target pressure / bar e＞41℃ No correction 25℃＜e≤41℃ 9～11 15℃＜e≤25℃ 8～10 e≤15℃ 7～9

[0078] In this embodiment, the capacity mapping table shown in Table 1 is used to select the corresponding capacity correction coefficient f according to the ratio a of the nominal capacity of the indoor unit to the nominal capacity of the outdoor unit (i.e., a = indoor unit nominal performance / outdoor unit nominal performance). n , and the specific mapping relationship is: when a < 0.8, take f n = 0.8; when 0.8 ≤ a ≤ 1.1, take f n = 1.0; when 1.1 < a ≤ 1.3, take f n = 1.2; when a > 1.3, take f n = 1.3; and the coefficient setting is completed through the outdoor unit DIP switches (00, 01, 10, 11).

[0079] In this embodiment, the pressure correction table shown in Table 2 is used to determine the target low-pressure pressure range of the system according to the real-time ambient temperature e, so as to dynamically adjust the pressure correction coefficient P: no correction is performed when e > 41°C; when 25°C < e ≤ 41°C, the target pressure is set to 9 - 11 bar; when 15°C < e ≤ 25°C, the target pressure is set to 8 - 10 bar; when e ≤ 15°C, the target pressure is set to 7 - 9 bar.

[0080] In this embodiment, through the capacity mapping table and the pressure correction table, the system can obtain and update the capacity correction coefficient and the pressure correction standard in real time for different unit matching states and different ambient temperature conditions, providing an accurate mapping basis for the energy demand calculation.

[0081] In this embodiment, by introducing a fixed-value judgment mechanism in different temperature ranges to determine the ambient temperature coefficient, the calculation process is simplified and the operation efficiency of the algorithm is improved. On this basis, combined with the capacity mapping table and the pressure correction table for differential parameter adjustment, the energy efficiency level under different temperature and pressure conditions can be accurately reflected, enhancing the adaptability and operation stability of the model. This parameter setting method supports flexible expansion and fine-tuning of working conditions, and is applicable to various types of air-conditioning systems and different geographical climate environments.

[0082] Step 103: Obtain the compressor operating status, and obtain the second internal unit energy demand value based on the preset energy demand prediction model, real-time ambient temperature, and compressor operating status; the energy demand prediction model is constructed based on machine learning algorithms and historical operating data;

[0083] In this embodiment, the energy demand prediction model is constructed based on machine learning algorithms and historical operating data, and includes:

[0084] Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity;

[0085] The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data.

[0086] Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated.

[0087] An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

[0088] In this embodiment, the compressor speed and start / stop signals are collected in real time by the compressor operation status monitoring unit in each control cycle, and combined with the outdoor temperature obtained by the ambient temperature sensor, and sent as input features to the pre-built energy demand prediction model to obtain the second indoor unit energy demand value.

[0089] In this embodiment, the energy demand prediction model is constructed using a random forest algorithm based on historical operating data. First, historical operating data is collected under various operating conditions, including ambient temperature, low pressure, compressor speed, start / stop signals, and the actual cooling / heating capacity at the corresponding time. The raw data is then cleaned to remove outliers and missing values, and the dimensions of each feature are normalized or standardized to obtain standardized data. Subsequently, temperature-pressure interaction terms, speed lag terms, and start / stop signal pulsation features are extracted from the standardized data, and redundant features are removed by feature importance screening to form training samples.

[0090] In this embodiment, the goal is to minimize the mean squared error (MAE) on the training set. Regression models such as gradient boosting regression trees, random forests, or lightweight neural networks are employed, and hyperparameters are optimized through K-fold cross-validation and iterative training is performed. Simultaneously, model performance is validated on an independent test set. Once an error metric such as MAE ≤ 5% is met, the final model is exported in a format suitable for embedded controller deployment. The exported model is then deployed to the controller, where the current ambient temperature and compressor operating status characteristics are read online. The second internal energy demand value, Q2, is inferred and output in real time, and periodically fine-tuned based on actual operating feedback each cycle. This scheme dynamically predicts the system's energy demand through a machine learning model, significantly improving control accuracy and adaptability under varying operating conditions.

[0091] In this embodiment, a random forest energy demand prediction model based on historical operating data is introduced to explore the nonlinear correlation between complex factors such as ambient temperature, low-pressure, speed lag, and start-stop signals and energy demand, thereby improving prediction accuracy under non-ideal operating conditions. By extracting temperature and pressure features, speed lag features, and signal pulsation features, the model's ability to perceive dynamic changes in the system is enhanced, making the prediction results more consistent with actual operating trends. The trained model can be embedded in the controller, supporting real-time inference and assisting control decisions, which is beneficial for improving the overall system response speed and energy-saving control performance.

[0092] Step 104: Perform a weighted fusion of the first internal function requirement value and the second internal function requirement value to determine the target internal function requirement value;

[0093] In this embodiment, the first intrinsic functional requirement value Q1 output by the physical mapping model and the second intrinsic functional requirement value Q2 output by the machine learning prediction model are calculated respectively, and the two are weighted and fused to obtain the final target intrinsic functional requirement value Q. t Specifically, the control module automatically generates fusion weights ω (0≤ω≤1) based on the online confidence level or historical prediction error rate of the machine learning model, where ω can be dynamically adjusted according to the model feedback accuracy; subsequently, the target energy requirement for formaldehyde is calculated based on the first internal energy requirement Q1 and the second internal energy requirement Q2.

[0094] Q t =ω×Q2+(1-ω)×Q1 (2)

[0095] In this embodiment, during the fusion process, when the confidence level of the machine learning model is high or the historical error rate is low, the value of ω is larger to enhance the dominant role of the data-driven model in energy demand estimation under complex operating conditions. Conversely, when the model confidence level decreases, the physical mapping value Q1 is used as the main factor to ensure the stability and reliability of the system. This fusion strategy takes into account both the interpretability of the physical model and the nonlinear fitting capability of the machine learning model, achieving robustness and high accuracy in energy demand estimation, and providing a more accurate target basis for subsequent compressor speed control.

[0096] In this embodiment, the user or maintenance personnel are allowed to manually set the fusion weight ω to 0 through the settings interface. At this time, the target internal function requirement value Q is... t This is the first intrinsic energy requirement value Q1 (i.e., Q) obtained from pure physical mapping calculation. t =0×Q2+1×Q1=Q1). In this operating mode, the system relies entirely on the physical model output determined by the indoor unit's nominal performance, capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient to calculate and generate the compressor's target speed control command, omitting the machine learning prediction model step. This approach ensures that the air conditioning system can still perform energy demand control in a stable and predictable manner even when the model is updated or online learning is unavailable, improving the system's reliability and maintainability in special application scenarios.

[0097] Step 105: Calculate the target speed based on the target internal function requirement value, generate a control command based on the target speed, and complete the air conditioning function requirement control based on the control command.

[0098] In this embodiment, the received fused target internal functional requirement value Q t Then, based on the pre-calibrated compressor single-speed capacity Cq, the compressor target speed n is calculated using the following formula:

[0099]

[0100] Here, 'n' is measured in revolutions per second (rps). The control module then converts the target speed 'n' into a corresponding variable frequency drive command (such as setting the motor driver output frequency or voltage / current amplitude) and sends it to the compressor via the compressor drive unit. Upon receiving the control command, the compressor dynamically adjusts its operating conditions according to its internal closed-loop speed regulation mechanism (e.g., PI or PID control algorithm), rapidly approaching and stabilizing the actual speed at 'nrps'. During operation, the system continuously monitors the actual speed, pipeline pressure, and ambient temperature signals, triggering a new calculation cycle at the end of each 120-second cycle to ensure consistency and real-time performance between the control command and the target energy demand, thereby achieving precise energy demand control and efficient operation of the air conditioning unit.

[0101] In this embodiment, by comprehensively considering multiple key factors such as start / stop signals, real-time ambient temperature, indoor unit nominal performance, and compressor operating status, a dual-path energy demand estimation mechanism integrating a physical model and a data-driven model is constructed. This improves the accuracy of energy demand prediction and the adaptability of system operation. The weighted fusion strategy used to determine the target energy demand effectively coordinates the differences between multi-source information, enhancing the robustness of the control strategy under complex operating conditions. Finally, the compressor speed is derived from the target energy demand to generate control commands, achieving dynamic response and precise adjustment of the air conditioning system to the indoor heat load, thus improving system energy efficiency and user comfort.

[0102] As a specific example of an embodiment of the present invention, a 14kW indoor unit, a 16kW outdoor unit, and ω being 0 are used for illustration. Please refer to Table 3, which is an air conditioner operation status table provided in an embodiment of the present invention.

[0103] Table 3

[0104]

[0105] In this embodiment, after the system starts up, the outdoor unit's performance is 16kW and the indoor unit's nominal performance is Q. n =14kW, the capacity ratio of the two is a=0.875, corresponding to a capacity correction factor fc=1 (DIP switch "01"), and an ambient temperature coefficient fe=1 at an ambient temperature of 35℃. After startup, the system collects low-pressure data every 2 minutes and calculates the pressure correction factor Pc and the internal working capacity requirement Q, then based on Q... n The energy demand is calculated using fc, fe, and Pc, and the target speed n is calculated by combining this with the compressor's single-speed capacity Cq = 0.22 kW / rps. At the 1st minute, the low-pressure area is 12 bar, Pc = 1.01, resulting in Q ≈ 14.14 kW and n ≈ 64 rps. By the 5th minute, the pressure drops to 11.3 bar, Pc increases to 1.03, Q ≈ 14.42 kW, and n ≈ 66 rps. By the 13th minute, the pressure continues to drop to 8.5 bar, Pc falls back to 1.01, Q ≈ 14.14 kW, and n ≈ 64 rps. By the 17th minute, the pressure rises and stabilizes at 9.5 bar, Pc returns to 1, Q ≈ 14 kW, and n ≈ 64 rps. It is evident that by real-time correction of pressure changes and speed adjustment, the system achieves rapid response and steady-state operation under dynamic pressure conditions.

[0106] This invention also provides an air conditioning function control system, including an indoor unit and an outdoor unit, and includes: a control module, wherein the control module is electrically connected to the indoor unit and the outdoor unit respectively;

[0107] The control module is used to acquire start / stop signals, initialize the compressor based on the start / stop signals, and acquire the indoor unit's nominal performance and compressor single speed; acquire the real-time ambient temperature, and determine a first indoor unit energy demand value based on the real-time ambient temperature, indoor unit nominal performance, compressor single speed, and a preset mapping rule; acquire the compressor's operating status, and acquire a second indoor unit energy demand value based on a preset energy demand prediction model, real-time ambient temperature, and compressor operating status; the energy demand prediction model is constructed based on machine learning algorithms and historical operating data; perform weighted fusion of the first and second indoor unit energy demand values ​​to determine a target indoor unit energy demand value; calculate a target speed based on the target indoor unit energy demand value, generate control commands based on the target speed, and complete the air conditioning unit energy demand control based on the control commands.

[0108] In this embodiment, the air conditioning energy demand control system includes an indoor unit, an outdoor unit, and a control module positioned between them. The control module is electrically connected to both the indoor and outdoor units to achieve adaptive energy demand adjustment for the entire system. Specifically, when the user sends a start / stop signal (24V high / low level) through the indoor unit, the control module first detects and generates a start-up or stop-down command. During the start-up phase, it initiates compressor initialization processing—including fan self-test and temperature and pressure sensor self-calibration—to read the nominal performance Q on the indoor unit's nameplate. n The compressor's single-speed capability Cq is obtained through experimental calibration or online calibration; subsequently, the control module acquires the real-time ambient temperature e through an ambient temperature sensor and combines it with Q. n The first internal energy demand value Q1 is calculated using the preset mapping rules formed by the capacity mapping table and pressure correction table, Cq, and Cq. Simultaneously, the control module collects the compressor's operating status (speed, start / stop signals) in real time and inputs it, along with e, into a pre-built energy demand prediction model based on historical operating data and machine learning algorithms (such as random forests) to quickly infer the second internal energy demand value Q2. Then, the control module dynamically sets the fusion weight ω according to the online confidence level or error feedback of the machine learning model, and combines Q1 and Q2 according to Q... t =ω·Q2+(1–ω)·Q1 weighted fusion, to obtain the target internal function requirement value Q t Finally, the control module, based on Q... t Cq is obtained through the formula n = Q t / Cq calculates the target compressor speed n and generates corresponding variable frequency drive commands to send to the compressor, so as to achieve precise energy demand control and efficient operation of the air conditioning unit under various operating conditions.

[0109] In this embodiment, the control module is used to acquire the real-time ambient temperature and determine the first indoor unit energy requirement value based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule, including:

[0110] The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule.

[0111] The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

[0112] In this embodiment, the mapping rules include a capacity mapping table and a pressure correction table; the control module is used to acquire the real-time ambient temperature, and determine the capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules, including:

[0113] The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value.

[0114] The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table.

[0115] The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

[0116] In this embodiment, the energy demand prediction model is constructed based on machine learning algorithms and historical operating data, and includes:

[0117] Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity;

[0118] The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data.

[0119] Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated.

[0120] An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

[0121] In this embodiment, the control module is used to acquire start / stop signals, perform compressor initialization processing based on the start / stop signals, and acquire the indoor unit's nominal performance and the compressor's single speed, including:

[0122] Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command.

[0123] Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration.

[0124] In this embodiment of the invention, a control device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the above-described air conditioning function control method.

[0125] In this embodiment of the invention, a computer-readable storage medium is also provided, which includes a stored computer program, wherein the computer program controls the device where the computer-readable storage medium is located to execute the above-described air conditioning function control method when it is running.

[0126] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. The one or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a control device.

[0127] The control device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The control device may include, but is not limited to, a processor, memory, and display. Those skilled in the art will understand that the above components are merely examples of the control device and do not constitute a limitation on the control device. It may include more or fewer components than the specified components, or a combination of certain components, or different components. For example, the control device may also include input / output devices, network access devices, buses, etc.

[0128] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the control equipment, connecting all parts of the control equipment through various interfaces and lines.

[0129] Memory can be used to store computer programs and / or modules. The processor, by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory, implements various functions of the controlled device. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one application program required for a function (such as sound playback, text conversion, etc.), etc.; the data storage area can store data created based on the use of the mobile phone (such as audio data, text message data, etc.). Furthermore, memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0130] In this invention, modules for processing multi-device access platforms, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Those skilled in the art can understand and implement this invention without any inventive effort.

[0131] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for controlling the function requirements of an air conditioner, characterized in that, include: Obtain start / stop signals, perform compressor initialization processing based on the start / stop signals, and obtain the indoor unit's nominal performance and the compressor's single speed. The real-time ambient temperature is obtained, and the first indoor unit energy requirement value is determined based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule. The compressor operating status is obtained, and a second indoor unit energy demand value is obtained based on a preset energy demand prediction model, real-time ambient temperature, and compressor operating status; the energy demand prediction model is constructed based on machine learning algorithms and historical operating data; The target internal function requirement value is determined by weighted fusion of the first internal function requirement value and the second internal function requirement value. The target speed is calculated based on the target internal function requirement, and a control command is generated based on the target speed. The air conditioning function requirement control is completed based on the control command.

2. The air conditioning function requirement control method as described in claim 1, characterized in that, The step of acquiring the real-time ambient temperature and determining the first indoor unit energy requirement based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule includes: The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule. The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

3. The air conditioning function requirement control method as described in claim 2, characterized in that, The mapping rules include a capacity mapping table and a pressure correction table; the process of obtaining the real-time ambient temperature and determining the capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules includes: The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value. The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table. The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

4. The air conditioning function requirement control method as described in claim 1, characterized in that, The energy demand prediction model is constructed based on machine learning algorithms and historical operating data, and includes: Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity; The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data. Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated. An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

5. A method for controlling air conditioning function requirements as described in any one of claims 1 to 4, characterized in that, The process of acquiring the start / stop signal, initializing the compressor based on the start / stop signal, and acquiring the indoor unit's nominal performance and the compressor's single speed includes: Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command. Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration.

6. An air conditioning function control system, comprising an indoor unit and an outdoor unit, characterized in that, include: The control module is electrically connected to both the indoor unit and the outdoor unit. The control module is used to acquire start / stop signals, perform compressor initialization processing based on the start / stop signals, and acquire the indoor unit's nominal performance and the compressor's single speed. The system acquires the real-time ambient temperature and determines a first indoor unit energy demand value based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule. It also acquires the compressor's operating status and obtains a second indoor unit energy demand value based on a preset energy demand prediction model, the real-time ambient temperature, and the compressor's operating status. The energy demand prediction model is constructed based on machine learning algorithms and historical operating data. A weighted fusion of the first and second indoor unit energy demand values ​​is performed to determine a target indoor unit energy demand value. A target speed is calculated based on the target indoor unit energy demand value, and a control command is generated based on the target speed. Finally, the system completes the air conditioning unit energy demand control based on the control command.

7. The air conditioning function control system as described in claim 6, characterized in that, The control module is used to acquire the real-time ambient temperature and determine the first indoor unit energy requirement based on the real-time ambient temperature, the indoor unit's nominal performance, the compressor's single speed, and a preset mapping rule, including: The system acquires the real-time ambient temperature and determines the capability correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and a preset mapping rule. The first indoor unit energy requirement is determined based on the indoor unit's nominal performance, capacity correction factor, ambient temperature coefficient, and pressure correction factor.

8. The air conditioning function control system as described in claim 7, characterized in that, The mapping rules include a capacity mapping table and a pressure correction table; the control module is used to acquire the real-time ambient temperature, and determine the capacity correction coefficient, ambient temperature coefficient, and pressure correction coefficient based on the real-time ambient temperature, the nominal performance of the indoor unit, and the preset mapping rules, including: The real-time ambient temperature is obtained. When the real-time ambient temperature is greater than or equal to a first preset threshold, the ambient temperature coefficient is set to a first value; otherwise, the ambient temperature coefficient is set to a second value. The pressure correction factor is determined based on the real-time ambient temperature and the pressure correction table. The capability correction coefficient is determined based on the nominal performance of the internal unit and the capability mapping table.

9. An air conditioning function control system as described in claim 6, characterized in that, The energy demand prediction model is constructed based on machine learning algorithms and historical operating data, and includes: Acquire historical operating data, including ambient temperature, low pressure, compressor speed, start / stop signals, and cooling capacity; The ambient temperature, low pressure, compressor speed, and cooling capacity are cleaned to obtain standardized data. Feature extraction is performed on the standardized data to obtain temperature and pressure features, speed lag features, and start / stop signal pulsation features, and training data is generated. An initial energy demand prediction model is constructed based on the random forest algorithm, and the initial energy demand prediction model is trained based on the training data to obtain the energy demand prediction model.

10. An air conditioning function control system as described in any one of claims 6 to 9, characterized in that, The control module is used to acquire start / stop signals, perform compressor initialization processing based on the start / stop signals, and acquire the indoor unit's nominal performance and the compressor's single speed, including: Obtain the start / stop signal; when the start / stop signal is high, generate a power-on command. Based on the power-on command, the compressor is initialized, and the nominal performance of the indoor unit and the compressor's single speed are obtained; the initialization process includes fan self-test and sensor self-calibration.

Images