Air source heat pump control method and device and electronic equipment

Abstract

The application discloses an air source heat pump control method and device and electronic equipment. The method comprises the following steps: in response to receiving alarm information of a target air source heat pump, obtaining the current running temperature and frosting state of the target air source heat pump; according to the difference between the running temperature and a preset temperature threshold and the frosting state, using a first machine learning model to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information; determining the components with the first correlation greater than a preset correlation threshold as adjustment candidate components and obtaining the current running state thereof; according to the current running state of the adjustment candidate components and the preset correlation of the adjustment candidate components, calculating a first processing scheme for the current alarm information and executing the same, and obtaining the first running data of each component during the execution; according to the first running data and the preset correlation between each component, calculating the current state of the target air source heat pump; according to the current state and the execution time of the first processing scheme, calculating the execution effectiveness of the first processing scheme, and controlling the running of the target air source heat pump according to the execution effectiveness. The application can realize component priority selection and processing scheme closed-loop evaluation under complex working conditions such as multiple component coupling and frosting, improve the accuracy and efficiency of abnormal disposal, and improve the running stability and heating guarantee capacity.

Description

Technical Field

[0001] This application relates to the field of energy technology, and in particular to an air source heat pump control method, apparatus, and electronic equipment. Background Technology

[0002] Air source heat pump control and operation management technology belongs to the field of HVAC and energy station automation control. Air source heat pumps utilize refrigerant compression cycles to absorb heat from outdoor air and supply heat to the user side, and are widely used in scenarios such as domestic hot water, underfloor heating, and air conditioning heating. A typical air source heat pump system usually includes key components such as compressors, evaporators, condensers, and expansion valves, and is equipped with auxiliary components such as fans, four-way valves, liquid receivers, gas-liquid separators, copper pipes, protectors, and electrical control systems. The number of components is large, the coupling relationship is complex, and the operating conditions are significantly affected by changes in ambient temperature, humidity, and load.

[0003] In actual operation, air source heat pumps often experience frost formation and decreased heat exchange efficiency due to low temperature and high humidity, or abnormal parameters such as suction and exhaust pressure, exhaust temperature, and current caused by load fluctuations, which can trigger alarms or protective shutdowns. For commercial or industrial heat pump water heaters and similar applications, common models can be divided into ordinary and low-temperature types. Different models have different applicable environments, resulting in differences in the causes and handling methods for the same alarm phenomenon on different equipment or in different environments, further increasing the difficulty of operation control and maintenance judgment.

[0004] In existing technologies, handling anomalies in air source heat pumps typically relies on alarm notifications, combined with single-point threshold alarms, empirical rules, or manual inspections. For example, exceeding certain temperature / pressure / current limits triggers shutdown, load reduction, or mode switching. Maintenance personnel then rely on experience to determine if frosting has occurred, if the load is abnormal, and which component's control parameters should be adjusted. While this approach is cost-effective, it often depends on fixed thresholds or static rules, making it difficult to adapt to changes in ambient temperature and frosting conditions. It also struggles to be compatible with different equipment models, installation conditions, and operating stages, and is prone to misjudgments or delayed responses.

[0005] Furthermore, since air source heat pumps consist of multiple components working collaboratively, there is often a many-to-many relationship between alarm information and component anomalies: the same alarm may be caused by the linkage deviation of multiple components, or an abnormal state of a certain component may be indirectly caused by changes in other related components. Existing solutions are insufficient in terms of component correlation modeling, candidate component screening, and adaptive generation of handling solutions. They can usually only provide a general "out of limit / fault" prompt, making it difficult to quickly locate the component that should be adjusted first and its adjustment direction. This results in inaccurate handling solutions, long troubleshooting paths, and low efficiency in restoring heating.

[0006] Therefore, there is an urgent need for an air source heat pump control method that, upon receiving an alarm message, can quantitatively evaluate the correlation between multiple components and the alarm message by combining operating characteristics such as operating temperature and frosting status, screen and adjust candidate components and generate a handling plan, and simultaneously use component operating data and component correlations to evaluate the current status and execution effectiveness during the plan execution process, thereby achieving adaptive control. This method aims to improve the timeliness and accuracy of handling heating anomalies, reduce maintenance dependence, and enhance system operational stability. Summary of the Invention

[0007] This application provides an air source heat pump control method, apparatus, and electronic device to address the shortcomings of poor control performance of air source heat pumps in the prior art.

[0008] To achieve the above objectives, embodiments of this application provide an air source heat pump control method, including: In response to receiving an alarm message from the target air source heat pump, the current operating temperature and frosting status of the target air source heat pump are obtained; Based on the difference between the operating temperature and the preset temperature threshold and the frosting status, the first machine learning model is used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information. Components with a first correlation greater than a preset correlation threshold are identified as adjustment candidate components, and the current running status of the adjustment candidate components is obtained; Based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components, calculate the first processing scheme for the current alarm information; Execute the first processing plan, and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing plan; Based on the acquired first operating data and the pre-defined correlation between each component, calculate the current state of the target air source heat pump; Calculate the effectiveness of the first processing solution based on the current state and the execution time of the first processing solution; Control the operation of the target air source heat pump based on the effectiveness of its implementation.

[0009] According to the air source heat pump control method of the present application, the multiple components include at least one of a compressor, an evaporator, a condenser, a throttle valve, a fan, a four-way valve, a liquid receiver, a gas-liquid separator, a copper pipe, a protector, and an electrical control system.

[0010] According to the air source heat pump control method of the present application embodiment, obtaining the current operating state of the adjustment candidate component includes obtaining at least one of temperature, pressure, current, frequency, speed, valve opening degree, flow rate, and start / stop state.

[0011] According to the air source heat pump control method of this application embodiment, controlling the operation of the target air source heat pump based on the effectiveness of execution includes: When the execution effectiveness is lower than the preset effectiveness threshold, the current running status of the adjustment candidate component is obtained, and the current running status of the associated components associated with each adjustment candidate component is obtained according to the preset correlation between each component. Based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing solution and its execution effectiveness, the second processing solution is calculated using the third machine learning model. The second treatment plan is implemented for the target air source heat pump.

[0012] The air source heat pump control method according to embodiments of this application further includes: calculating the energy efficiency and load of the target air source heat pump based on the current state, and The effectiveness of the first processing scheme is calculated based on the current state and the execution time of the first processing scheme, including: calculating the effectiveness based on energy efficiency, load and execution time.

[0013] According to the air source heat pump control method of this application embodiment, calculating the energy efficiency and load of the target air source heat pump based on the current state includes: calculating the real-time COP, energy consumption, and heat supply.

[0014] According to the air source heat pump control method of the present application, the first processing scheme includes performing at least one control operation on the adjustment candidate component, and the control operation includes at least one of load reduction operation, shutdown protection, switching operation mode, triggering alarm and recording fault event.

[0015] This application also provides an air source heat pump control device, including: The acquisition module is used to acquire the current operating temperature and frosting status of the target air source heat pump in response to receiving alarm information from the target air source heat pump. The first correlation calculation module is used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information based on the difference between the operating temperature and the preset temperature threshold and the frosting status using the first machine learning model. The candidate determination module is used to determine the components with a first correlation greater than a preset correlation threshold as adjustment candidate components, and to obtain the current running status of the adjustment candidate components; The first scheme calculation module is used to calculate the first processing scheme for the current alarm information based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components. The execution and acquisition module is used to execute the first processing scheme and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing scheme; The status calculation module is used to calculate the current status of the target air source heat pump based on the first operating data and the preset correlation between the components. The validity calculation module is used to calculate the execution validity of the first processing solution based on the current state and the execution time of the first processing solution; The control module is used to control the operation of the target air source heat pump based on the effectiveness of the execution.

[0016] In the air source heat pump control device according to the embodiments of this application, the control module is further configured to trigger the calculation and execution of a second processing scheme when the execution effectiveness is lower than a preset effectiveness threshold. It also includes a second scheme calculation module, which is used to obtain the current operating status of the adjustment candidate components and the current operating status of the associated components associated with each adjustment candidate component, and calculate the second processing scheme using a third machine learning model based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing scheme and the execution effectiveness, so that the execution and acquisition module executes the second processing scheme on the target air source heat pump.

[0017] This application also provides an electronic device, including: Memory, used to store programs; A processor is configured to run the program stored in the memory, wherein the program executes the air source heat pump control method provided in the embodiments of this application.

[0018] This application also provides a computer-readable storage medium storing a computer program executable by a processor, wherein the program, when executed by the processor, implements the air source heat pump control method provided in this application.

[0019] The air source heat pump control method, apparatus, and electronic device provided in this application obtain the current operating temperature and frosting state of the target air source heat pump. Based on the difference between the operating temperature and a preset temperature threshold, and the frosting state, a first machine learning model is used to calculate the first correlation between multiple components of the heat pump and the current alarm information. Components with a first correlation greater than a preset correlation threshold are identified as adjustment candidate components, and their current operating state is obtained. Further, a first processing scheme for the current alarm information is calculated and executed based on the current operating state of the adjustment candidate components and the preset correlation of the adjustment candidate components. During the execution, first operating data of each component is collected. Subsequently, the current state of the target air source heat pump is calculated based on the first operating data and the preset correlation between each component. The execution effectiveness of the first processing scheme is calculated in combination with the execution time of the first processing scheme. Finally, the target air source heat pump is controlled based on the execution effectiveness. Compared with existing anomaly handling methods that rely on fixed thresholds, empirical rules, or manual investigation, this application introduces "quantitative screening of the correlation between components and alarm information + status assessment based on component correlation + closed-loop evaluation of the effectiveness of the handling solution." This enables more accurate identification of priority adjustment objects and generation of more suitable handling solutions in scenarios with multiple component coupling and operating conditions significantly affected by the environment. It reduces misjudgment and blind adjustment, improves the pertinence and response efficiency of anomaly handling, shortens fault location and recovery time, and enhances the operational stability and heating guarantee capability of air source heat pumps under typical operating conditions such as frosting and load fluctuations.

[0020] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating an embodiment of the air source heat pump control method provided in this application; Figure 2 A schematic diagram of the air source heat pump control device provided in this application; Figure 3 A schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0022] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0023] Example 1 like Figure 1 As shown, Figure 1 This is a schematic flowchart illustrating an air-source heat pump control method according to an embodiment of this application. Figure 1 The method shown may include: S101, in response to receiving alarm information from the target air source heat pump, obtains the current operating temperature and frosting status of the target air source heat pump.

[0024] In step S101, upon receiving alarm information from the target air source heat pump, the current operating temperature and frosting status of the target air source heat pump can be obtained. For example, after receiving alarm information reported by the target air source heat pump, the heat pump main controller, station control system, or edge gateway triggers a closed-loop handling process for that alarm. The alarm information can come from protectors, electrical control systems, drivers, or host computer platforms, and the alarm content may include alarm code, alarm level, occurrence time, trigger source component identifier, etc.

[0025] When obtaining the current operating temperature and frosting status, the operating temperature can be obtained by reading the outdoor heat exchanger side temperature sensor, the refrigerant temperature / pressure conversion value, and the evaporator surface temperature. The frosting status can be determined by the defrosting flag, frosting criterion output (such as low evaporation temperature and decreased heat exchange efficiency), or image / frosting sensor detection results. To ensure the timeliness of subsequent analysis, the operating temperature and frosting status are preferably aligned with the alarm timestamp to form a snapshot of the status at the same moment or within the same time window.

[0026] S102, based on the difference between the operating temperature and the preset temperature threshold and the frosting status, use the first machine learning model to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information.

[0027] In step S102, based on the difference between the operating temperature and the preset temperature threshold, and the frosting status, a first machine learning model can be used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information. The preset temperature threshold is pre-configured and can be set according to the model (normal / low-temperature), operating mode (heating / hot water / defrosting), season, and regional environment. The difference between the operating temperature and the preset temperature threshold can be expressed as a difference, the absolute value of the difference, a segmented interval of the difference, or a normalized deviation; the frosting status can be input in the form of binary values ​​(frost / no frost), multi-level (mild / moderate / severe), or continuous scoring.

[0028] The first machine learning model maps information such as "temperature deviation + frosting status + alarm type" to "correlation score between component and alarm," thereby quantifying the degree of association between each component and the current alarm. The first correlation can be a continuous value from 0 to 1 or multiple levels; a higher score indicates that the component is more likely to trigger the alarm or is the object that needs priority adjustment. The model can be logistic regression, gradient boosting tree, neural network, or a fusion model, and can be deployed on the device side or the cloud side.

[0029] In addition, in the embodiments of this application, there may be at least one of the following components: a compressor, an evaporator, a condenser, a throttle valve, a fan, a four-way valve, a liquid receiver, a gas-liquid separator, copper pipes, a protector, and an electrical control system.

[0030] In this optional implementation, the output dimension of the first machine learning model corresponds one-to-one with the component set, such as outputting "compressor correlation, evaporator correlation, throttle valve correlation, and electronic control system correlation". The system can select a subset of components based on the actual configuration of the field equipment (e.g., models without a four-way valve will not output this item) to avoid invalid judgments on non-existent components and improve the interpretability of correlation ranking.

[0031] S103, the component with the first correlation greater than the preset correlation threshold is identified as the adjustment candidate component, and the current running status of the adjustment candidate component is obtained.

[0032] In step S103, components with a first correlation greater than a preset correlation threshold can be identified as adjustment candidate components, and the current operating status of the adjustment candidate components can be obtained. For example, a preset correlation threshold can be set to filter the candidate set that needs to be prioritized for adjustment from multiple components. The threshold can be a fixed value or can be adaptively adjusted according to the alarm level: for example, a lower threshold is used for high-risk alarms to expand the candidate set, and a higher threshold is used for low-risk alarms to reduce false adjustments. In addition to threshold filtering, the first correlation can also be sorted and Top-K components can be selected as adjustment candidate components to balance coverage and computational overhead.

[0033] When acquiring the current operating status of candidate components for adjustment, key operating parameters of the candidate components can be read from sensors, drivers, controller registers, or real-time platform data and aligned with the time of alarm occurrence. This provides more refined input for subsequent "treatment scheme calculations," such as determining whether it is "overload" or "insufficient heat exchange," or whether it is "abnormal valve opening" or "insufficient airflow."

[0034] Furthermore, embodiments of this application may also: obtain the current operating status of the regulating candidate component, including obtaining at least one of temperature, pressure, current, frequency, speed, valve opening, flow rate, and start / stop status.

[0035] In this optional implementation, different sets of state parameters can be configured for different candidate components: for example, for compressors, the focus is on collecting current, frequency, exhaust temperature, and suction and exhaust pressures; for throttle valves, the focus is on collecting valve opening and superheat-related parameters; for fans, the focus is on collecting speed and estimated air volume; and for electronic control systems, the focus is on collecting fault codes, drive temperature, and power supply voltage. Quality flags (missing, abnormal jumps, out-of-bounds) can be added to the collected data to reduce the weight or eliminate unreliable data during the scheme calculation.

[0036] S104. Calculate the first processing scheme for the current alarm information based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components.

[0037] In step S104, a first processing scheme for the current alarm information can be calculated based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components. For example, a "correlation of adjustment candidate components" can be established in advance to describe the coupling and influence relationships between candidate components and between candidate components and other components, such as "the influence of compressor frequency changes on evaporation pressure and condensation pressure" and "the influence of fan speed changes on evaporation temperature and frosting trend". This correlation can be derived from mechanistic knowledge, historical operating data statistics or calibration results, and stored in the form of a correlation matrix, rule table or graph structure.

[0038] When calculating the first treatment plan, the system comprehensively considers the current operating status, alarm type, operating temperature deviation, frosting status, and component correlation of candidate components to generate a targeted combination of adjustment actions and their parameters (such as adjustment range, execution duration, execution priority, and interlocking conditions). The first treatment plan can be a single-component adjustment (e.g., adjusting only the throttle valve opening) or a multi-component coordinated adjustment (e.g., simultaneously limiting the compressor frequency and increasing the fan speed) to restore heating and stable operation as quickly as possible while ensuring safety.

[0039] In addition, in the embodiments of this application, the first processing scheme may include performing at least one control operation on the adjustment candidate component, the control operation including at least one of load reduction operation, shutdown protection, switching operation mode, triggering alarm and recording fault event.

[0040] In this optional implementation, load reduction operation can be achieved by limiting the compressor frequency / current limit, lowering the target outlet water temperature setting, or reducing the flow rate on the heating side; shutdown protection is used to quickly cut off operation to protect the equipment when the risk is uncontrollable; switching operating modes can include entering defrost, standby, or bypass operation; triggering an alarm can push handling suggestions to the operation and maintenance platform; recording fault events can save alarm context, candidate components and their status, correlation results, and parameters of the proposed execution plan, which is convenient for subsequent traceability and model iteration.

[0041] S105, execute the first processing plan, and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing plan.

[0042] In step S105, a first processing scheme can be executed, and during the execution of the first processing scheme, the first operating data of each component of the target air source heat pump can be acquired. For example, the first processing scheme can be converted into control commands and sent to the corresponding actuators or control loops, such as adjusting the compressor target frequency, adjusting the electronic expansion valve opening, changing the fan speed, triggering the four-way valve switching, or entering the defrosting process. The execution process must meet equipment safety interlocks and control constraints, such as minimum start-up and shutdown intervals, valve switching delays, and compressor protection logic priorities, to avoid secondary faults caused by the control actions themselves.

[0043] During the execution of the first processing scheme, the system collects initial operating data from each component at a preset sampling period, forming "scheme execution process data." This initial operating data may include key parameters such as temperature, pressure, current, frequency, valve opening, speed, flow rate, and start / stop status, and records the scheme start time, current stage, execution parameters, and alarm status changes. This process data is used for subsequent status assessment and effectiveness evaluation, thereby supporting closed-loop adaptive control.

[0044] S106, calculate the current state of the target air source heat pump based on the acquired first operating data and the preset correlation between the components.

[0045] In step S106, the current state of the target air source heat pump can be calculated based on the acquired first operating data and the preset correlations between the components. For example, a "correlation between components" can be established to characterize the coupling relationships and transmission effects of variables within the system, such as the impact of refrigerant pressure changes on the heat exchanger temperature difference, the impact of fan airflow changes on the evaporation temperature and frosting trend, and the impact of load changes on the compressor current. This correlation can be stored in the form of an correlation matrix, a causal graph, a state-space model, or empirical rules.

[0046] When calculating the current state, the system maps the initial operating data into a comprehensive state index or state vector to characterize whether the overall operation of the heat pump is stabilizing, whether it is still abnormal, and the propagation of the abnormality among various components. The current state may include, but is not limited to: heating capacity status, refrigerant circulation status, frosting risk status, and protection risk status. By incorporating correlations into the state calculation, misjudgments caused by relying solely on single-point parameters can be reduced, and the actual operating status after alarm handling can be more accurately reflected at the system level.

[0047] S107, Calculate the execution effectiveness of the first processing plan based on the current state and the execution time of the first processing plan.

[0048] In step S107, the effectiveness of the first processing plan can be calculated based on the current state and the execution time of the first processing plan. For example, the effectiveness can be used to measure whether the first processing plan improves the current alarm and the anomaly it caused. The system can evaluate the effectiveness based on the trend of the current state (e.g., whether key indicators have returned to the normal range, whether the alarm has been cleared, whether the frosting condition has been alleviated, and whether the outlet water temperature / temperature difference on the heating side has recovered) and the execution time: if the state improves significantly within a reasonable time window, the effectiveness is high; if the state does not improve significantly or continues to deteriorate, the effectiveness is low.

[0049] Execution time can include cumulative execution time, phased execution time, or the time required to reach a certain state target. By incorporating both "state outcome" and "time cost" into the effectiveness evaluation, we can avoid one-sided judgments caused by using "whether the alarm has been cleared" as a single indicator, thus providing a more reliable basis for subsequent control decisions.

[0050] Furthermore, embodiments of this application may also include: calculating the energy efficiency and load of the target air source heat pump based on the current state; and calculating the execution effectiveness of the first processing scheme based on the current state and the execution time of the first processing scheme, including: calculating the execution effectiveness based on the energy efficiency, load, and execution time.

[0051] In this optional implementation, the system not only focuses on alarm clearance but also incorporates operational economy into the evaluation: if a solution can clear the alarm but leads to a significant decrease in energy efficiency or a deterioration in load matching, its effectiveness score can be reduced; if a solution restores stability in a short time and maintains a reasonable level of energy efficiency, its effectiveness score can be increased. This allows the control strategy to be upgraded from "being able to operate" to "operating more economically and stably".

[0052] Furthermore, embodiments of this application may also: calculate the energy efficiency and load of the target air source heat pump based on the current state, including: calculating real-time COP, energy consumption, and heating capacity.

[0053] In this optional implementation, real-time COP can be estimated from heating output and input power. Heating output can be calculated based on the supply and return water temperature difference and flow rate, while energy consumption can be obtained based on power metering or current and voltage estimation. The system can calculate average COP and cumulative energy consumption within a sliding time window to reduce the impact of instantaneous fluctuations and link with load indicators (such as heating output per unit time or heat load rate) to build a more robust effectiveness evaluation.

[0054] S108, based on the effectiveness of execution, control the operation of the target air source heat pump.

[0055] In step S108, the operation of the target air source heat pump can be controlled based on the effectiveness of execution. For example, a final control decision can be output based on the effectiveness of execution: when the effectiveness of execution is high, the first treatment plan is maintained or gradually retreated to the normal control curve; when the effectiveness of execution is low, an escalation treatment is triggered, such as expanding the adjustment range, adjusting control parameters, or entering a stronger intervention mode such as protection / defrosting. This closed-loop mechanism allows the treatment process to converge adaptively, reducing manual intervention.

[0056] Specific methods for controlling the operation of the target air source heat pump may include: updating controller settings, adjusting operating modes, updating limiting parameters, locking / unlocking certain protection strategies, and reporting the results to the operation and maintenance platform to create traceable records. Through "effectiveness-driven control," the system can avoid continuously wasting time and energy on ineffective solutions, thereby improving anomaly recovery efficiency and operational stability.

[0057] Furthermore, embodiments of this application may also include: controlling the operation of the target air source heat pump based on the effectiveness of execution, including: when the effectiveness of execution is lower than a preset effectiveness threshold, obtaining the current operating status of the adjustment candidate components and, based on the preset correlation between the components, obtaining the current operating status of the associated components associated with each adjustment candidate component; calculating a second processing scheme using a third machine learning model based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing scheme and the effectiveness of execution; and executing the second processing scheme for the target air source heat pump.

[0058] In this optional implementation, when the system determines that the first processing solution is ineffective, it automatically expands the observation and adjustment scope based on component correlation: on the one hand, it reacquires the state of the candidate adjustment component, and on the other hand, it simultaneously acquires the state of strongly correlated components, and calculates the state difference of these correlated components between the "alarm trigger time" and the "current time" to identify possible root cause migration or coupling mismatch. Subsequently, the third machine learning model integrates the difference features, the content of the first processing solution and its effectiveness to generate a more targeted second processing solution (e.g., upgrading from "only adjusting the throttle valve" to "compressor frequency limiting + fan speed increase + entering defrosting" or "triggering protection shutdown and reporting"), which is then executed by the system, thereby achieving multi-round iterative adaptive handling and improving the success rate and safety in complex abnormal scenarios.

[0059] The air source heat pump control method provided in this application obtains the current operating temperature and frosting status of the target air source heat pump. Based on the difference between the operating temperature and a preset temperature threshold, and the frosting status, a first machine learning model is used to calculate the first correlation between multiple components of the heat pump and the current alarm information. Components with a first correlation greater than a preset correlation threshold are identified as adjustment candidate components, and their current operating status is obtained. Further, a first processing scheme for the current alarm information is calculated and executed based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components. During the execution, first operating data of each component is collected. Subsequently, the current status of the target air source heat pump is calculated based on the first operating data and the preset correlation between each component. The execution effectiveness of the first processing scheme is calculated in combination with the execution time of the first processing scheme. Finally, the target air source heat pump is controlled based on the execution effectiveness. Compared with existing anomaly handling methods that rely on fixed thresholds, empirical rules, or manual investigation, this application introduces "quantitative screening of the correlation between components and alarm information + status assessment based on component correlation + closed-loop evaluation of the effectiveness of the handling solution." This enables more accurate identification of priority adjustment objects and generation of more suitable handling solutions in scenarios with multiple component coupling and operating conditions significantly affected by the environment. It reduces misjudgment and blind adjustment, improves the pertinence and response efficiency of anomaly handling, shortens fault location and recovery time, and enhances the operational stability and heating guarantee capability of air source heat pumps under typical operating conditions such as frosting and load fluctuations.

[0060] Example 2 Figure 2 This is a schematic diagram of the air source heat pump control device provided in this application. This air source heat pump control device can be used to implement, for example, [reference needed]. Figure 1The air source heat pump control method described in this application embodiment may include: an acquisition module 21, a first correlation calculation module 22, a candidate determination module 23, a first scheme calculation module 24, an execution and acquisition module 25, a state calculation module 26, an effectiveness calculation module 27, a control module 28, and a second scheme calculation module 29.

[0061] The acquisition module 21 can be used to acquire the current operating temperature and frosting status of the target air source heat pump in response to receiving alarm information from the target air source heat pump.

[0062] The acquisition module 21 can be used to connect the alarm channel and the operation data channel on the device side. Upon receiving alarm information (e.g., alarm code, alarm level, occurrence time, trigger source information) from the target air source heat pump, the acquisition module 21 reads the operating temperature and frosting status corresponding to the alarm time point, forming an input feature set for subsequent correlation calculations. The operating temperature can be obtained from the outdoor / refrigerant side temperature sensor or by pressure-temperature conversion; the frosting status can be obtained from the defrost flag, frosting judgment logic output, or other frosting detection signals. To improve consistency, the acquisition module 21 preferably performs timestamp alignment and validity verification on the data (e.g., missing, out-of-bounds, and abnormal jump markers).

[0063] The first correlation calculation module 22 can be used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information based on the difference between the operating temperature and the preset temperature threshold and the frosting state, using a first machine learning model.

[0064] The first correlation calculation module 22 can receive the operating temperature, frosting status, and alarm information output by the acquisition module 21, and calculate temperature difference characteristics (such as difference, normalization deviation, or segmentation encoding) in combination with a preset temperature threshold. Subsequently, module 22 calls the first machine learning model to output the first correlation score or level between the target air source heat pump and the current alarm information for multiple components, which is used to characterize the degree to which "this component needs to be prioritized for adjustment / this component may be related to the cause of the alarm". The first machine learning model can be deployed on an edge controller or cloud platform, and module 22 is responsible for completing the model input splicing, feature standardization, and model inference result encapsulation.

[0065] Furthermore, in this embodiment, the first correlation calculation module 22 can be configured to calculate the first correlation for at least one of the compressor, evaporator, condenser, throttle valve, fan, four-way valve, liquid receiver, gas-liquid separator, copper pipe, protector and electrical control system.

[0066] In this optional implementation, the output dimension of module 22 matches the actual set of components configured in the device; when a certain model does not have a specific component, the corresponding relevant output may not be generated or the item may be marked as invalid, thereby avoiding meaningless candidate screening and improving the interpretability and stability of the relevant ranking.

[0067] The candidate determination module 23 can be used to determine the components with a first correlation greater than a preset correlation threshold as adjustment candidate components, and obtain the current running status of the adjustment candidate components.

[0068] The candidate determination module 23 can perform threshold judgment on the output of the first correlation calculation module 22: adding components with a first correlation greater than a preset correlation threshold to the set of adjustment candidate components. The preset correlation threshold can be configured according to alarm level, device model, or operating mode, or a "threshold + Top-K" combination strategy can be used to control the candidate size. After the candidate set is determined, module 23 reads the current operating status of the candidate components and outputs the candidate component identifier, correlation score, and operating status together to the first scheme calculation module 24.

[0069] In addition, in this embodiment of the application, the candidate determination module 23 may also be configured to obtain the current operating status of the adjustment candidate component, the current operating status including at least one of temperature, pressure, current, frequency, speed, valve opening, flow rate and start / stop status.

[0070] In this optional implementation, module 23 can adopt a differentiated data acquisition strategy based on the candidate component type: for example, prioritizing the acquisition of current, frequency, and exhaust temperature for compressors, prioritizing the acquisition of valve opening and superheat-related quantities for throttle valves, and prioritizing the acquisition of speed and estimated air volume for fans. Module 23 can also label the acquired data with quality tags (missing / abnormal / reliable) to facilitate subsequent scheme calculations using weight reduction or rollback strategies.

[0071] The first scheme calculation module 24 can be used to calculate the first processing scheme for the current alarm information based on the current operating status of the adjustment candidate component and the preset correlation of the adjustment candidate component.

[0072] The first scheme calculation module 24 can calculate a first processing scheme for the current alarm information based on the candidate component status output by the candidate determination module 23 and in combination with preset adjustment candidate component correlations (e.g., component coupling relationships stored in the form of correlation matrices, rule tables, or correlation graphs). The first processing scheme may include control action type, action parameters (amplitude / limit / target value), action sequence, execution duration, interlock conditions, and exit conditions, so that the device can issue executable control commands to the heat pump control system.

[0073] Furthermore, in this embodiment of the application, the first processing scheme generated by the first scheme calculation module 24 may maliciously include performing at least one control operation on the adjustment candidate component, and the control operation includes at least one of load reduction operation, shutdown protection, switching operation mode, triggering alarm and recording fault event.

[0074] In this optional implementation, load reduction operation can correspond to compressor frequency / current limiting, reducing the target outlet water temperature setting, or limiting load output; shutdown protection can correspond to safe shutdown and protection lockout; switching operating modes can correspond to entering defrost, standby, or other preset modes; triggering alarms can synchronize handling suggestions to the operation and maintenance platform; recording fault events can write alarm context, candidate components, correlations, scheme parameters, and execution results to local or cloud logs for traceability and model iteration.

[0075] The execution and acquisition module 25 can be used to execute the first processing scheme and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing scheme.

[0076] The execution and acquisition module 25 can parse the first processing scheme into control commands that the device can recognize, and execute corresponding control actions according to the scheme requirements (such as adjusting the compressor frequency, adjusting the throttle valve opening, changing the fan speed, switching the trigger mode, etc.). At the same time, during the execution of the scheme, it collects the first operating data of each component of the target air source heat pump at a preset sampling period. The first operating data may include key temperatures, pressures, currents, frequencies, speeds, valve openings, flow rates, start / stop statuses, and alarm status changes, and is marked with timestamps and execution stage markers to support subsequent status assessments and effectiveness calculations.

[0077] The status calculation module 26 can be used to calculate the current status of the target air source heat pump based on the first operating data and the preset correlation between the components.

[0078] The state calculation module 26 can receive the first operating data from the execution and acquisition module 25 and, in conjunction with the preset correlations between various components (such as component coupling models, causal relationship diagrams, or empirical association rules), calculate the current state of the target air source heat pump. The current state can be represented as a state vector or a set of state indicators, used to reflect whether the system is stabilizing, whether alarms are alleviated, whether the risk of frosting has decreased, and whether heating capacity has recovered. Compared with judgments based solely on single-point thresholds, module 26, by introducing the correlations between components, can more comprehensively characterize the "abnormal propagation and regression caused by multi-component coupling," thereby providing a more reliable state basis for effectiveness evaluation.

[0079] The validity calculation module 27 can be used to calculate the execution validity of the first processing scheme based on the current state and the execution time of the first processing scheme.

[0080] The effectiveness calculation module 27 can calculate the effectiveness of the first processing solution based on the current state output by the state calculation module 26 and the execution time of the first processing solution. The execution time can be the cumulative duration, the duration in stages, or the time taken to reach the preset state target; the effectiveness of the execution can be output as a continuous score or a grade result, which is used to reflect "the degree of improvement of the anomaly within a given time cost". Module 27 can be implemented using a preset evaluation function, weighted index fusion, or an evaluation model trained based on historical data, so as to balance safety, stability, and response speed.

[0081] In addition, in this embodiment, the effectiveness calculation module 27 can also be configured to calculate the energy efficiency and load of the target air source heat pump based on the current state, and calculate the execution effectiveness based on the energy efficiency, load and execution time.

[0082] In this optional implementation, module 27 introduces economic constraints while evaluating "whether it has returned to normal": when a certain solution clears the alarm but causes a significant reduction in energy efficiency or a deterioration in load matching, the effectiveness score can be reduced; when the solution recovers to stability in a short time and the energy efficiency remains within a reasonable range, the effectiveness score can be increased, thereby avoiding one-sided control that "only pursues clearing the alarm".

[0083] Furthermore, in this embodiment, the effectiveness calculation module 27 can be configured to calculate real-time COP, energy consumption, and heat supply based on the current state to obtain energy efficiency and load.

[0084] In this optional implementation, the heat supply can be estimated based on the temperature difference and flow rate of the supply and return water, the energy consumption can be estimated based on electrical power metering or electrical parameters, and the real-time COP is calculated based on the heat supply and input power. Module 27 can perform sliding time window averaging or filtering on the above quantities to reduce the impact of instantaneous fluctuations on the effectiveness evaluation.

[0085] The control module 28 can be used to control the operation of the target air source heat pump based on the effectiveness of the execution.

[0086] For example, in this embodiment of the application, the control module 28 can also be used to trigger the calculation and execution of the second processing scheme when the execution validity is lower than a preset validity threshold.

[0087] The control module 28 receives the execution validity output from the validity calculation module 27 and controls the operation of the target air source heat pump accordingly: when the execution validity meets the requirements, the first treatment plan can be maintained or gradually reverted to the conventional control strategy; when the execution validity is low, the escalation treatment path is entered. Control actions may include updating setpoints, adjusting limit parameters, switching operating modes, triggering protection strategies, or reporting the treatment results.

[0088] When the execution validity is lower than the preset validity threshold, the control module 28 triggers the second scheme calculation module 29 to generate a second processing scheme, and coordinates the execution and acquisition module 25 to execute the second processing scheme to form an "validity-driven closed-loop iterative handling" to improve the recovery success rate and security under complex abnormal conditions.

[0089] The second scheme calculation module 29 can be used to obtain the current operating status of the adjustment candidate components and the current operating status of the associated components associated with each adjustment candidate component. Based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing scheme and the execution effectiveness, the second processing scheme is calculated using a third machine learning model so that the execution and acquisition module executes the second processing scheme on the target air source heat pump.

[0090] The second scheme calculation module 29 executes after being triggered by the control module 28: First, it acquires the current operating status of the candidate adjustment component, and then, based on the preset correlation between components, it acquires the current operating status of the associated components related to the candidate adjustment component; subsequently, it calculates the difference characteristics (e.g., difference, rate of change, deviation level) between the "current operating status of the associated component" and the "operating status of the associated component when receiving alarm information," and uses these difference characteristics, the content of the first processing scheme, and the execution effectiveness as inputs to call the third machine learning model to output the second processing scheme. The second processing scheme can be a combination of stronger intervention or more coordinated control actions, such as expanding the adjustment object, changing the adjustment order, switching the operating mode, or entering the protection strategy, and is executed by the execution and acquisition module 25, thereby realizing automatic correction and escalation processing of ineffective handling.

[0091] Example 3 The above describes the internal functions and structure of the air source heat pump control device, which can be implemented as an electronic device. Figure 3 A schematic diagram illustrating the structure of an embodiment of the electronic device provided in this application. (See attached diagram.) Figure 3 As shown, the electronic device includes a memory 31 and a processor 32.

[0092] Memory 31 is used to store programs. In addition to the programs described above, memory 31 can also be configured to store various other data to support operation on the electronic device. Examples of this data include instructions for any application or method used to operate on the electronic device, contact data, phonebook data, messages, pictures, videos, etc.

[0093] The memory 31 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0094] Processor 32 is not limited to a processor (CPU), but may also be a graphics processing unit (GPU), a field-programmable gate array (FPGA), an embedded neural network processor (NPU), or an artificial intelligence (AI) chip. Processor 32 is coupled to memory 31 and executes the program stored in memory 31 to perform the air source heat pump control method of Embodiment 1 described above.

[0095] Furthermore, such as Figure 3 As shown, the electronic device may also include other components such as a communication component 33, a power supply component 34, an audio component 35, and a display 36. Figure 3 The diagram only shows some components and does not mean that the electronic device includes only these components. Figure 3 The components shown.

[0096] Communication component 33 is configured to facilitate wired or wireless communication between electronic devices and other devices. The electronic devices can access wireless networks based on communication standards, such as WiFi, 3G, 4G, or 5G, or combinations thereof. In one exemplary embodiment, communication component 33 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 33 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0097] Power supply component 34 provides power to various components of the electronic device. Power supply component 34 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device.

[0098] Audio component 35 is configured to output and / or input audio signals. For example, audio component 35 includes a microphone (MIC) configured to receive external audio signals when the electronic device is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 31 or transmitted via communication component 33. In some embodiments, audio component 35 also includes a speaker for outputting audio signals.

[0099] Display 36 includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation.

[0100] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling an air source heat pump, characterized in that, include: In response to receiving an alarm message from the target air source heat pump, the current operating temperature and frosting status of the target air source heat pump are obtained; Based on the difference between the operating temperature and the preset temperature threshold and the frosting status, the first machine learning model is used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information. Components with a first correlation greater than a preset correlation threshold are identified as adjustment candidate components, and the current running status of the adjustment candidate components is obtained; Based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components, calculate the first processing scheme for the current alarm information; Execute the first processing plan, and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing plan; Based on the acquired first operating data and the pre-defined correlation between each component, calculate the current state of the target air source heat pump; Calculate the effectiveness of the first processing solution based on the current state and the execution time of the first processing solution; Control the operation of the target air source heat pump based on the effectiveness of its implementation.

2. The method according to claim 1, characterized in that, Multiple components include at least one of a compressor, evaporator, condenser, expansion valve, fan, four-way valve, liquid receiver, gas-liquid separator, copper tubing, protector, and electrical control system.

3. The method according to claim 1, characterized in that, Obtaining the current operating status of the candidate regulating components includes obtaining at least one of the following: temperature, pressure, current, frequency, speed, valve opening, flow rate, and start / stop status.

4. The method according to claim 1, characterized in that, The operation of the air source heat pump is controlled according to the effectiveness of its implementation, including: When the execution effectiveness is lower than the preset effectiveness threshold, the current running status of the adjustment candidate component is obtained, and the current running status of the associated components associated with each adjustment candidate component is obtained according to the preset correlation between each component. Based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing solution and its execution effectiveness, the second processing solution is calculated using the third machine learning model. The second treatment plan is implemented for the target air source heat pump.

5. The method according to claim 1, characterized in that, Also includes: Calculate the energy efficiency and load of the target air source heat pump based on the current conditions; Furthermore, the execution effectiveness of the first processing scheme is calculated based on the current state and the execution time of the first processing scheme, including: calculating the execution effectiveness based on energy efficiency, load, and execution time.

6. The method according to claim 5, characterized in that, Calculating the energy efficiency and load of the target air source heat pump based on the current state includes: calculating real-time COP, energy consumption, and heating output.

7. The method according to claim 1, characterized in that, The first processing scheme includes performing at least one control operation on the candidate component to be regulated, the control operation including at least one of load reduction operation, shutdown protection, switching operating mode, triggering alarm and recording fault event.

8. An air source heat pump control device, characterized in that, include: The acquisition module is used to acquire the current operating temperature and frosting status of the target air source heat pump in response to receiving alarm information from the target air source heat pump. The first correlation calculation module is used to calculate the first correlation between multiple components in the target air source heat pump and the current alarm information based on the difference between the operating temperature and the preset temperature threshold and the frosting status using the first machine learning model. The candidate determination module is used to determine the components with a first correlation greater than a preset correlation threshold as adjustment candidate components, and to obtain the current running status of the adjustment candidate components; The first scheme calculation module is used to calculate the first processing scheme for the current alarm information based on the current operating status of the adjustment candidate components and the preset correlation of the adjustment candidate components. The execution and acquisition module is used to execute the first processing scheme and acquire the first operating data of each component of the target air source heat pump during the execution of the first processing scheme; The status calculation module is used to calculate the current status of the target air source heat pump based on the first operating data and the preset correlation between the components. The validity calculation module is used to calculate the execution validity of the first processing solution based on the current state and the execution time of the first processing solution; The control module is used to control the operation of the target air source heat pump based on the effectiveness of the execution.

9. The apparatus according to claim 8, characterized in that, The control module is also used to trigger the calculation and execution of a second processing scheme when the execution validity is lower than a preset validity threshold. The device further includes a second scheme calculation module, which is used to obtain the current operating status of the adjustment candidate components and the current operating status of the associated components associated with each adjustment candidate component, and calculate the second processing scheme using a third machine learning model based on the difference between the current operating status of each associated component and the operating status of each associated component when the alarm information is received, as well as the first processing scheme and the execution effectiveness, so that the execution and acquisition module executes the second processing scheme on the target air source heat pump.

10. An electronic device, characterized in that, include: Memory, used to store programs; A processor for running the program stored in the memory, wherein the program executes the air source heat pump control method as described in any one of claims 1 to 7.

Images