Air conditioning equipment, fault diagnosis method and device thereof and computer readable storage medium

Abstract

The invention belongs to the technical field of air conditioners, and discloses air conditioning equipment, a fault diagnosis method and device of the air conditioning equipment and a computer readable storage medium. The fault diagnosis method for the air conditioning equipment comprises the steps that a fault-free time sequence prediction model of the air conditioning equipment, operation parameters of the air conditioning equipment at all moments within a first preset duration, corresponding actual state parameters of the air conditioning equipment under the operation parameters and the change trend of the actual state parameters are obtained, and all the moments comprise the current moment; inputting the operation parameters into a fault-free time sequence prediction model to obtain prediction state parameter distribution information of the air conditioning equipment; determining a fault diagnosis threshold range of the air conditioning equipment according to the predicted state parameter distribution information; and performing fault diagnosis on the air conditioning equipment according to the actual state parameter at the current moment, the fault diagnosis threshold range and the change trend. According to the method, dynamic self-adaptive generation of the diagnosis threshold can be realized, the diagnosis accuracy and the working condition robustness are improved, and the early warning capability for early-stage slowly-changing faults is remarkably enhanced.

Description

Technical Field

[0001] This invention relates to the field of air conditioning technology, and in particular to an air conditioning device and its fault diagnosis method and apparatus, and a computationally readable storage medium. Background Technology

[0002] Energy consumption of HVAC equipment constitutes a major component of building operational energy consumption, typically accounting for 30%–40% of total energy consumption in various public and civil buildings. Its operational efficiency directly impacts energy consumption levels and is also related to indoor thermal comfort and the continuity of industrial and commercial production. System malfunctions can lead to anything from reduced energy efficiency and decreased user comfort to serious consequences such as equipment shutdowns and significant economic losses due to work stoppages and production stoppages. Therefore, timely and accurate fault diagnosis of HVAC equipment is crucial for ensuring efficient and stable system operation and enabling predictive maintenance.

[0003] Related technologies have proposed rule-based diagnostic algorithms based on expert knowledge and experience. This method, based on an understanding of the system's thermophysical processes, summarizes the impact of different faults on key characteristic parameters (such as system pressure and superheat), and designs diagnostic rules in the form of "if…then…". In practice, it determines whether a fault has occurred by comparing the system's direct or indirect measured values ​​with preset thresholds. However, this method has significant limitations: the setting of diagnostic rules, especially thresholds, heavily relies on expert experience and extensive testing and verification for specific equipment or operating conditions. It also exhibits poor generalization ability when facing different system forms or equipment types, requiring a large amount of tedious manual parameter tuning. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, the first objective of this invention is to propose a fault diagnosis method for air conditioning equipment, which can achieve dynamic adaptive generation of diagnostic thresholds, significantly improving the accuracy and robustness of diagnosis, and by integrating the dual criteria of temporal change trends and threshold exceedance, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0005] A second objective of this invention is to provide a computer-readable storage medium.

[0006] The third objective of this invention is to provide a fault diagnosis device for air conditioning equipment.

[0007] The fourth objective of this invention is to provide an air conditioning device.

[0008] To achieve the above objectives, a first aspect of the present invention provides a fault diagnosis method for an air conditioning device, wherein the method includes: acquiring a fault-free time-series prediction model of the air conditioning device, operating parameters of the air conditioning device at each moment within a first preset time period, actual state parameters of the air conditioning device corresponding to the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment; inputting the operating parameters into the fault-free time-series prediction model to obtain predicted state parameter distribution information of the air conditioning device; determining a fault diagnosis threshold range of the air conditioning device based on the predicted state parameter distribution information; and performing fault diagnosis on the air conditioning device based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend.

[0009] The fault diagnosis method for air conditioning equipment according to embodiments of the present invention acquires a fault-free time-series prediction model of the air conditioning equipment, operating parameters of the air conditioning equipment at each moment within a first preset time period, actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters. It then inputs the operating parameters into the fault-free time-series prediction model to obtain the distribution information of the predicted state parameters of the air conditioning equipment. Based on the distribution information of the predicted state parameters, it determines the fault diagnosis threshold range of the air conditioning equipment, and then performs fault diagnosis on the air conditioning equipment based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend. This enables dynamic adaptive generation of the diagnostic threshold, significantly improving the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of time-series changing trends and exceeding the threshold, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0010] In addition, the fault diagnosis method for air conditioning equipment according to the above embodiments of the present invention may further include the following additional technical features: According to an embodiment of the present invention, the operating parameters are input into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. The method further includes: when the fault-free timing prediction model performs prediction processing based on the operating parameters, the Monte Carlo random inactivation method is used to inactivate the neurons of the fault-free timing prediction model to reduce the deviation between the predicted state parameter distribution information and the fault-free state parameter distribution information. The fault-free state parameter distribution information is obtained by the fault-free timing prediction model based on the fault-free operating parameters.

[0011] According to one embodiment of the present invention, the predicted state parameter distribution information includes predicted state parameter values ​​and the probability corresponding to each predicted state parameter value. Determining the fault diagnosis threshold range of the air conditioning equipment based on the predicted state parameter distribution information includes: determining the average value and standard deviation of the predicted state parameters based on the predicted state parameter values ​​and the probability corresponding to each predicted state parameter value; and determining the upper limit and lower limit of the fault diagnosis threshold range based on the average value and the standard deviation.

[0012] According to one embodiment of the present invention, fault diagnosis of the air conditioning equipment based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the change trend includes: determining an over-threshold diagnosis result of the actual state parameters based on the actual state parameters at the current moment and the fault diagnosis threshold range; determining a change trend diagnosis result of the actual state parameters based on the change trend; and performing fault diagnosis of the air conditioning equipment based on the over-threshold diagnosis result and the change trend diagnosis result.

[0013] According to one embodiment of the present invention, the over-threshold diagnostic result of the actual state parameter includes a lower limit value that is less than the fault diagnosis threshold range and an upper limit value that is greater than the fault diagnosis threshold range, and the change trend diagnostic result of the actual state parameter includes an upward trend and a downward trend.

[0014] According to one embodiment of the present invention, fault diagnosis of the air conditioning equipment based on the over-threshold diagnosis result and the trend diagnosis result includes: when the over-threshold diagnosis result is less than the lower limit of the fault diagnosis threshold range and the trend diagnosis result is a downward trend, marking the actual state parameter at the current moment with a first fault identifier; when the over-threshold diagnosis result is greater than the upper limit of the fault diagnosis threshold range and the trend diagnosis result is an upward trend, marking the actual state parameter at the current moment with a second fault identifier; and determining the fault type of the air conditioning equipment based on the first fault identifier and / or the second fault identifier of the actual state parameter at the current moment.

[0015] According to one embodiment of the present invention, the fault types include refrigerant leakage fault, abnormal liquid line resistance fault, evaporator scaling fault, and condenser scaling fault.

[0016] According to an embodiment of the present invention, the method further includes: preprocessing the operating parameters and status parameters of the air conditioning equipment, wherein the preprocessing includes at least one of abnormal data removal, operating mode filtering, start-up / shutdown segment division, time-series data slicing, feature value calculation, and normalization processing.

[0017] According to one embodiment of the present invention, the operating parameters include at least one of ambient temperature, cooling water temperature, compressor speed, throttle valve opening, fan speed and water pump speed, and the state parameters include at least one of suction superheat, subcooling, evaporation pressure and condensation pressure.

[0018] To achieve the above objectives, a second aspect of the present invention provides a computer-readable storage medium storing a fault diagnosis program for an air conditioning device, wherein the fault diagnosis program for the air conditioning device, when executed by a processor, implements the fault diagnosis method for the air conditioning device described in the aforementioned embodiments of the present invention.

[0019] According to embodiments of the present invention, a computer-readable storage medium can dynamically and adaptively generate diagnostic thresholds by executing a fault diagnosis program for an air conditioning device through a processor, which greatly improves the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of temporal change trends and exceeding thresholds, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0020] To achieve the above objectives, a third aspect of the present invention provides a fault diagnosis device for an air conditioning unit, wherein the device comprises: an acquisition module, configured to acquire a fault-free time-series prediction model of the air conditioning unit, operating parameters of the air conditioning unit at each moment within a first preset time period, actual state parameters of the air conditioning unit corresponding to the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment; a processing module, configured to input the operating parameters into the fault-free time-series prediction model to obtain predicted state parameter distribution information of the air conditioning unit; a determination module, configured to determine a fault diagnosis threshold range of the air conditioning unit based on the predicted state parameter distribution information; and a diagnosis module, configured to perform fault diagnosis on the air conditioning unit based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend.

[0021] According to an embodiment of the present invention, the fault diagnosis device for air conditioning equipment acquires, through an acquisition module, a fault-free time-series prediction model of the air conditioning equipment, operating parameters of the air conditioning equipment at each moment within a first preset time period, actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters. A processing module inputs the operating parameters into the fault-free time-series prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. A determination module then determines the fault diagnosis threshold range of the air conditioning equipment based on the predicted state parameter distribution information. Finally, a diagnosis module performs fault diagnosis on the air conditioning equipment based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend. This enables dynamic adaptive generation of the diagnostic threshold, significantly improving the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of time-series change trend and exceeding the threshold, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0022] To achieve the above objectives, a fourth aspect of the present invention provides an air conditioning device, including the fault diagnosis device for the air conditioning device described in the foregoing embodiments of the present invention.

[0023] According to the air conditioning equipment of the present invention, by adopting the fault diagnosis device of the air conditioning equipment of the present invention described above, the diagnostic threshold can be dynamically and adaptively generated, which greatly improves the accuracy of diagnosis and the robustness of operating conditions. Furthermore, by integrating the dual criteria of time-series change trend and exceeding the threshold, the early warning capability for early-stage slow-change faults is significantly enhanced.

[0024] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to an embodiment of the present invention; Figure 2 This is a technical schematic diagram of the fault-free timing prediction model according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to another embodiment of the present invention; Figure 4 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to yet another embodiment of the present invention; Figure 5 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to another embodiment of the present invention; Figure 6 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to a specific embodiment of the present invention; Figure 7 This is a block diagram of a fault diagnosis device for air conditioning equipment according to an embodiment of the present invention; Figure 8 This is a block diagram of an air conditioning device according to an embodiment of the present invention. Detailed Implementation

[0026] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0027] The following description, with reference to the accompanying drawings, describes an air conditioning equipment fault diagnosis method, a computer-readable storage medium, an air conditioning equipment fault diagnosis device, and an air conditioning equipment according to embodiments of the present invention.

[0028] Figure 1 This is a flowchart illustrating a fault diagnosis method for an air conditioning device according to an embodiment of the present invention.

[0029] Specifically, in some embodiments of the present invention, such as Figure 1 As shown, the fault diagnosis methods for air conditioning equipment include: S101, acquire the fault-free timing prediction model of the air conditioning equipment, the operating parameters of the air conditioning equipment at each moment within a first preset time period, the actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment.

[0030] Specifically, in this embodiment, the fault-free timing prediction model is a long short-term memory network model pre-trained based on historical fault-free operation data. The first preset duration can be set according to diagnostic sensitivity requirements, such as the most recent 24 hours or a complete operating cycle. Operating parameters are control variables and environmental variables that affect the system state, including at least one of parameters such as ambient temperature, cooling water temperature, compressor speed, throttle valve opening, fan speed, and water pump speed. Actual state parameters are key variables reflecting the thermodynamic state of the refrigeration cycle, including at least one of states such as suction superheat, subcooling, evaporation pressure, and condensation pressure. The trend is obtained by performing a Mann-Kendall nonparametric trend test on the time series of actual state parameters within the first preset duration (such as the most recent 7 days or 30 days), used to determine whether the parameter has a significant upward or downward monotonic trend. The fault-free timing prediction model of the air conditioning equipment can be directly loaded from the local non-volatile memory of the air conditioning equipment. The operating parameters of the air conditioning equipment at each moment within the first preset time period can be obtained from the sensor network or data acquisition and monitoring control system. For directly measurable state parameters (such as evaporation pressure and condensation pressure), they are obtained from the real-time readings of the corresponding pressure sensors. For state parameters that need to be calculated (such as suction superheat and subcooling), they are calculated in real time based on the original sensor readings (such as suction temperature, exhaust temperature, and pressure values) at the current moment and necessary adjacent moments, according to thermodynamic formulas and refrigerant property tables.

[0031] S102, input the operating parameters into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment.

[0032] Specifically, in this embodiment, such as Figure 2As shown, the operating parameters at each time point within the first preset time period T are organized into an n×T matrix (n being the number of operating parameters), which serves as the input to the fault-free time-series prediction model. For example, the operating parameters at time tT, t-T+1, ..., up to time t are used as input to obtain the distribution information of the predicted state parameters at the current time t. During the prediction process, a Monte Carlo random deactivation method is also used to deactivate the neurons of the fault-free time-series prediction model. That is, during each prediction process, the neurons in each hidden layer of the neural network are randomly set to zero according to the probability set during training (e.g., ...). Figure 2 The A1, A2, ..., An in the model reduce the deviation between the predicted state parameter distribution information and the fault-free state parameter distribution information.

[0033] S103, determine the fault diagnosis threshold range of the air conditioning equipment based on the predicted state parameter distribution information.

[0034] Specifically, in this embodiment, the predicted state parameter distribution information includes the predicted state parameter value and the probability corresponding to each predicted state parameter value. The average value and standard deviation of the predicted state parameter are determined based on the predicted state parameter value and the probability corresponding to each predicted state parameter value. The upper limit and lower limit of the fault diagnosis threshold range are determined based on the average value and standard deviation.

[0035] S104, perform fault diagnosis on the air conditioning equipment based on the actual status parameters, fault diagnosis threshold range and change trend at the current moment.

[0036] Specifically, in this embodiment, the diagnostic process consists of two independent judgments and one comprehensive decision. First, the actual state parameter calculated at the current moment (such as the measured intake superheat) is compared with the corresponding dynamic diagnostic threshold range to determine whether it exceeds the boundary, resulting in a "normal," "too high," or "too low" threshold diagnosis result. Second, the diagnostic result (rising, falling, or no trend) is determined based on the obtained trend of the actual state parameter. Third, a comprehensive decision is made according to a preset fault type and fault identifier mapping rule table. For example, when a state parameter's threshold diagnosis result is less than the lower limit of the fault diagnosis threshold range and the trend diagnosis result is a downward trend, a first fault identifier is used to mark the actual state parameter at the current moment. When another state parameter's threshold diagnosis result is greater than the upper limit of the fault diagnosis threshold range and the trend diagnosis result is an upward trend, a second fault identifier is used to mark the actual state parameter at the current moment. After marking all state parameters with fault identifiers, the rule table is consulted based on the fault identifiers of multiple state parameters to determine the specific fault type of the air conditioning equipment.

[0037] Furthermore, in some embodiments of the present invention, the method further includes: obtaining a set of historical fault-free operating parameters of the air conditioning equipment and a set of historical state parameters corresponding to the air conditioning equipment under the set of historical fault-free operating parameters; and training a model based on the set of historical fault-free operating parameters and the set of historical state parameters to obtain a fault-free timing prediction model.

[0038] Specifically, in this embodiment, historical operating data generated by the target air conditioning equipment or its equivalent equipment during a known fault-free phase is obtained, thereby forming a historical fault-free operating parameter set and a historical state parameter set. The historical fault-free operating parameter set and the historical state parameter set are preprocessed, including at least one of abnormal data removal, operating mode filtering, start-up and shutdown segment division, time series data slicing, feature value calculation, and normalization processing.

[0039] A time series prediction model is constructed using a Long Short-Term Memory (LSTM) neural network, such as... Figure 2 As shown, the model's input is a time series segment of operating parameters of length T (historical window size), i.e., [X(tT), X(t-T+1), ..., X(t)], where X(tT) is a vector of n operating parameters at time tT, such as X(1,1), X(2,1), ..., X(n,1). The model's output is the predicted value of one or more key state parameters at the current time t, i.e., [Y(1,t), Y(2,t), ..., Y(n,t)]. The training objective of the model is to minimize the difference between the predicted state parameter vector and the true fault-free value calculated from historical data.

[0040] The preprocessed sample dataset is divided into a training set and a validation set. During the training phase, a Monte Carlo random deactivation technique is introduced, and a Dropout mechanism is added to the LSTM layer to randomly drop some neurons (e.g., ...) with a preset probability. Figure 2 The parameters A1, A2, ..., An in the model name are used to prevent overfitting of the training data and enhance its generalization ability. The random deactivation setting here is the basis for enabling Monte Carlo random deactivation prediction in the subsequent online diagnostic phase. The final model obtained after training is the fault-free timing prediction model. The fault-free timing prediction model encapsulates the dynamic behavior patterns of the air conditioning system under healthy conditions. When a set of operating parameter timings is input, it can output the corresponding predicted values ​​of the state parameters that the system "should have" under fault-free conditions.

[0041] Furthermore, in some embodiments of the present invention, the method further includes inputting operating parameters into a fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. When the fault-free timing prediction model performs prediction processing based on operating parameters, the Monte Carlo random inactivation method is used to inactivate the neurons of the fault-free timing prediction model to reduce the deviation between the predicted state parameter distribution information and the fault-free state parameter distribution information. The fault-free state parameter distribution information is obtained by the fault-free timing prediction model based on the fault-free operating parameters.

[0042] Specifically, in this embodiment, when the fault-free timing prediction model performs prediction processing based on operating parameters, the Monte Carlo random inactivation method is used to inactivate the neurons of the fault-free timing prediction model. That is, during each prediction processing, the neurons in each hidden layer of the neural network are randomly set to zero according to the probability set during training (e.g., ...). Figure 2 The system uses A1, A2, ..., An to reduce the deviation between the predicted state parameter distribution information and the fault-free state parameter distribution information. This allows for a more comprehensive coverage of the possible distribution of fault-free states and effectively improves the robustness of the prediction results, making the generated dynamic thresholds more adaptable to the inherent randomness of the system and measurement noise.

[0043] Furthermore, in some embodiments of the present invention, such as Figure 3 As shown, the predicted state parameter distribution information includes the predicted state parameter values ​​and the probability corresponding to each predicted state parameter value. Based on the predicted state parameter distribution information, the fault diagnosis threshold range for the air conditioning equipment is determined, including: S201, determine the mean and standard deviation of the predicted state parameters based on the predicted state parameter values ​​and the probability corresponding to each predicted state parameter value.

[0044] Specifically, in this embodiment, the probability corresponding to each predicted state parameter value is the calculated weight corresponding to each predicted state parameter value. After obtaining the predicted state parameter values ​​and the probabilities corresponding to each predicted state parameter value, the average value of the predicted state parameters can be determined according to the following formula:

[0045] in, This represents the average value of the predicted state parameters. Indicates the first The probability corresponding to each predicted state parameter value Indicates the first One predicted state parameter.

[0046] The standard deviation of the predicted state parameters can be determined using the following formula:

[0047] in, The standard deviation of the predicted state parameters is represented by... This represents the average value of the predicted state parameters. Indicates the first The probability corresponding to each predicted state parameter value Indicates the first The present invention does not impose specific limitations on the calculation methods for determining the mean and standard deviation of the predicted state parameters.

[0048] S202, determine the upper and lower limits of the fault diagnosis threshold range based on the mean and standard deviation.

[0049] Specifically, in this embodiment, the lower limit of the fault diagnosis threshold range can be the sum of negative 3 times the standard deviation and the mean, and the upper limit of the fault diagnosis threshold range can be the sum of positive 3 times the standard deviation and the mean, as shown in the following formula:

[0050]

[0051]

[0052] in, Let t be the set of actual state parameter values. Let be the set of predicted state parameter values ​​at time t. To predict the average value of the state parameters, The standard deviation is used to predict the state parameters. It should be noted that this invention does not require a specific limit on the multiple of the standard deviation in the formula; for example, it can be taken as plus or minus two times the standard deviation. The specific multiple can also be limited based on expert experience rules.

[0053] Furthermore, in some embodiments of the present invention, such as Figure 4 As shown, fault diagnosis of air conditioning equipment is performed based on the actual state parameters, fault diagnosis threshold range, and changing trends at the current moment, including: S301, determine the over-threshold diagnosis result of the actual state parameters based on the actual state parameters at the current moment and the fault diagnosis threshold range.

[0054] Specifically, in this embodiment, when the actual state parameter at the current moment is less than the lower limit of the fault diagnosis threshold range, it can be determined that the actual state parameter is too small; when the actual state parameter at the current moment is greater than the upper limit of the fault diagnosis threshold range, it can be determined that the actual state parameter is too large; and when the actual state parameter at the current moment is within the fault diagnosis threshold range, it can be determined that the actual state parameter is normal.

[0055] S302, Determine the diagnostic results of the changing trend of actual state parameters based on the changing trend.

[0056] Specifically, in this embodiment, the trend diagnosis result is a longitudinal analysis result based on time series, independent of instantaneous value comparisons. By analyzing the numerical sequence of the state parameter over a relatively long time window, a trend statistic Z is calculated using nonparametric statistical tests (such as the Mann-Kendall test). If the Z value is greater than a positive threshold (e.g., 1.96), it is diagnosed as having a statistically significant "upward trend"; if the Z value is less than a negative threshold (e.g., -1.96), it is diagnosed as having a "downward trend"; otherwise, it is diagnosed as having "no significant trend". This result reflects whether the parameter is undergoing a systematic and directional shift.

[0057] S303, perform fault diagnosis on air conditioning equipment based on the results of over-threshold diagnosis and the results of change trend diagnosis.

[0058] Specifically, in this embodiment, the comprehensive diagnosis is based on a predefined fault type and fault identifier logic matrix (or expert rule table). This matrix specifies the combination of "exceeding threshold diagnostic results" (too high or too low) and "changing trend diagnostic results" (rising or falling) that a set of state parameters should have for each fault type. The system checks whether the actual diagnostic results of all relevant state parameters completely match the description of a certain row (corresponding to a fault type) in the matrix. Only when the fault identifiers of all associated parameters simultaneously meet the corresponding conditions is the fault finally determined to have occurred. This "AND" logic greatly improves the rigor of the diagnosis and reduces false alarms caused by single instantaneous anomalies or short-term fluctuations.

[0059] Furthermore, in some embodiments of the present invention, the threshold diagnosis results of the actual state parameters include a lower limit value that is less than the fault diagnosis threshold range and an upper limit value that is greater than the fault diagnosis threshold range, and the trend diagnosis results of the actual state parameters include an upward trend and a downward trend.

[0060] Furthermore, in some embodiments of the present invention, such as Figure 5 As shown, fault diagnosis of air conditioning equipment is performed based on the results of threshold exceedance diagnosis and the results of trend change diagnosis, including: S401, when the over-threshold diagnosis result is less than the lower limit of the fault diagnosis threshold range and the trend diagnosis result is a downward trend, the first fault identifier is used to mark the actual state parameters at the current moment.

[0061] Specifically, in this embodiment, as shown in Table 1 below, when a certain state parameter (such as evaporation pressure Pe) is simultaneously detected to be "too low" and "showing a downward trend," it is marked with a "↓" symbol (i.e., the first fault symbol). This means that the parameter not only has a low current value but is also in a continuous deterioration process, constituting a strong fault symptom. This symbol will be recorded and associated with the parameter. Furthermore, the present invention does not specifically limit the representation of the first fault symbol.

[0062] Table 1

[0063] S402, when the over-threshold diagnosis result is greater than the upper limit of the fault diagnosis threshold range and the trend diagnosis result is an upward trend, the actual state parameters at the current moment are marked by the second fault identifier.

[0064] Specifically, in this embodiment, as shown in Table 1 above, when both "too high" and "rising trend" are detected for a certain state parameter (such as condensing pressure Pc), it is marked with an "↑" symbol (i.e., a second fault symbol). This means that the state parameter not only has a current value that is too high, but the pressure is also continuously rising, constituting a strong fault symptom in another direction, and this symbol is also recorded. Furthermore, the present invention does not specifically limit the representation of the second fault symbol.

[0065] S403, determine the fault type of the air conditioning equipment based on the first fault identifier and / or the second fault identifier of the actual status parameters at the current moment.

[0066] Specifically, in this embodiment, the system maintains a fault determination lookup table, as shown in Table 1 above. To determine if an air conditioning unit has a "refrigerant leakage fault," the corresponding identifiers for "evaporating pressure Pe," "condensing pressure Pc," "suction superheat SH," and "subcooling SC" must all meet certain conditions, i.e., "evaporating pressure Pe" is marked as "↓," "condensing pressure Pc" is marked as "↓," "suction superheat SH" is marked as "↑," and "subcooling SC" is marked as "↓." The diagnostic engine summarizes all currently marked state parameters and their identifiers and compares them with each row in the lookup table. Only when the identifiers of all actual state parameters meet the conditions is the corresponding fault type conclusion output. This ensures the reliability of the diagnostic conclusion.

[0067] Furthermore, in some embodiments of the present invention, the fault types include refrigerant leakage faults, abnormal liquid line resistance faults, evaporator scaling faults, and condenser scaling faults. Moreover, the present invention may not specifically limit the fault types.

[0068] Furthermore, in some embodiments of the present invention, the method further includes: preprocessing the operating parameters and status parameters of the air conditioning equipment, wherein the preprocessing includes at least one of abnormal data removal, operating mode screening, start-up and shutdown segment division, time series data slicing, feature value calculation and normalization processing.

[0069] Specifically, in this embodiment, to ensure the quality and consistency of the data input to the fault-free timing prediction model and subsequent diagnostic logic, a complete data preprocessing process is constructed, namely, abnormal data removal, operation mode screening, power on / off segment division, timing data slicing, feature value calculation, and normalization processing.

[0070] For outlier removal, the raw data stream reported by the device is first cleaned for data quality. Outliers are identified and removed based on preset physical and data validity rules, including but not limited to: identifying and removing null and missing values; filtering out invalid readings that clearly exceed the device's physical operating range (e.g., specific error codes indicating sensor failure, such as -40℃); and removing outliers caused by momentary communication interruptions or drastic jumps between preceding and following data points. This effectively eliminates interference from noise and erroneous information at the source.

[0071] For operating mode selection, HVAC equipment exhibits fundamental differences in its thermodynamic cycle and control logic under different operating modes (such as cooling, heating, defrosting, and standby). To prevent confusion between dynamic characteristics under different modes, which could lead to difficulties in model learning or misdiagnosis, this method filters data according to operating mode. Specifically, based on the operating mode signals reported by the equipment, continuous data segments under the target diagnostic mode (mainly stable cooling or heating modes) are selected, while data under transient or non-target modes such as defrosting, oil return, and start-stop transitions are removed. This ensures that the data used for modeling and diagnosis is consistent and representative in terms of thermophysical processes.

[0072] For the division of start-up and shutdown segments and the slicing of time-series data, continuous equipment operation data consists of multiple independent start-up and shutdown cycles. First, based on the compressor start-up and shutdown flag signals, the time-series data is divided into several independent operating segments. Each segment represents a complete working cycle from start-up to shutdown, and the data between different segments are not directly related in time. Next, within each independent operating segment, a sliding window of fixed length T (historical window size) is used to slide along the time axis to generate continuous time-series data slices. Each slice is a subsequence of data with T time steps, serving as a sample for model input. This allows the construction of training and prediction samples that meet the requirements of the LSTM model and possess temporal continuity, while preserving the natural boundaries of the physical process.

[0073] For eigenvalue calculations, raw sensor data (such as suction temperature, discharge pressure, and ambient temperature) are transformed into higher-order characteristic parameters with greater diagnostic significance using thermodynamic formulas. Typical calculations include: obtaining the saturated evaporation temperature from a table or by fitting the compressor suction pressure, and then calculating the suction superheat by combining it with the measured suction temperature; deriving the saturated condensation temperature from the condensation pressure, and calculating the subcooling by combining it with the liquid line temperature; in addition, derived indicators such as indoor and outdoor heat exchange temperature difference and energy efficiency ratio (EER) can also be calculated. These characteristic parameters better reflect the intrinsic performance state of the system, improving the model's generalization ability and fault sensitivity.

[0074] For normalization, the operating and state parameters obtained after the above processing have different units and numerical ranges (e.g., pressure in MPa, temperature in °C, and rotational speed in rpm). To eliminate the adverse effects of these dimensional differences on the model's training convergence speed and numerical stability, normalization is employed. Specifically, the Min-Max normalization method is used to linearly scale the value of each feature parameter (variable) to the [0, 1] interval. This ensures that all input features are at similar numerical scales, thereby optimizing model performance.

[0075] Furthermore, in some embodiments of the present invention, the operating parameters include at least one of ambient temperature, cooling water temperature, compressor speed, throttle valve opening, fan speed, and water pump speed, and the state parameters include at least one of suction superheat, subcooling, evaporation pressure, and condensation pressure. Moreover, the present invention may not specifically limit the types of parameters included in the operating parameters and state parameters.

[0076] In summary, in one specific embodiment of the present invention, such as Figure 6 As shown, firstly, the historical fault-free operating parameters and corresponding status parameters of the air conditioning equipment are obtained. Then, abnormal data is removed, operating mode is filtered, start-up and shutdown segments are divided, time series data is sliced, feature values ​​are calculated and normalized. A time series prediction model is constructed using a long short-term memory (LSTM) neural network. The model is trained based on the historical fault-free operating parameter set and the historical status parameter set to obtain the fault-free time series prediction model.

[0077] The system acquires a fault-free timing prediction model for the air conditioning equipment, its operating parameters at each moment within a first preset time period, the actual state parameters of the air conditioning equipment under the operating parameters, and the changing trends of the actual state parameters. It then performs abnormal data removal, operating mode filtering, on / off segment division, timing data slicing, feature value calculation, and normalization on the operating parameters and corresponding state parameters. The processed operating parameters are input into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. Based on the predicted state parameter distribution information, the fault diagnosis threshold range of the air conditioning equipment is determined.

[0078] Based on the processed status parameters and the fault diagnosis threshold range, over-threshold diagnosis is performed. For example, if the actual status parameter at the current moment is less than the lower limit of the fault diagnosis threshold range, it can be determined that the actual status parameter is too small. If the actual status parameter at the current moment is less than the lower limit of the fault diagnosis threshold range, it can be determined that the actual status parameter is too large. If the actual status parameter at the current moment is within the fault diagnosis threshold range, it can be determined that the actual status parameter is normal.

[0079] The actual state parameter change trend diagnosis result is determined based on the processed state parameter parameters and change trends. For example, the Mann-Kendall trend detection algorithm is used in the time series to determine whether the change trend diagnosis result is an upward trend, a downward trend, or no significant trend.

[0080] Furthermore, the results of threshold-exceeding diagnosis and trend-changing diagnosis can be fused to determine the final diagnostic result for the air conditioning equipment. For example, based on a predefined fault type and fault identifier logic matrix, this matrix specifies the combination of "threshold-exceeding diagnosis result" (too high or too low) and "trend-changing diagnosis result" (rising or falling) that a set of state parameters should have for each fault type. The system checks whether the actual diagnostic results of all relevant state parameters completely match the description of a certain row (corresponding to a fault) in the matrix. Only when the fault identifiers of all associated parameters simultaneously meet the corresponding conditions is the fault ultimately determined to have occurred.

[0081] In summary, the fault diagnosis method for air conditioning equipment according to embodiments of the present invention acquires a fault-free time-series prediction model of the air conditioning equipment, operating parameters of the air conditioning equipment at each moment within a first preset time period, actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters. It then inputs the operating parameters into the fault-free time-series prediction model to obtain the distribution information of the predicted state parameters of the air conditioning equipment. Based on the distribution information of the predicted state parameters, it determines the fault diagnosis threshold range of the air conditioning equipment, and then performs fault diagnosis on the air conditioning equipment based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend. Therefore, it can achieve dynamic adaptive generation of the diagnostic threshold, significantly improving the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of time-series changing trends and exceeding the threshold, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0082] Based on the fault diagnosis method for air conditioning equipment proposed in the foregoing embodiments of the present invention, the present invention also proposes a computer-readable storage medium storing a fault diagnosis program for air conditioning equipment thereon. When the fault diagnosis program for air conditioning equipment is executed by a processor, it implements the fault diagnosis method for air conditioning equipment of the above embodiments of the present invention.

[0083] According to embodiments of the present invention, a computer-readable storage medium can dynamically and adaptively generate diagnostic thresholds by executing a fault diagnosis program for an air conditioning device through a processor, which greatly improves the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of temporal change trends and exceeding thresholds, it significantly enhances the early warning capability for early-stage, slowly changing faults.

[0084] Figure 7 This is a block diagram of a fault diagnosis device for an air conditioning unit according to an embodiment of the present invention.

[0085] Specifically, such as Figure 7 As shown, the fault diagnosis device 100 for air conditioning equipment includes an acquisition module 10, a processing module 20, a determination module 30, and a diagnosis module 40.

[0086] The acquisition module 10 is used to acquire the fault-free timing prediction model of the air conditioning equipment, the operating parameters of the air conditioning equipment at each moment within a first preset time period, the actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment; the processing module 20 is used to input the operating parameters into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment; the determination module 30 is used to determine the fault diagnosis threshold range of the air conditioning equipment based on the predicted state parameter distribution information; and the diagnosis module 40 is used to perform fault diagnosis on the air conditioning equipment based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend.

[0087] In some embodiments of the present invention, the processing module 20 is specifically used to inactivate the neurons of the faultless time-series prediction model by using the Monte Carlo random inactivation method when the faultless time-series prediction model performs prediction processing based on the operating parameters, so as to reduce the deviation between the predicted state parameter distribution information and the faultless state parameter distribution information. The faultless state parameter distribution information is obtained by the faultless time-series prediction model based on the faultless operating parameters.

[0088] In some embodiments of the present invention, the predicted state parameter distribution information includes the predicted state parameter value and the probability corresponding to each predicted state parameter value. The determining module 30 is specifically used to determine the average value and standard deviation of the predicted state parameter based on the predicted state parameter value and the probability corresponding to each predicted state parameter value; and to determine the upper limit and lower limit of the fault diagnosis threshold range based on the average value and standard deviation.

[0089] In some embodiments of the present invention, the diagnostic module 40 is specifically used to determine the over-threshold diagnostic result of the actual state parameters based on the actual state parameters at the current moment and the fault diagnosis threshold range; determine the change trend diagnostic result of the actual state parameters based on the change trend; and perform fault diagnosis on the air conditioning equipment based on the over-threshold diagnostic result and the change trend diagnostic result.

[0090] In some embodiments of the present invention, the over-threshold diagnostic results of the actual state parameters include a lower limit value that is less than the fault diagnosis threshold range and an upper limit value that is greater than the fault diagnosis threshold range, and the change trend diagnostic results of the actual state parameters include an upward trend and a downward trend.

[0091] In some embodiments of the present invention, the diagnostic module 40 is further configured to mark the actual state parameters at the current moment using a first fault identifier when the over-threshold diagnostic result is less than the lower limit of the fault diagnosis threshold range and the change trend diagnostic result is a downward trend; and to mark the actual state parameters at the current moment using a second fault identifier when the over-threshold diagnostic result is greater than the upper limit of the fault diagnosis threshold range and the change trend diagnostic result is an upward trend; and to determine the fault type of the air conditioning equipment based on the first fault identifier and / or the second fault identifier of the actual state parameters at the current moment.

[0092] In some embodiments of the present invention, the fault types include refrigerant leakage fault, abnormal liquid line resistance fault, evaporator scaling fault, and condenser scaling fault.

[0093] In some embodiments of the present invention, the processing module 20 is further configured to preprocess the operating parameters and status parameters of the air conditioning equipment. The preprocessing includes at least one of abnormal data removal, operating mode filtering, start-up and shutdown segment division, time series data slicing, feature value calculation and normalization processing.

[0094] It should be noted that other specific implementations of the fault diagnosis device for air conditioning equipment proposed in the embodiments of the present invention can be found in the specific implementations of the fault diagnosis method for air conditioning equipment described in the foregoing embodiments of the present invention. To reduce redundancy, they will not be repeated here.

[0095] In summary, the fault diagnosis device for air conditioning equipment according to embodiments of the present invention acquires, through an acquisition module, a fault-free time-series prediction model of the air conditioning equipment, operating parameters of the air conditioning equipment at each moment within a first preset time period, actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters; and through a processing module, inputs the operating parameters into the fault-free time-series prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. Then, through a determination module, the fault diagnosis threshold range of the air conditioning equipment is determined based on the predicted state parameter distribution information. Finally, through a diagnosis module, the air conditioning equipment is fault-diagnosed based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend. Thus, dynamic adaptive generation of the diagnostic threshold can be achieved, significantly improving the accuracy and robustness of the diagnosis. Furthermore, by integrating the dual criteria of time-series changing trends and exceeding the threshold, the early warning capability for early-stage, slowly changing faults is significantly enhanced.

[0096] Figure 8This is a block diagram of an air conditioning device according to an embodiment of the present invention.

[0097] like Figure 8 As shown, the air conditioning equipment 1000 includes the fault diagnosis device 100 of the air conditioning equipment according to the above embodiment of the present invention.

[0098] According to the air conditioning equipment of the present invention, by adopting the fault diagnosis device of the air conditioning equipment of the present invention described above, the diagnostic threshold can be dynamically and adaptively generated, which greatly improves the accuracy of diagnosis and the robustness of operating conditions. Furthermore, by integrating the dual criteria of time-series change trend and exceeding the threshold, the early warning capability for early-stage slow-change faults is significantly enhanced.

[0099] Furthermore, other components and functions of the air conditioning equipment in the embodiments of the present invention are known to those skilled in the art, and will not be described in detail here to reduce redundancy.

[0100] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0101] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0102] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0103] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0104] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0105] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0106] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0107] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A fault diagnosis method for air conditioning equipment, characterized in that, The method includes: The system acquires a fault-free timing prediction model for the air conditioning equipment, the operating parameters of the air conditioning equipment at each moment within a first preset time period, the actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment. The operating parameters are input into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment; The fault diagnosis threshold range of the air conditioning equipment is determined based on the predicted state parameter distribution information. The air conditioning equipment is diagnosed based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the trend of change.

2. The fault diagnosis method for air conditioning equipment according to claim 1, characterized in that, The method further includes inputting the operating parameters into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment, and inputting the operating parameters into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment. When the fault-free time-series prediction model performs prediction processing based on the operating parameters, the Monte Carlo random inactivation method is used to inactivate the neurons of the fault-free time-series prediction model to reduce the deviation between the predicted state parameter distribution information and the fault-free state parameter distribution information. The fault-free state parameter distribution information is obtained by the fault-free time-series prediction model based on the fault-free operating parameters.

3. The fault diagnosis method for air conditioning equipment according to claim 2, characterized in that, The predicted state parameter distribution information includes predicted state parameter values ​​and the probability corresponding to each predicted state parameter value. Determining the fault diagnosis threshold range for the air conditioning equipment based on the predicted state parameter distribution information includes: The mean and standard deviation of the predicted state parameters are determined based on the predicted state parameter values ​​and the probability corresponding to each predicted state parameter value. The upper and lower limits of the fault diagnosis threshold range are determined based on the average value and the standard deviation.

4. The fault diagnosis method for air conditioning equipment according to claim 3, characterized in that, Fault diagnosis of the air conditioning equipment is performed based on the actual state parameters at the current moment, the fault diagnosis threshold range, and the changing trend, including: The over-threshold diagnosis result of the actual state parameters is determined based on the actual state parameters at the current moment and the fault diagnosis threshold range; Based on the described trend of change, determine the diagnostic result of the actual state parameter's change trend; The air conditioning equipment is diagnosed based on the over-threshold diagnosis results and the change trend diagnosis results.

5. The fault diagnosis method for air conditioning equipment according to claim 4, characterized in that, The threshold diagnosis results of the actual state parameters include lower values ​​that are less than the fault diagnosis threshold range and upper values ​​that are greater than the fault diagnosis threshold range. The trend diagnosis results of the actual state parameters include upward trends and downward trends.

6. The fault diagnosis method for air conditioning equipment according to claim 5, characterized in that, Based on the above-threshold diagnostic results and the trend diagnostic results, fault diagnosis is performed on the air conditioning equipment, including: When the over-threshold diagnosis result is less than the lower limit of the fault diagnosis threshold range and the change trend diagnosis result is a downward trend, the actual state parameter at the current moment is marked by the first fault identifier; When the over-threshold diagnosis result is greater than the upper limit of the fault diagnosis threshold range and the change trend diagnosis result is an upward trend, the actual state parameters at the current moment are marked using a second fault identifier. The fault type of the air conditioning equipment is determined based on the first fault identifier and / or the second fault identifier of the actual status parameters at the current moment.

7. The fault diagnosis method for air conditioning equipment according to claim 6, characterized in that, The fault types include refrigerant leakage faults, abnormal liquid line resistance faults, evaporator scaling faults, and condenser scaling faults.

8. The fault diagnosis method for air conditioning equipment according to any one of claims 1-7, characterized in that, The method further includes: The operating parameters and status parameters of the air conditioning equipment are preprocessed, and the preprocessing includes at least one of the following: abnormal data removal, operating mode screening, start-up and shutdown segment division, time series data slicing, feature value calculation and normalization processing.

9. The fault diagnosis method for air conditioning equipment according to any one of claims 1-7, characterized in that, The operating parameters include at least one of ambient temperature, cooling water temperature, compressor speed, throttle valve opening, fan speed, and water pump speed, and the state parameters include at least one of suction superheat, subcooling, evaporation pressure, and condensation pressure.

10. A computer-readable storage medium, characterized in that, It stores a fault diagnosis program for air conditioning equipment, which, when executed by a processor, implements the fault diagnosis method for air conditioning equipment according to any one of claims 1-9.

11. A fault diagnosis device for air conditioning equipment, characterized in that, The device includes: The acquisition module is used to acquire the fault-free timing prediction model of the air conditioning equipment, the operating parameters of the air conditioning equipment at each moment within a first preset time period, the actual state parameters of the air conditioning equipment under the operating parameters, and the changing trend of the actual state parameters, wherein each moment includes the current moment; The processing module is used to input the operating parameters into the fault-free timing prediction model to obtain the predicted state parameter distribution information of the air conditioning equipment; The determination module is used to determine the fault diagnosis threshold range of the air conditioning equipment based on the predicted state parameter distribution information; The diagnostic module is used to diagnose the air conditioning equipment based on the actual status parameters at the current moment, the fault diagnosis threshold range, and the trend of change.

12. An air conditioning device, characterized in that, It includes the fault diagnosis device for air conditioning equipment as described in claim 11.

Images