A method for monitoring fouling failure of an evaporative condenser of a direct expansion air conditioning unit

Abstract

This application relates to the field of air conditioning unit fault diagnosis technology, specifically to a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit. The method includes: acquiring temperature data for each inlet and outlet, and acquiring current data during the operation of the water pump and fan; analyzing the changes in temperature differences between each inlet and outlet, and the degree of difference between different inlet and outlet temperature differences, and combining this with the temporal trend of the temperature differences between each inlet and outlet to obtain characteristic coefficients for the response of each inlet and outlet to changes in heat transfer thermal resistance; adjusting the characteristic coefficients based on the correlation between local temperature changes at each inlet and outlet and local current changes in the water pump and fan, for use in diagnosing scaling faults in the evaporative condenser of the direct expansion air conditioning unit. This application can improve the accuracy of monitoring scaling faults in the evaporative condenser of air conditioning units.

Description

Technical Field

[0001] This application relates to the field of air conditioning unit fault diagnosis technology, specifically to a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit. Background Technology

[0002] As the core heat exchange structure of a direct expansion air conditioning unit, the evaporative condenser cools the refrigerant by absorbing heat through the evaporation of a sprayed water film. The performance of the evaporative condenser directly affects the cooling efficiency, operational stability, and overall energy consumption of the air conditioning system. However, during long-term operation, calcium and magnesium particles and suspended solids in the circulating cooling water can form scale on the surface of the condenser coils. This increases the thermal resistance during heat exchange, leading to higher condensing temperatures, significantly increased compressor power consumption, and reduced cooling capacity. It can even trigger high-pressure protection, frequent shutdowns, and pipe corrosion and leaks, affecting the safe and stable operation of the air conditioning unit.

[0003] However, in practical applications, due to the complex operating environment of evaporative condensers, they are subject to interference from multiple factors such as ambient temperature and humidity, load fluctuations, water quality changes, and air volume drift. Traditional monitoring of scaling fault responses in direct expansion air conditioning units does not integrate multi-dimensional related parameters such as water pump and fan current. This results in poor anti-interference ability and high misjudgment rate in the diagnosis of scaling fault responses during the control and monitoring of air conditioning units. It is also impossible to distinguish the severity of scaling, leading to delayed fault response warnings and untimely responses during the control of air conditioning units, which affects the stable operation of direct expansion air conditioning units. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit, thereby resolving the existing issues.

[0005] The scaling fault monitoring method for the evaporative condenser of a direct expansion air conditioning unit disclosed in this application adopts the following technical solution: One embodiment of this application provides a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit, including the following steps: Acquire temperature data for each inlet and outlet, including spray water inlet and outlet, refrigerant inlet and outlet, and air inlet and outlet, and acquire current data for water pump and fan operation. The changes in temperature difference between each inlet and outlet are analyzed, as well as the degree of difference between the changes in temperature difference between different inlets and outlets. Combined with the time-series change trend of temperature difference between each inlet and outlet, characteristic coefficients of the response of each inlet and outlet to changes in heat transfer resistance are obtained. By analyzing the correlation between the local temperature changes at each inlet and outlet and the local current changes at the water pump and fan, the characteristic coefficients of the response of each inlet and outlet to changes in heat exchange thermal resistance are adjusted. Each time series array is constructed, and the time series array is divided according to the characteristic coefficients of the response of each inlet and outlet to the change in heat exchange thermal resistance after adjustment, so as to diagnose the scaling fault of the evaporative condenser of the direct expansion air conditioning unit.

[0006] Preferably, the difference between the outlet temperature and the inlet temperature of each channel at the same acquisition time is calculated as the temperature difference between the inlet and outlet of each channel, and the temperature differences at all acquisition times are arranged in time sequence to form the temperature difference sequence of the inlet and outlet of each channel.

[0007] Preferably, the distance between the temperature difference sequence of each inlet and outlet and the temperature difference sequence of each other inlet and outlet is calculated as the deviation characteristic coefficient between each inlet and outlet and the other inlet and outlet under the influence of heat transfer resistance change; and the absolute value of the slope of the fitted line corresponding to the temperature difference sequence of each inlet and outlet is used as the trend characteristic coefficient of each inlet and outlet under the influence of heat transfer resistance change.

[0008] Preferably, the method for obtaining the characteristic coefficients of the response of each inlet and outlet to changes in heat transfer thermal resistance is as follows: ,in, For the first Characteristic coefficients of the response of the inlet and outlet of the road to changes in heat exchange thermal resistance; For the first Road and the The deviation characteristic coefficient between the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance; The quantity of imports and exports; and The first Road and the The trend characteristic coefficients of the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance To avoid adjustment coefficients with a denominator of zero.

[0009] Preferably, the temperature change rate of each inlet and outlet is calculated within a preset sliding window, and all the temperature change rates are arranged in chronological order to form a temperature change rate sequence for each inlet and outlet. Correspondingly, for the current data of the water pump and the fan, the current change rate sequence of the water pump and the fan is obtained.

[0010] Preferably, before adjusting the characteristic coefficients, the correlation coefficients between the temperature change rate sequence of each inlet and outlet and the current change rate sequence of the water pump and fan are calculated, and the sum of the absolute values ​​of all correlation coefficients is used as the temperature-flow coordinated response coefficient of each inlet and outlet.

[0011] Preferably, the ratio of the temperature-flow synergistic response coefficient of each inlet and outlet to the sum of the temperature-flow synergistic response coefficients of all inlets and outlets is used as the temperature-flow synergistic weighting coefficient of each inlet and outlet.

[0012] Preferably, adjusting the characteristic coefficients includes: using the product of the characteristic coefficients of each inlet / outlet's response to changes in heat exchange thermal resistance and the temperature-flow synergistic weighting coefficient as the adjusted characteristic coefficients of each inlet / outlet's response to changes in heat exchange thermal resistance.

[0013] Preferably, the temperature data of all inlets and outlets at each acquisition time and the current data of the water pump and fan during operation are used to form time series arrays. The time series arrays of multiple direct expansion air conditioning units with different known scale thicknesses within the preset monitoring period are obtained and used as fault response analysis samples. The adjustment characteristic coefficients of each inlet and outlet corresponding to each fault response analysis sample in response to the change in heat exchange thermal resistance are used to form the coordinates of each fault response analysis sample.

[0014] Preferably, the coordinates of each fault response analysis sample are clustered and the neural network model is trained. The scale coefficient corresponding to the current direct expansion air conditioning unit is output through the trained neural network model. The probability of scale failure in the evaporator condenser of the direct expansion air conditioning unit is positively correlated with the scale coefficient.

[0015] This application has at least the following beneficial effects: To address the shortcomings of traditional methods for diagnosing scaling faults in evaporative condensers of direct expansion air conditioning units, such as weak anti-interference capabilities, high false alarm rates, delayed fault warnings, and untimely responses, this application first analyzes the multi-physics coupling characteristics and time-varying features of the air conditioning unit during operation, focusing on the inlet and outlet temperatures and the currents of water pumps and fans. Then, based on the fault response impact characteristics of heat exchange efficiency decay, increased water circulation resistance, and ventilation load shift during unit operation, it accurately analyzes the sensitivity characteristics of inlet and outlet temperature changes under different scaling degrees. According to the analysis results, data with different fault responses under different monitoring periods are precisely divided, and fault samples with different sensitivity characteristics are balanced to avoid a decrease in the accuracy of condenser scaling fault analysis due to uneven distribution of samples with different sensitivity characteristics. Furthermore, based on this, an LSTM time-series prediction model and an ANN fault diagnosis model are combined to achieve quantitative assessment of scaling degree, advanced prediction of operating trends, and intelligent early warning of fault classification. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the steps of a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit provided in this application. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the scaling fault monitoring method for the evaporative condenser of a direct expansion air conditioning unit provided in this application.

[0021] This application provides a method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit, as illustrated in one embodiment. For details, please refer to [link to specific documentation]. Figure 1 This includes the following steps: Step 1: Obtain temperature data for each inlet and outlet, including spray water inlet and outlet, refrigerant inlet and outlet, and air inlet and outlet, and obtain current data for the water pump and fan during operation.

[0022] During the operation of a direct expansion air conditioning unit, scaling in the evaporative condenser can lead to decreased heat exchange efficiency, increased water circulation resistance, and changes in ventilation load. Single fault response monitoring cannot fully reflect the state changes caused by the fault response. Therefore, in this embodiment, high-precision sensors are deployed at key heat exchange and power nodes of the evaporative condenser in the direct expansion air conditioning unit to complete multi-dimensional operational data acquisition and achieve comprehensive coverage of scaling fault-sensitive parameters. Specifically, in this embodiment, PT100 platinum resistance temperature sensors are installed in the condenser spray circuit, refrigerant circuit, and air circulation circuit to collect data at the spray water inlet and outlet, refrigerant inlet and outlet, and air circulation circuit, respectively. Temperature data from six inlet and outlet channels are used to monitor and analyze minute temperature changes in each channel. Simultaneously, high-precision current sensors are installed in the power supply circuits of the circulating water pump and axial fan to collect operating current data, which is used to analyze load fluctuations caused by changes in water circuit blockage resistance and airflow resistance. All sensors sample synchronously at a fixed frequency, with temperature data sampling at 1Hz and current data sampling at 0.003333Hz. Data is aligned using a unified timestamp. For cases with different sampling frequencies, linear interpolation is used to extend and map the current data to a 1Hz time node sequence for analyzing the real-time operating status of the condenser.

[0023] Furthermore, considering that in practical applications, sensors may be affected by electromagnetic interference, momentary power outages, communication packet loss, water flow pulsations, and other scenarios, leading to significant impacts on the quality of the collected data due to noise interference, this embodiment will preprocess the collected data of various types. Specifically, this embodiment uses a data filtering algorithm to filter each type of collected data. This embodiment employs a moving average filtering algorithm and uses max-min normalization to map the processed data to the [0,1] interval, avoiding the influence of dimensional differences on the calculation and analysis. The specific processes of the moving average filtering algorithm and the max-min normalization method are existing technologies known to those skilled in the art and will not be elaborated upon in this embodiment. In practical applications, implementers may also use other existing data filtering algorithms and data normalization methods.

[0024] This concludes the data collection and processing for the diagnosis of scaling faults in evaporative condensers during the operation and control of direct expansion air conditioning units.

[0025] Step 2: Analyze the changes in temperature difference between each inlet and outlet and the degree of difference between different inlet and outlet temperature differences. Combine this with the time-series change trend of the temperature difference between each inlet and outlet to obtain the characteristic coefficients of the response of each inlet and outlet to changes in heat exchange thermal resistance.

[0026] During the operation of a direct expansion air conditioning unit, calcium and magnesium particles and suspended solids in the circulating cooling water will form scale on the surface of the condenser coils due to long-term use. This scale forms a heat insulation layer on the condenser coil wall, increasing the thermal resistance during heat exchange. Furthermore, as the scale thickness increases, it hinders the stability of water and air flow channels such as pipes, spray nozzles, and filters, affecting heat exchange efficiency and the stability of the air conditioning unit's operation control. Therefore, to accurately diagnose the fault response caused by scale formation in the evaporator condenser of a direct expansion air conditioning unit, this embodiment considers the slow-change fault characteristics caused by scale formation and analyzes the fault response characteristics of the condenser based on the heat exchange change characteristics and the load change characteristics of the water pump and fan during the operation of the air conditioning unit.

[0027] First, considering the significant impact of scaling in the evaporative condenser on heat exchange during the operation of a direct expansion air conditioning unit, the pre-processed temperature data is used to calculate the refrigerant inlet / outlet temperature difference, the spray water inlet / outlet temperature difference, and the air inlet / outlet temperature difference. For example, for the refrigerant inlet / outlet, the difference between the outlet and inlet temperatures at the same sampling time is calculated; this difference represents the refrigerant inlet / outlet temperature difference. Since scaling in pipes has a correlation with heat exchange resistance and heat transfer efficiency during actual operation—for example, thicker scaling results in greater pipe wall heat exchange resistance, less heat carried away by the cooling water, and a smaller inlet / outlet temperature difference that continuously deviates from the normal value—temperature difference data is acquired for each inlet / outlet at all sampling times. All these temperature difference data are then sorted according to their corresponding time sequence, resulting in a temperature difference sequence for each inlet / outlet. This yields the refrigerant inlet / outlet temperature difference sequence, the spray water inlet / outlet temperature difference sequence, and the air inlet / outlet temperature difference sequence, reflecting the heat exchange capacity decay characteristics and the gradual impact of scaling during continuous operation of the evaporative condenser.

[0028] Furthermore, based on the above calculations and analysis, scaling failures may cause disordered changes in the inlet and outlet temperature differences of the refrigerant, spray water, and air, as well as phase shifts and amplitude differences, disrupting the coordinated changes during normal operation. Therefore, for each obtained inlet and outlet temperature difference sequence, the DTW distance between each inlet and outlet temperature difference sequence and the temperature difference sequences of other inlets and outlets is calculated and denoted as the deviation characteristic coefficient under the influence of heat transfer resistance changes. The larger the deviation characteristic coefficient, the worse the operational coordination between the temperatures, the higher the degree of heat transfer imbalance, and the higher the possibility of system anomalies caused by scaling. Furthermore, the least squares method is used to obtain the fitted straight line of the temperature difference sequence of each inlet and outlet, and the slope of the fitted straight line is calculated. The absolute value of the slope is denoted as the trend characteristic coefficient under the influence of heat transfer resistance changes. The larger the trend characteristic coefficient, the faster the rate of heat transfer efficiency decay of the loop may be, and the more significant the development trend of scaling deposition and deterioration.

[0029] Based on the above calculations and analysis, due to the different degrees of heat transfer path and resistance disturbance caused by scaling faults in the evaporative condenser of the air conditioning unit on the refrigerant side, water side, and air side, the temperature change response of different inlets and outlets in the air conditioning unit varies under the influence of changes in heat exchange thermal resistance. Therefore, the deviation characteristic coefficient and trend characteristic coefficient of the three inlets and outlets under the influence of changes in heat exchange thermal resistance are combined to analyze the change response characteristics of each inlet and outlet under the influence of changes in heat exchange thermal resistance. The characteristic coefficient of the sensitivity response of each inlet and outlet to changes in heat exchange thermal resistance is calculated. The larger the characteristic coefficient, the more sensitive the temperature change of that inlet and outlet is to scaling faults, and the higher the accuracy of its temperature change response characteristics for early fault identification and analysis. The specific calculation relationship is as follows: ,in, For the first Characteristic coefficients of the response of the inlet and outlet of the road to changes in heat exchange thermal resistance; For the first Road and the The deviation characteristic coefficient between the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance; The quantity of imports and exports; and The first Road and the The trend characteristic coefficients of the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance This represents an adjustment factor to avoid zero denominators, used to prevent calculation overflow caused by zero denominators. In this embodiment... The value is the minimum non-zero trend characteristic coefficient calculated under the historical fault-free operation of the target unit, so as to avoid obscuring the relative proportional relationship of the heat exchange slopes of each path.

[0030] in, Reflecting the Road entrances and exits compared to the first The relative intensity of heat exchange trend changes at road inlets and outlets. The larger the value, the more likely it is that the process is in the [number]th [stage]. Road and the The first step in the analysis of road entrances and exits The heat exchange attenuation trend of the road is relatively stronger, and the response to scaling failure is more significant.

[0031] Step 3: By analyzing the correlation between the local temperature changes at each inlet and outlet and the local current changes at the water pump and fan, the characteristic coefficients of the response of each inlet and outlet to changes in heat exchange thermal resistance are adjusted.

[0032] Based on the above calculations and analysis, the sensitivity characteristics of each temperature inlet and outlet are analyzed to determine the sensitivity characteristics of each inlet and outlet to the changes in heat exchange resistance caused by scaling faults. However, due to the influence of multiple environmental interference factors during the actual control and operation of the air conditioning unit, the single temperature characteristic may fluctuate, be misjudged, or deviate from the actual fault state. Therefore, it is further considered that scaling faults will directly change the resistance of the water circulation system, making the circulating water pump current characteristics show a regular and identifiable response trend. For example, scaling will block pipes, spray nozzles, and filters, and the water flow resistance will continue to rise as scaling worsens. The water pump load and current are stably positively correlated. At the same time, scaling will narrow the airflow channel and increase the wind resistance. At this time, the heat exchange efficiency will decrease, causing the fan to deviate from the design operating point. The change in wind resistance will cause the fan current to deviate from the steady state. Therefore, combining the coupling characteristics of water pump current, fan current, and temperature sensitivity, an anti-interference weighted correction analysis is performed on the sensitivity characteristics of each inlet and outlet temperature change to scaling faults to improve the stability, accuracy, and early identification capability of scaling fault diagnosis.

[0033] Specifically, since the temperature changes caused by scaling faults during the operation and control of the air conditioning unit are strongly coupled and synchronously linked with the changes in the load of the water pump and fan, for each inlet and outlet, the temperature change rate of the pre-processed temperature data of that inlet and outlet within a preset sliding window is calculated. In this embodiment, the sliding window duration is 10 minutes. The sequence of all the temperature change rates arranged in chronological order is recorded as the temperature change rate sequence of that inlet and outlet. The current change rate of the pre-processed current data of the water pump and fan within the sliding window is also calculated. The rates are sequenced in chronological order and denoted as the current change rate sequences of the water pump and the fan, respectively. Due to the synchronous correlation of heat exchange state, water flow resistance, and ventilation resistance between the inlet and outlet and the water pump and fan under the influence of scaling failure, for example, the increased scaling will cause the temperature change rate to continuously decrease, while the current change rate of the water pump and the fan will increase synchronously and slowly. The three trends have temporal consistency. Therefore, the correlation coefficient between the temperature change rate sequence of each inlet and outlet and the current change rate sequence of the water pump and the fan is calculated, and the sum of the absolute values ​​of all correlation coefficients is taken as the temperature-flow synergistic response coefficient of each inlet and outlet. Preferably, in this embodiment, the Pearson correlation coefficient between the temperature change rate sequence of each inlet and outlet and the current change rate sequence of the water pump, and the Pearson correlation coefficient between the temperature change rate sequence of each inlet and outlet and the current change rate sequence of the fan are calculated respectively. The sum of the absolute values ​​of all the Pearson correlation coefficients is recorded as the temperature-flow synergistic response coefficient of each inlet and outlet. The larger the temperature-flow synergistic response coefficient, the higher the degree of synergistic change between the temperature of the inlet and outlet and the current of the water pump and the fan, and the more significant the temperature change response of the inlet and outlet is to the fault response of the air conditioning unit caused by scaling.

[0034] Furthermore, based on the above calculation and analysis, for each inlet and outlet, the ratio of the temperature-current synergistic response coefficient of each inlet and outlet to the sum of the temperature-current synergistic response coefficients of all inlets and outlets is calculated and denoted as the temperature-current synergistic weight coefficient of each inlet and outlet, reflecting the contribution ratio of the temperature-current correlation change characteristics of that inlet and outlet in the overall fault characteristic system.

[0035] Since the single temperature-flow synergy characteristic is easily affected by factors such as ambient temperature and humidity, load fluctuations, and water quality changes, it may cause misjudgment or missed judgment in early scaling identification. Therefore, anti-interference weighted correction adjustment is performed on the sensitivity of each inlet and outlet temperature change to scaling faults to obtain the characteristic coefficients of each inlet and outlet response to changes in heat exchange thermal resistance after adjustment.

[0036] In this embodiment, the product of the characteristic coefficients of each inlet / outlet's response to changes in heat exchange thermal resistance and the temperature-current synergistic weighting coefficient is used as the adjusted characteristic coefficients of each inlet / outlet's response to changes in heat exchange thermal resistance. The larger the temperature-current synergistic weighting coefficient, the more significant the coordinated change characteristics of temperature and current at the corresponding inlet / outlet, the stronger the response capability to scaling faults, and the less affected by multiple environmental factors. Therefore, by weighting and adjusting the original sensitivity characteristics, a more stable and realistic adjusted characteristic coefficient that better reflects the actual scaling fault response is obtained, effectively improving the accuracy and reliability of fault diagnosis.

[0037] Step 4: Construct each time series array. Divide each time series array according to the characteristic coefficients of the response of each inlet and outlet to the change in heat exchange thermal resistance after adjustment, and then diagnose the scaling fault of the evaporative condenser of the direct expansion air conditioning unit.

[0038] Based on the above calculations and analysis, and combined with the characteristics of the impact of scaling faults on the temperature change response of different circuits during the operation and control of direct expansion air conditioning units, the adjustment characteristic coefficients of the response of each inlet and outlet to the change in heat exchange thermal resistance are obtained. Furthermore, considering that scaling is a gradual fault that accumulates slowly and develops gradually, the time-series characteristic analysis under the scaling fault response is the key to distinguishing scaling from transient interference.

[0039] Because scale formation in the actual operation and control of air conditioning units is a long-term cumulative process, temperature and current do not change abruptly but drift slowly and change monotonically. Environmental fluctuations, load jumps, and sensor interference often manifest as short-term spikes and random fluctuations. Therefore, this embodiment uses a 24-hour monitoring cycle. The temperature data of each inlet and outlet at each acquisition time, as well as the current data of the water pump and fan during operation, are used to form time series arrays. In this embodiment, it is an eight-dimensional time series array, including temperature data of the spray water inlet and outlet, temperature data of the refrigerant inlet and outlet, temperature data of the air inlet and outlet, and current data of the water pump and fan during operation. Based on the above calculations and analysis, the adjustment characteristic coefficients of the response of all inlets and outlets to changes in heat exchange resistance sensitivity in each monitoring cycle are determined. The eight-dimensional time series arrays within each monitoring cycle are used as fault response analysis samples. The adjustment characteristic coefficients of each inlet and outlet corresponding to each fault response analysis sample are used to form the coordinates of each fault response analysis sample. In this embodiment, the coordinates corresponding to each fault response analysis sample are marked as follows: ,in These are the adjustment characteristic coefficients for the refrigerant inlet / outlet, spray water inlet / outlet, and air inlet / outlet, respectively, in response to changes in heat transfer thermal resistance. In this embodiment, they are used as... The coordinates are constructed using three orthogonal axes. In this embodiment, the coordinates are mapped based on the adjustment of the characteristic coefficients, with the aim of highlighting the differences in the distribution of samples under the influence of different scaling fault responses.

[0040] Furthermore, to achieve accurate classification of detection data from different monitoring cycles under scaling fault response, according to the above process in this embodiment, an 8-dimensional time series array of 500 direct expansion air conditioning units with different known scaling thicknesses is obtained within the monitoring cycle. This array is used to construct a fault response analysis sample set with sufficient spatiotemporal distribution characteristics. All fault response analysis samples in the fault response analysis sample set are mapped to a three-dimensional spatial coordinate system according to their corresponding coordinates. The resulting sample set composed of the mapped three-dimensional coordinates is used for subsequent sample partitioning, equalization processing, and fault diagnosis model training.

[0041] Furthermore, based on the mapped sample set obtained above, the density peak clustering algorithm is used to divide the samples. The purpose is to accurately divide the samples with different fault change responses by combining the sensitivity features of the response to scaling faults. Furthermore, considering that if the differences in the responses to scaling faults among different classes of samples are large during the operation and control of the air conditioning unit, it may cause the model training to be biased towards the class with the larger number of samples, resulting in a bias in the fault diagnosis model and a low recognition rate for samples of the minority class with low sensitivity to scaling fault responses. Therefore, in this embodiment, the ADASYN adaptive synthetic sampling method is used to balance the samples of different sensitivity feature classes based on the division results. This eliminates the bias caused by the imbalance in the number of samples affecting the differences in scaling fault responses during the actual operation of the air conditioning unit, and improves the ability of the subsequent fault diagnosis model to identify scaling.

[0042] Based on the characteristics of the responses of various inlet / outlet points and water pumps / fans to scaling faults under different monitoring periods, the 8-dimensional time series arrays of 500 direct expansion air conditioning units with different known scaling thicknesses for each monitoring period were divided and balanced. This achieved equalization of samples with varying sensitivities to scaling fault responses. Each sample was labeled according to its corresponding scaling thickness. Specifically, before calibration, the minimum-maximum normalization method was used to map the scaling thickness values ​​to a distribution within [0,1]. The normalized results within the interval are then used as target labels for network learning. Further, the balanced and labeled sample sets are used to construct training, testing, and validation sets in an 8:1:1 ratio. A fault diagnosis model is trained using a two-layer fully connected ANN neural network structure model, where the loss function is the binary cross-entropy loss function, the optimizer is the Adam optimizer, and the output is the scaling coefficient within the [0,1] interval, used to quantitatively characterize the scaling severity of the evaporative condenser in a direct expansion air conditioning unit. The specific training process of the two-layer fully connected ANN neural network structure model is well-known to those skilled in the art and will not be elaborated further. The probability of scaling failure in the evaporative condenser of a direct expansion air conditioning unit is positively correlated with the scaling coefficient; that is, the higher the scaling coefficient, the higher the probability of scaling failure in the evaporative condenser of the direct expansion air conditioning unit.

[0043] For the current 8-dimensional time series array of the direct expansion air conditioning unit within the current monitoring period, the fouling coefficient of the current direct expansion air conditioning unit is predicted using the aforementioned fault diagnosis model to obtain the corresponding fouling coefficient. Specifically, for the current 8-dimensional time series array of the current direct expansion air conditioning unit within the current monitoring period, the same calculation rules as the training sample extraction are first used to analyze and extract the adjustment characteristic coefficients (i.e., ...) of each inlet and outlet in response to changes in heat transfer thermal resistance sensitivity within the current monitoring period. , , Based on this, the three-dimensional coordinates containing the adjusted characteristic coefficients are fed into the above fault diagnosis model as input, so as to predict the scaling coefficient of the current direct expansion air conditioning unit and obtain the scaling coefficient corresponding to the current direct expansion air conditioning unit.

[0044] Furthermore, to achieve early warning, graded handling, and precise maintenance of scaling faults, this embodiment preferably divides the status into four levels according to the scaling coefficient: a scaling coefficient ≤ 0.2 is normal, requiring continuous monitoring and routine inspections; a scaling coefficient ≤ 0.5 (0.2 < scaling coefficient ≤ 0.5) is a level three warning, requiring enhanced monitoring and preparation for cleaning; a scaling coefficient ≤ 0.8 (0.5 < scaling coefficient ≤ 0.8) is a level two warning, requiring immediate cleaning; and a scaling coefficient ≥ 0.8 is a level one warning, requiring emergency shutdown and forced cleaning. This enables early identification, precise location, and graded handling of scaling, improving the accuracy of response and diagnosis of scaling faults in the evaporative condenser of the direct expansion air conditioning unit.

[0045] It is understood that references to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include the specific features, structures, or characteristics described in connection with that embodiment. Therefore, the appearance of phrases such as "in one embodiment," "in some embodiments," "in other embodiments," or "in still other embodiments" in different parts of this specification does not necessarily refer to the same embodiment, but rather means "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0046] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous. Moreover, the sequence numbers of the steps in the embodiments do not imply a specific order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments in this specification.

[0047] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 of the technical features. Such 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 this application, and should all be included within the protection scope of this application.

Claims

1. A method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit, characterized in that, Includes the following steps: Acquire temperature data for each inlet and outlet, including spray water inlet and outlet, refrigerant inlet and outlet, and air inlet and outlet, and acquire current data for water pump and fan operation. The changes in temperature difference between each inlet and outlet are analyzed, as well as the degree of difference between the changes in temperature difference between different inlets and outlets. Combined with the time-series change trend of temperature difference between each inlet and outlet, characteristic coefficients of the response of each inlet and outlet to changes in heat transfer resistance are obtained. By analyzing the correlation between the local temperature changes at each inlet and outlet and the local current changes at the water pump and fan, the characteristic coefficients of the response of each inlet and outlet to changes in heat exchange thermal resistance are adjusted. Each time series array is constructed, and the time series array is divided according to the characteristic coefficients of the response of each inlet and outlet to the change in heat exchange thermal resistance after adjustment, so as to diagnose the scaling fault of the evaporative condenser of the direct expansion air conditioning unit.

2. The method of claim 1, wherein the method is characterized by: The difference between the outlet temperature and the inlet temperature of each channel at the same acquisition time is calculated as the temperature difference between the inlet and outlet of each channel, and the temperature differences at all acquisition times are arranged in time sequence to form the temperature difference sequence between the inlet and outlet of each channel.

3. The fouling failure monitoring method of an evaporative condenser of a direct expansion air handling unit as set forth in claim 2, wherein, Calculate the distance between the temperature difference sequence of each inlet and outlet and the temperature difference sequence of each other inlet and outlet, and use it as the deviation characteristic coefficient between each inlet and outlet and the other inlet and outlet under the influence of heat transfer resistance change; and use the absolute value of the slope of the fitted line corresponding to the temperature difference sequence of each inlet and outlet as the trend characteristic coefficient of each inlet and outlet under the influence of heat transfer resistance change.

4. The method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit as described in claim 3, characterized in that, The method for obtaining the characteristic coefficients of the response of each inlet and outlet to changes in heat transfer thermal resistance is as follows: ,in, For the first Characteristic coefficients of the response of the inlet and outlet of the road to changes in heat exchange thermal resistance; For the first Road and the The deviation characteristic coefficient between the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance; The quantity of imports and exports; and The first Road and the The trend characteristic coefficients of the inlet and outlet of the road under the influence of changes in heat exchange thermal resistance To avoid adjustment coefficients with a denominator of zero.

5. The method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit as described in claim 1, characterized in that, Calculate the temperature change rate of each inlet and outlet within a preset sliding window, arrange all the temperature change rates in time sequence to form a temperature change rate sequence for each inlet and outlet, and correspondingly, obtain the current change rate sequence for the water pump and fan based on the current data.

6. A method of monitoring fouling of an evaporative condenser of a direct expansion air handling unit as set forth in claim 5, wherein, Before adjusting the characteristic coefficients, calculate the correlation coefficient between the temperature change rate sequence of each inlet and outlet and the current change rate sequence of the water pump and fan, and use the sum of the absolute values ​​of all correlation coefficients as the temperature-flow coordinated response coefficient of each inlet and outlet.

7. A method of monitoring fouling failure of an evaporative condenser of a direct expansion air handling unit as set forth in claim 6, wherein, The ratio of the temperature-flow synergistic response coefficient of each inlet and outlet to the sum of the temperature-flow synergistic response coefficients of all inlets and outlets is used as the temperature-flow synergistic weighting coefficient of each inlet and outlet.

8. The method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit as described in claim 7, characterized in that, Adjusting the characteristic coefficients involves multiplying the characteristic coefficients of each inlet and outlet response to changes in heat exchange thermal resistance with the temperature-flow synergistic weighting coefficient, and using the result as the adjusted characteristic coefficients for each inlet and outlet response to changes in heat exchange thermal resistance.

9. A method of monitoring fouling failure of an evaporative condenser of a direct expansion air handling unit as set forth in claim 8, wherein, The temperature data of all inlets and outlets at each acquisition time, as well as the current data of the water pump and fan during operation, are used to form time series arrays. The time series arrays of multiple direct expansion air conditioning units with different known scale thicknesses within the preset monitoring period are obtained and used as fault response analysis samples. The adjustment characteristic coefficients of each inlet and outlet corresponding to each fault response analysis sample in response to the change in heat exchange thermal resistance are used to form the coordinates of each fault response analysis sample.

10. The method for monitoring scaling faults in the evaporative condenser of a direct expansion air conditioning unit as described in claim 9, characterized in that, The coordinates of each fault response analysis sample are clustered and the neural network model is trained. The scale coefficient corresponding to the current direct expansion air conditioning unit is output through the trained neural network model. The probability of scale failure in the evaporator condenser of the direct expansion air conditioning unit is positively correlated with the scale coefficient.

Images