Method for detecting abnormality of refrigeration coefficient and related device

Abstract

The application discloses a refrigeration coefficient anomaly detection method and related equipment. The refrigeration coefficient anomaly detection method comprises the following steps: determining a first refrigeration coefficient of a refrigeration system; determining a refrigeration mode and a time length during which the refrigeration system is in the refrigeration mode; if the time length is greater than or equal to a target time length, inputting a plurality of first refrigeration coefficients into a variable point detection model to obtain a variable point index of the refrigeration system; determining a second refrigeration coefficient corresponding to each variable point index; if it is detected according to the second refrigeration coefficient that the refrigeration system is abnormal, obtaining a plurality of target point position data corresponding to each refrigeration device in the refrigeration system; inputting the plurality of target point position data into a preset anomaly detection model to obtain a plurality of anomaly point position data; selecting, from the plurality of anomaly point position data, anomaly point position data with an importance greater than a preset importance; and determining an anomaly cause of the refrigeration system based on the selected anomaly point position data. The above method can improve the accuracy and speed of refrigeration coefficient anomaly detection.

Description

Technical Field

[0001] This application belongs to the field of testing technology, and in particular relates to a method and related equipment for detecting abnormal coefficient of performance (COP). Background Technology

[0002] The coefficient of performance (COP) is the amount of cooling output per unit of power consumed. It is also known as the coefficient of performance and is an important indicator for evaluating whether a refrigeration system is operating normally and efficiently.

[0003] In related technologies, the coefficient of performance (COP) is usually compared with a set threshold to determine if there is an anomaly. However, since anomalies in refrigeration systems generally persist for a considerable period, and the COP itself fluctuates, using a threshold to determine the presence of anomalies often results in multiple instances of both abnormal and normal performance within a given timeframe, leading to low accuracy in COP anomaly detection. Furthermore, refrigeration systems have multiple operating modes, and setting corresponding thresholds for each mode is inefficient, further contributing to low efficiency in COP anomaly detection. Summary of the Invention

[0004] This application provides a method and related equipment for detecting abnormal coefficient of performance (COP) to solve the problems of low accuracy and speed in detecting COP.

[0005] The first aspect of this application provides a method for detecting abnormal coefficient of performance (COP) in a refrigeration system, applied to a control device. The control device and multiple refrigeration devices form a refrigeration system. The method includes: determining and collecting a first COP corresponding to the refrigeration system at preset time intervals; determining the refrigeration mode corresponding to the refrigeration system and the duration the refrigeration system is in the refrigeration mode; if the duration is greater than or equal to a target duration, inputting multiple first COPs into a preset variable point detection model to obtain multiple variable point indices corresponding to the refrigeration system; determining a second COP corresponding to each variable point index; if an abnormality is detected in the refrigeration system based on the second COP, acquiring multiple target point data corresponding to each refrigeration device in the refrigeration system; inputting the multiple target point data into a preset abnormality detection model to obtain multiple abnormal point data; selecting abnormal point data with an importance greater than a preset importance from the multiple abnormal point data, and determining the cause of the abnormality in the refrigeration system based on the selected abnormal point data.

[0006] Furthermore, in the above-mentioned method for detecting abnormal coefficient of performance (COP) provided in this application embodiment, the step of determining and collecting the first COP corresponding to the refrigeration system at each preset time interval includes: selecting the first total point data and the second total point data corresponding to the refrigeration system at each preset time interval; determining and collecting the first COP based on the first total point data and the second total point data; and combining multiple first COPs in the order of the time axis to obtain the multiple first COPs corresponding to the refrigeration system.

[0007] Furthermore, in the above-mentioned method for detecting abnormal coefficient of performance (COP) provided in the embodiments of this application, determining the second COP corresponding to each variable point index includes: determining the time axis corresponding to each variable point index; determining the average of multiple first COPs corresponding to the time axis to obtain the second COP.

[0008] Furthermore, in the above-mentioned method for detecting abnormal coefficient of performance (COP) provided in the embodiments of this application, the method further includes: selecting the maximum value from a plurality of second COPs as the target second COP; determining the target time axis corresponding to the target second COP; if the target time axis is located on a preset time axis and the target second COP is greater than a preset COP threshold, then determining that the refrigeration system has an abnormality.

[0009] Furthermore, in the above-described method for detecting abnormal coefficient of performance (COP) provided in this application embodiment, the step of acquiring multiple target point data corresponding to each refrigeration device in the refrigeration system includes: collecting multiple initial point data corresponding to each refrigeration device in the refrigeration system according to a preset frequency; if there are missing point data in the multiple initial point data, filling in the missing point data according to a preset interpolation method to obtain multiple first intermediate point data corresponding to each refrigeration device; consolidating the multiple first intermediate point data at preset time intervals to obtain multiple second intermediate point data; and selecting the multiple target point data from the multiple second intermediate point data according to the target time axis.

[0010] Furthermore, in the above-described method for detecting abnormal coefficient of performance (COP) provided in this application embodiment, before inputting the multiple target point data into a preset anomaly detection model to obtain multiple abnormal point data, the method further includes: determining the amount of data corresponding to the target point data; if the amount of data exceeds a preset data amount threshold, then the method of inputting the multiple target point data into the preset anomaly detection model to obtain multiple abnormal point data is executed.

[0011] Furthermore, in the above-mentioned method for detecting abnormal coefficient of performance (COP) provided in the embodiments of this application, the step of inputting the multiple target point data into a preset anomaly detection model to obtain multiple abnormal point data includes: inputting the multiple target point data into the anomaly detection model to obtain a label corresponding to each target point data; and selecting multiple target point data whose labels are target labels as the multiple abnormal point data.

[0012] Furthermore, in the above-described method for detecting abnormal coefficient of performance (COP) provided in this application embodiment, the step of selecting abnormal point data with a greater importance than a preset importance from the plurality of abnormal point data, and determining the cause of the abnormality of the refrigeration system based on the selected abnormal point data, includes: normalizing the plurality of abnormal point data to obtain a plurality of normalized point data; processing the plurality of normalized point data according to a recursive feature elimination algorithm to obtain the importance of each normalized point data; selecting a preset number of normalized point data with the highest ranking of the importance, and determining the cause of the abnormality of the refrigeration system based on the selected normalized point data.

[0013] A second aspect of this application also provides a cooling coefficient anomaly detection device, applied to a control device. The control device and multiple cooling devices form a cooling system. The cooling coefficient anomaly detection device includes: a first cooling coefficient determination module, used to determine a first cooling coefficient corresponding to the cooling system at preset time intervals; a cooling mode determination module, used to determine the cooling mode corresponding to the cooling system and the duration for which the cooling system is in the cooling mode; and a change point index determination module, used to input multiple first cooling coefficients into a preset change point detection model when the duration is greater than or equal to a target duration, to obtain multiple change points corresponding to the cooling system. The system includes: an index; a second refrigeration coefficient determination module for determining the second refrigeration coefficient corresponding to each variable point index; a target point data acquisition module for acquiring multiple target point data corresponding to each refrigeration device in the refrigeration system if an anomaly is detected in the refrigeration system based on the second refrigeration coefficient; an anomaly point data determination module for inputting the multiple target point data into a preset anomaly detection model to obtain multiple anomaly point data; and an anomaly cause selection module for selecting anomaly point data with a greater importance than a preset importance from the multiple anomaly point data, and determining the cause of the anomaly in the refrigeration system based on the selected anomaly point data.

[0014] A third aspect of this application also provides a control device, which includes a controller and a memory. The controller is used to execute a computer program stored in the memory to implement the cooling coefficient anomaly detection method described in any one of the above embodiments.

[0015] A fourth aspect of this application also provides a computer-readable storage medium storing a computer program, which, when executed by a controller, implements the above-described method for detecting abnormal coefficient of performance (COP).

[0016] The cooling coefficient anomaly detection method provided in this application embodiment determines the first cooling coefficient corresponding to the cooling system at preset time intervals; determines the cooling mode corresponding to the cooling system and the duration of the cooling system in the cooling mode; if the duration is greater than or equal to the target duration, inputs multiple first cooling coefficients into a preset change point detection model to obtain multiple change point indices corresponding to the cooling system. The change point detection model can quickly and accurately predict the time point (i.e., change point index) where the cooling coefficient changes abruptly in the cooling system, and detects whether there is an anomaly in the cooling system based on the second cooling coefficient corresponding to the change point index. For different modes of the ice machine system, data is selected according to the mode to identify anomalies, which can improve the accuracy and speed of cooling coefficient anomaly detection. Furthermore, when an anomaly is detected in the cooling system, this application obtains multiple target point data corresponding to each cooling device in the cooling system; inputs multiple target point data into a preset anomaly detection model to obtain multiple anomaly point data, and selects anomaly point data with an importance greater than a preset importance from the multiple anomaly point data. Based on the selected anomaly point data, the cause of the anomaly in the cooling system is determined, further improving the accuracy and speed of cooling coefficient anomaly detection. Test results show that within one hour of an anomaly occurring in the refrigeration system's data points, the response rate of the control equipment increased by approximately 30%; the method of detecting anomalies in the coefficient of performance (COP) using the control equipment can quickly pinpoint the cause of the anomaly, with a response speed approximately 80 times faster than manual troubleshooting; the embodiments of this application can achieve fully automated anomaly detection, alarming, and cause location, saving over 67% of labor costs. Attached Figure Description

[0017] Figure 1 This is an application scenario diagram of a method for detecting abnormal coefficient of performance (COP) provided in an embodiment of this application;

[0018] Figure 2 This is a flowchart illustrating a method for detecting abnormal coefficient of performance (COP) provided in an embodiment of this application.

[0019] Figure 3 This is a schematic diagram illustrating the process for determining the first coefficient of performance (COP) according to an embodiment of this application.

[0020] Figure 4 This is a flowchart illustrating the determination of a second coefficient of performance (COP) provided in an embodiment of this application.

[0021] Figure 5 This is a flowchart of a refrigeration system anomaly determination provided in an embodiment of this application;

[0022] Figure 6 This is a flowchart illustrating the acquisition of target location data provided in an embodiment of this application;

[0023] Figure 7 This is a flowchart illustrating a data volume determination method provided in an embodiment of this application;

[0024] Figure 8 This is a flowchart illustrating the determination of an abnormal cause provided in an embodiment of this application;

[0025] Figure 9 This is a schematic diagram of the structure of a cooling coefficient anomaly detection device provided in an embodiment of this application;

[0026] Figure 10 This is a schematic diagram of the structure of a control device provided in an embodiment of this application. Detailed Implementation

[0027] The embodiments of this application are described in detail below. Examples of these embodiments are shown 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 are only used to explain this application, and should not be construed as limiting this application.

[0028] In the description of this application, it should be understood that the terms indicating orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. 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 indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, it should be noted that "a plurality of" means two or more, unless otherwise explicitly specified.

[0029] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; 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. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0030] Figure 1This is an application scenario diagram of a method for detecting abnormal coefficient of performance (COP) provided in an embodiment of this application. For example... Figure 1 As shown, the refrigeration system includes control equipment and multiple refrigeration devices. The number of refrigeration devices can be set according to actual needs. For example, the refrigeration system includes three refrigeration devices: refrigeration device A, refrigeration device B, and refrigeration device C. In one embodiment, a refrigeration device refers to a device used to generate a cooling effect. Although not shown, the refrigeration device includes a compressor, a condenser, a throttling valve, and an evaporator. First, the compressor rotates continuously under the drive of a motor, drawing in low-temperature, low-pressure vaporized refrigerant from the evaporator and compressing it into a high-temperature, high-pressure state. Then, the high-temperature, high-pressure vaporized refrigerant enters the condenser, where excess heat is dissipated, cooling the high-temperature, high-pressure vaporized refrigerant into a room-temperature, room-pressure liquid refrigerant. Next, the pressure of the liquid refrigerant is reduced by the expansion valve, thereby lowering the temperature of the liquid refrigerant. The high-pressure, room-temperature liquid refrigerant is then throttled through the throttling valve into a low-temperature, low-pressure liquid refrigerant, creating conditions for refrigerant evaporation. Finally, the liquid refrigerant in the evaporator absorbs heat and evaporates, thus producing a cooling effect.

[0031] In one embodiment, the control device is communicatively connected to each refrigeration unit and receives point data transmitted by each refrigeration unit. Point data refers to data collected during the operation of each refrigeration unit to monitor and control the operating status of its various working modules, reflecting the unit's operational status. The working modules of the refrigeration unit may include, but are not limited to, cooling modules, main body modules, freezing modules, and chemical dosing modules. Point data may include, but is not limited to, inlet and outlet water temperatures of the cooling module, equipment load rate, chilled water temperature / pressure difference, pH value / conductivity of the chemical dosing module, cooling capacity, and power consumption. This point data can be collected using relevant monitoring equipment, such as water temperature sensors or pressure difference sensors. By monitoring and analyzing the point data of each refrigeration unit within the refrigeration system, the control device can promptly detect anomalies in the refrigeration system and ensure its normal operation.

[0032] Figure 2 This is a flowchart illustrating a method for detecting abnormal coefficient of performance (COP) in a refrigeration system, provided in an embodiment of this application. This method can be applied to a control device connected to multiple refrigeration devices to form a refrigeration system. Figure 2 As shown, the method for detecting abnormal coefficient of performance (COP) can include the following steps. Depending on different requirements, the order of the steps in this flowchart can be changed, and some steps can be omitted.

[0033] S11, at preset time intervals, determine and collect the first refrigeration coefficient corresponding to the refrigeration system.

[0034] In one embodiment, the refrigeration system includes multiple refrigeration devices that are in an on state, while refrigeration devices that are in a off state are not considered part of the refrigeration system. For example, in a refrigeration scenario, the system includes refrigeration devices A, B, C, and D, wherein refrigeration devices A, B, and C are all in an on state, and refrigeration device D is in a off state. Thus, refrigeration devices A, B, and C constitute the refrigeration system.

[0035] In one embodiment, each refrigeration device has corresponding location data, and the refrigeration system has corresponding total location data. The total location data refers to the data obtained by combining the location data of multiple refrigeration devices that are in an active state. In one embodiment, the total location data of the refrigeration system can be obtained through direct measurement, for example, by installing a cooling capacity monitor within the refrigeration system to monitor the cooling capacity of the entire system. In other embodiments, the total location data of the refrigeration system can be obtained through indirect measurement, for example, by monitoring the location data of the refrigeration devices that are in an active state within the refrigeration system, and then processing the location data of each refrigeration device accordingly to obtain the total location data. No limitation is imposed here. For example, the refrigeration system includes refrigeration devices A, B, and C that are in an active state, wherein refrigeration device A has an applied cooling capacity A and a power consumption A, refrigeration device B has an applied cooling capacity B and a power consumption B, and refrigeration device C has an applied cooling capacity C and a power consumption C. A refrigeration system corresponds to a total cooling capacity and a total power consumption. The total cooling capacity can be obtained from cooling capacity A, cooling capacity B, and cooling capacity C. For example, the sum of cooling capacity A, cooling capacity B, and cooling capacity C is taken as the total cooling capacity. Similarly, the total power consumption can be obtained from power consumption A, power consumption B, and power consumption C. For example, the sum of power consumption A, power consumption B, and power consumption C is taken as the total cooling capacity.

[0036] In one embodiment, the first coefficient of performance (COP) is an important indicator for evaluating whether a refrigeration system is operating normally and efficiently. The COP refers to the electrical energy consumed by the entire refrigeration system per unit cooling capacity. A higher COP indicates greater energy consumption; a lower COP indicates greater efficiency and energy saving. In one embodiment, the first COP is determined using the following formula 1:

[0037] Formula 1: First coefficient of performance (COP) = Total power consumption of the refrigeration system / Total cooling capacity of the refrigeration system.

[0038] The total cooling capacity and total power consumption can be acquired at preset intervals, which can be set according to actual needs, for example, 5 minutes. Within a certain period, the total cooling capacity and total power consumption of the refrigeration system are acquired every 5 minutes, and the first coefficient of performance (COP) of the refrigeration system is calculated according to Formula 1. Multiple COPs can be collected within a certain period.

[0039] S12, determine the cooling mode corresponding to the cooling system and the duration for which the cooling system is in the cooling mode.

[0040] In one embodiment, before inputting multiple first cooling coefficients into the change point detection model, it is necessary to determine the cooling mode of the cooling system and the duration of the current cooling mode. If the duration meets the requirements, change point detection can be performed on multiple first cooling coefficients. The cooling mode can be determined based on the number of cooling devices in the cooling system that are in the on state. There is a correspondence between the number of cooling devices in the cooling system that are in the on state and the cooling mode. Based on this correspondence, the corresponding cooling mode of the cooling system can be determined quickly and accurately. For example, when the cooling system contains cooling devices A, B, and C that are in the on state, it can be determined that the cooling system is in the first cooling mode; when the cooling system contains cooling devices A and B that are in the on state, it can be determined that the cooling system is in the second cooling mode.

[0041] In one embodiment, when the refrigeration device is in the on state, it sends a power-on flag to the control device at intervals (e.g., every 5 minutes), enabling the control device to determine the on / off status of the refrigeration device. Taking a refrigeration system containing refrigeration devices A, B, and C in the on state as an example, refrigeration device A sends power-on flag A to the control device at intervals, refrigeration device B sends power-on flag B to the control device at intervals, and refrigeration device C sends power-on flag C to the control device at intervals. The control device can determine the duration of receiving power-on flag A, the duration of receiving power-on flag B, and the duration of receiving power-on flag C. In one embodiment, the average of these three durations can be determined as the duration of the refrigeration system in the current refrigeration mode; in other embodiments, the maximum duration among the three durations can also be determined as the duration of the refrigeration system in the current refrigeration mode, without limitation.

[0042] S13, if the duration is greater than or equal to the target duration, then input multiple first cooling coefficients into a preset change point detection model to obtain multiple change point indices corresponding to the cooling system.

[0043] In one embodiment, the target duration is a pre-set duration suitable for change point detection, for example, 15 minutes. The change point detection model is used to detect the time point at which statistical characteristics (e.g., distribution type, distribution parameters) change due to systematic rather than random factors; this time point can be called a change point. The input data of the change point detection model consists of multiple refrigeration coefficients, and the output data consists of the time points where abrupt changes in the refrigeration coefficients occur (also simplified as "change point index" in this embodiment). In one embodiment, the change point indices are denoted as index1, index2, ..., indexN in order along the time axis.

[0044] This application embodiment detects the cooling mode of the cooling system and the duration of the current cooling mode. When the duration is greater than or equal to the target duration, it performs change point detection on multiple first cooling coefficients. This avoids the problem that the number of first cooling coefficients is too small due to the short duration of the cooling mode, and can improve the accuracy of change point detection and identification.

[0045] S14, determine the second cooling coefficient corresponding to each of the variable point indices.

[0046] In one embodiment, the time point at which data collection begins for the total points corresponding to the refrigeration system is designated as the first time point, and the time point at which data collection ends is designated as the second time point. A time axis is established based on the first and second time points. Continuing with the above embodiment, since the number of variable point indices is N, the time axis is divided into N+1 segments according to the variable point indices: the first time point - index1 segment, the index1 - index2 segment, ..., the index(N-1) - indexN segment, and the indexN - second time point segment. For each of the above time axes, there exists at least one first refrigeration coefficient. The multiple first refrigeration coefficients corresponding to this time axis segment are processed accordingly to obtain a second refrigeration coefficient. For example, the average value corresponding to multiple first refrigeration coefficients is calculated as the second refrigeration coefficient. For example, the average value of the multiple first cooling coefficients contained in the first time point - index1 segment is taken as the second cooling coefficient of the variable point index index1, and the second cooling coefficient is denoted as x1; the average value of the multiple first cooling coefficients contained in the index1-index2 segment is taken as the second cooling coefficient of the variable point index index2, and the second cooling coefficient is denoted as x2; and so on, until the average value of the multiple first cooling coefficients contained in the index(N-1)-indexN segment is taken as the second cooling coefficient of the variable point index indexN, and the second cooling coefficient is xn. The details are not elaborated here.

[0047] S15, if an abnormality is detected in the refrigeration system based on the second refrigeration coefficient, then acquire multiple target point data corresponding to each refrigeration device in the refrigeration system.

[0048] In one embodiment, when the refrigeration system malfunctions, the corresponding second coefficient of performance (COP) will rise to a higher level. Thus, by analyzing the refrigeration system over a period of time with a relatively high COP, the presence of an malfunction can be detected. Following the above embodiment, multiple second COPs exist, namely: x1, x2, ..., xn. The maximum value among x1, x2, ..., xn is selected as the target second COP. The presence of an malfunction in the refrigeration system is detected based on the time axis corresponding to the target second COP and the relationship between the target second COP and a preset threshold.

[0049] In one embodiment, if an anomaly is detected in the refrigeration system, anomaly analysis is performed on the target point data of each refrigeration device in the refrigeration system to determine the cause of the anomaly. The target point data refers to the point data collected after a target second refrigeration coefficient is detected (i.e., an anomaly in the refrigeration system is determined). For example, a time axis corresponding to the target point data can be determined based on the target second refrigeration coefficient, and then point data corresponding to each refrigeration device within the refrigeration system on that time axis can be collected as the target point data. If there are anomalous point data within the target point data, the anomalous point data within the target point data of each refrigeration device needs to be identified and analyzed to determine the cause of the refrigeration system anomaly. For example, it can be determined that there is an anomalous refrigeration device within the refrigeration system and that anomalous point data exists within that refrigeration device.

[0050] S16, input the multiple target point data into the preset anomaly detection model to obtain multiple anomaly point data.

[0051] In one embodiment, the anomaly detection model refers to pre-trained point data (also simplified as "abnormal point data" in this application embodiment) used to identify anomalies in target point data. The anomaly detection model can be a binary classification model. The input data of this model consists of multiple point data points, and the output data is a label corresponding to each point data point. The label is used to indicate whether the point data is abnormal. For example, the label can be a number label, a letter label, or a color label. This application embodiment only uses a number label as an example. For instance, when the label of a point data point is 1, the point data is determined to be abnormal; when the label of a point data point is 0, the point data is determined to be normal. In one embodiment, when training the anomaly detection model, normal first point data points are selected, and a first label is added to the first point data point; abnormal second point data points are selected, and a second label is added to the second point data point. The first label can be 0, and the second label can be 1. Using the first point data point and the second point data point as input data, and the corresponding first label and second label as output data, the binary classification model is trained to obtain the anomaly detection model.

[0052] In one embodiment, multiple target point data points of each refrigeration device are input into a preset anomaly detection model to obtain a label corresponding to each target point data point. Target point data points with the label "target label" (e.g., target label 1) are selected as anomaly point data points. In one embodiment, if there is only one anomaly point data point, the cause of the refrigeration system's anomaly is determined based on this anomaly point data point; if there are multiple anomaly point data points, the importance of the anomaly point data points needs to be identified, and the cause of the refrigeration system's anomaly is determined based on the anomaly point data points with higher importance.

[0053] S17, Select abnormal point data with a greater importance than a preset importance from the multiple abnormal point data, and determine the cause of the abnormality of the refrigeration system based on the selected abnormal point data.

[0054] In one embodiment, a performance evaluation model for assessing the performance of a refrigeration system is pre-set. Based on a recursive feature elimination algorithm and the performance evaluation model, high-importance outlier data is selected. The causes of the refrigeration system's anomalies are determined based on the selected outlier data, such as identifying malfunctioning refrigeration equipment and the presence of abnormal data points within that equipment. The performance evaluation model's input data consists of multiple outlier data points, and its output data consists of multiple performance indicators of the refrigeration system. These performance indicators can be set according to actual needs; for example, they may include, but are not limited to, ice production capacity, power consumption, and energy efficiency rating. Ice production capacity measures the equipment's ice-making ability and is a key parameter for judging equipment performance; power consumption reflects the equipment's energy efficiency level; and energy efficiency rating measures the level of energy saving. Recursive Feature Elimination (RFE) is a feature selection method based on a performance evaluation model. It repeatedly trains the model and eliminates the least important outlier data points until the required number is reached. In one embodiment, the output data of the recursive feature elimination method may include the weight of each outlier and the selected outlier data with higher importance, wherein the weight of each outlier can reflect the magnitude of the impact of the outlier on the performance index.

[0055] The cooling coefficient anomaly detection method provided in this application embodiment determines the first cooling coefficient corresponding to the cooling system at preset time intervals; determines the cooling mode corresponding to the cooling system and the duration of the cooling system in the cooling mode; if the duration is greater than or equal to a target duration, inputs multiple first cooling coefficients into a preset change point detection model to obtain multiple change point indices corresponding to the cooling system. The change point detection model can quickly and accurately predict the time point (i.e., change point index) where the cooling coefficient changes abruptly in the cooling system, and detects whether there is an anomaly in the cooling system based on the second cooling coefficient corresponding to the change point index. For different modes of the ice machine system, data is selected according to the mode to identify anomalies, which can improve the accuracy and speed of cooling coefficient anomaly detection. Furthermore, when an anomaly is detected in the cooling system, this application obtains multiple target point data corresponding to each cooling device in the cooling system; inputs the multiple target point data into a preset anomaly detection model to obtain multiple anomaly point data, and selects anomaly point data with an importance greater than a preset importance from the multiple anomaly point data as the cause of the anomaly in the cooling system, further improving the accuracy and speed of cooling coefficient anomaly detection. Test results show that within one hour of an anomaly occurring in the refrigeration system's data points, the response rate of the control equipment increased by approximately 30%; the method of detecting anomalies in the coefficient of performance (COP) using the control equipment can quickly pinpoint the cause of the anomaly, with a response speed approximately 80 times faster than manual troubleshooting; the embodiments of this application can achieve fully automated anomaly detection, alarming, and cause location, saving over 67% of labor costs.

[0056] In one embodiment, the first coefficient of performance (COP) refers to the amount of electricity consumed by the entire refrigeration system per unit of cooling capacity. The larger the COP, the greater the energy consumption of the refrigeration system; the smaller the COP, the more efficient and energy-saving the refrigeration system. Figure 3 This is a schematic diagram illustrating a process for determining the first coefficient of performance (COP) according to an embodiment of this application. The method for determining the first COP is applied to control equipment. Figure 3 As shown, it includes the following steps:

[0057] S21, at each preset time interval, select the first total point data and the second total point data corresponding to the refrigeration system.

[0058] In one embodiment, the preset duration can be set according to actual needs, for example, the preset duration can be 5 minutes. Within a certain period of time, the first total point data and the second total point data corresponding to the refrigeration system are acquired every 5 minutes. The first total point data may refer to the total power consumption of the refrigeration system, and the second total point data may refer to the total cooling capacity of the refrigeration system.

[0059] S22, determine and collect the first refrigeration coefficient based on the first total point data and the second total point data.

[0060] In one embodiment, the ratio of the first total point data to the second total point data (i.e., first total point data / second total point data) is calculated to obtain the first coefficient of performance (COP).

[0061] S23, combine multiple first refrigeration coefficients in order of time axis to obtain the multiple first refrigeration coefficients corresponding to the refrigeration system.

[0062] In one embodiment, the first coefficient of performance (COP) of the refrigeration system is acquired every 5 minutes over a period of time, resulting in multiple COPs. Each COP has a corresponding time point, which is used to identify the calculation time of the COP. Multiple COPs are combined in chronological order according to their time points to obtain multiple COPs corresponding to the refrigeration system.

[0063] This application embodiment determines multiple first refrigeration coefficients of the refrigeration system over a period of time, and then inputs these multiple first refrigeration coefficients into a preset change point detection model to obtain multiple change point indices corresponding to the refrigeration system. By using the change point detection model, the time point at which the refrigeration coefficient changes abruptly in the refrigeration system can be predicted quickly and accurately, thereby improving the accuracy and speed of refrigeration coefficient anomaly detection.

[0064] In one embodiment, after performing change point detection on multiple first refrigeration coefficients to obtain multiple change point indices, a second refrigeration coefficient corresponding to each change point index can be determined, thereby determining the anomaly of the refrigeration system based on the second refrigeration coefficient. Figure 4 This is a flowchart illustrating the determination of a second coefficient of performance (COP) according to an embodiment of this application. The method for determining the second COP is applied to control equipment. Figure 4 As shown, it includes the following steps:

[0065] S41, determine the time axis corresponding to each variable point index.

[0066] In one embodiment, the time point at which data collection begins for the total points corresponding to the refrigeration system is designated as the first time point, and the time point at which data collection ends for the total points corresponding to the refrigeration system is designated as the second time point. The variable point indices are denoted as index1, index2, ..., indexN in order along the time axis. The segment from the first time point to index1 is designated as the time axis corresponding to index1, the segment from index1 to index2 is designated as the time axis corresponding to index2, ..., and the segment from index(N-1) to indexN is designated as the time axis corresponding to indexN.

[0067] S42, determine the average of the multiple first refrigeration coefficients corresponding to the time axis to obtain the second refrigeration coefficient.

[0068] In one embodiment, for each of the aforementioned time axes, there exists at least one first cooling coefficient. Multiple first cooling coefficients corresponding to that time axis are processed accordingly to obtain a second cooling coefficient. For example, the average of the multiple first cooling coefficients is calculated as the second cooling coefficient. In one embodiment, the first time point - index1 corresponds to the second cooling coefficient x1, the index1-index2 segment corresponds to the second cooling coefficient x2, ..., the index(N-1)-indexN segment corresponds to the second cooling coefficient xn.

[0069] For example, if the first time point is 08:30 and index1 is 09:00, and the first cooling coefficient is determined every five minutes, then the segment from the first time point to index1 contains 6 first cooling coefficients. The average of the 6 first cooling coefficients is calculated to obtain the second cooling coefficient x1 corresponding to the variable point index index1. If index1 is 09:00 and index2 is 09:15, and the first cooling coefficient is determined every five minutes, then the segment from index1 to index2 contains 3 first cooling coefficients. The average of the 3 first cooling coefficients is calculated to obtain the second cooling coefficient x2 corresponding to the variable point index index2.

[0070] This application embodiment determines the time axis corresponding to each variable point index and multiple first cooling coefficients corresponding to each time axis. The multiple first cooling coefficients are averaged to obtain a second cooling coefficient. The second cooling coefficient is used to determine the anomaly of the cooling system. When the second cooling coefficient rises to a high level, the anomaly of the cooling system can be detected in time, thereby improving the accuracy and speed of anomaly detection in the cooling system.

[0071] In one embodiment, the presence of an abnormality in the refrigeration system is determined based on a plurality of second refrigeration coefficients. Figure 5 This is a flowchart illustrating a refrigeration system anomaly determination method provided in an embodiment of this application. The refrigeration system anomaly determination method is applied to control equipment. Figure 5 As shown, it includes the following steps:

[0072] S51 selects the maximum value from multiple second coefficients of performance (COPs) as the target second COP.

[0073] In one embodiment, following the above embodiment, the second coefficients of performance are x1, x2, ..., xn. Considering that the second coefficients of performance will rise to a higher level after the refrigeration system fails, the maximum value among x1, x2, ..., xn is selected as the target second coefficient of performance.

[0074] S52, determine the target time axis corresponding to the target second cooling coefficient.

[0075] In one embodiment, following the above embodiments, there is a correspondence between the second cooling coefficient and the time axis. For example, the first time point - index1 corresponds to the second cooling coefficient x1. By querying this correspondence, the target time axis corresponding to the target second cooling coefficient can be obtained.

[0076] S53, if the target time axis is on a preset time axis and the target second cooling coefficient is greater than a preset coefficient threshold, then it is determined that the cooling system is abnormal.

[0077] In one embodiment, the preset time axis can be pre-set according to actual needs. For example, the preset time axis can be the index(N-1)-indexN segment. If the target time axis is in the index(N-1)-indexN segment, that is, when the target second cooling coefficient is xn, it is determined whether xn is greater than the preset coefficient threshold. If xn is greater than the preset coefficient threshold, it is determined that the cooling system is abnormal. If the target time axis is not in the preset time axis position, or the target second cooling coefficient is less than or equal to the preset coefficient threshold, it is determined that the cooling system is normal. The preset coefficient threshold can be pre-set according to the cooling mode of the cooling system. There is a correspondence between the preset coefficient threshold and the cooling mode. By determining the cooling mode of the cooling system and traversing the correspondence according to the cooling mode, the preset coefficient threshold can be obtained.

[0078] This application embodiment selects the maximum value from multiple second cooling coefficients as the target second cooling coefficient, and then determines the position of the target second cooling coefficient on the time axis and its relationship with a preset coefficient threshold to determine whether there is an anomaly in the cooling system, which can improve the accuracy of anomaly detection in the cooling system.

[0079] In one embodiment, target point data refers to point data collected after the target second cooling coefficient is detected (i.e., it is determined that there is an anomaly in the cooling system). There are abnormal point data in the target point data. It is necessary to identify and analyze the abnormal point data in the target point data to determine the cause of the anomaly in the cooling system. Figure 6 This is a flowchart illustrating the acquisition of target point data according to an embodiment of this application. The method for acquiring target point data is applied to control equipment. Figure 6 As shown, it includes the following steps:

[0080] S61, collect multiple initial point data corresponding to each refrigeration device in the refrigeration system according to a preset frequency.

[0081] In one embodiment, the preset frequency can be set according to actual needs; for example, the preset frequency can be 1 minute / time. For each refrigeration device in the refrigeration system, there is corresponding initial point data. For example, the initial point data may include, but is not limited to, the inlet and outlet water temperatures of the cooling module, the equipment load rate, the chilled water temperature / pressure difference, the pH value / conductivity of the dosing module, the cooling capacity, and the power consumption.

[0082] S62, if there are missing point data in the multiple initial point data, the missing point data is filled in according to the preset interpolation method to obtain multiple first intermediate point data corresponding to each refrigeration device.

[0083] In one embodiment, data gaps may occur during data acquisition. For example, if initial point data is collected every minute from 08:00 to 08:05, initial point data should be collected at 08:01, 08:02, 08:03, 08:04, and 08:05. However, only the initial point data corresponding to 08:01, 08:02, and 08:05 are actually collected, while the initial point data corresponding to 08:03 and 08:04 are missing. The initial point data corresponding to 08:03 and 08:04 are then considered as missing point data. In one embodiment, the preset interpolation method can be linear interpolation, which estimates the value of unknown points based on the linear relationship between two known points. That is, the missing point data corresponding to 08:03 and 08:04 can be estimated based on the initial point data corresponding to 08:01 and 08:02. In other embodiments, besides the preset interpolation method, the missing point data can also be filled in by predicting the mean.

[0084] S63, the multiple first intermediate point data are aggregated at each preset time interval to obtain multiple second intermediate point data.

[0085] In one embodiment, the preset duration can be set according to actual needs, for example, the preset duration can be 5 minutes. Within a certain period, data from multiple first intermediate points of the refrigeration equipment are aggregated every 5 minutes to obtain multiple second intermediate point data. The data type of the first intermediate point data can include categorical variables and continuous variables, with different data types corresponding to their respective data aggregation methods. For example, the first intermediate point data can also include a power-on identifier for the refrigeration equipment, which can be 1 or 0, where 1 indicates the refrigeration equipment is in the power-on state and 0 indicates the refrigeration equipment is in the power-off state. The data type corresponding to the power-on identifier is a categorical variable; aggregating the power-on identifier every 5 minutes yields a string of encoded data composed of 1s and 0s. Another example is that the first intermediate point data can be the inlet and outlet water temperatures of the cooling tower, with the data type corresponding to the cooling tower inlet and outlet water temperatures being a continuous variable. Aggregating the cooling tower inlet and outlet water temperatures every 5 minutes means determining the cooling tower inlet and outlet water temperatures collected every minute, calculating the average of the five cooling tower inlet and outlet water temperatures, and obtaining the aggregated value of the cooling tower inlet and outlet water temperatures.

[0086] S64, Select the plurality of target point data from the plurality of second intermediate point data according to the target time axis.

[0087] In one embodiment, each second intermediate point data has a corresponding time point, which is used to identify the calculation time of the second intermediate point data. Based on this time point, the time axis to which each second intermediate point data belongs can be determined. From the time axes corresponding to multiple second intermediate point data, a target time axis is determined, and the second intermediate point data corresponding to the next time axis of the target time axis is taken as the target point data. Continuing with the above embodiment, the target time axis is the index(N-1)-indexN segment, and the next time axis of the target time axis is the indexN-second time point segment. The second intermediate point data within the indexN-second time point segment may have anomalies, so the second intermediate point data corresponding to the indexN-second time point segment is taken as the target point data.

[0088] This application embodiment preprocesses the initial point data of each refrigeration device to avoid the impact of missing values ​​on the anomaly detection of the refrigeration system. Furthermore, by aggregating the initial point data, the amount of data to be analyzed for anomaly detection of the refrigeration system can be reduced, thereby improving the efficiency of anomaly detection. In addition, this application embodiment selects multiple target point data from the multiple second intermediate point data according to the target time axis, thereby improving the accuracy of anomaly point data selection and thus improving the accuracy of anomaly detection of the refrigeration system.

[0089] In one embodiment, before inputting multiple target point data into a preset anomaly detection model, the amount of data at the target points needs to be monitored to avoid the inability to accurately identify the cause of the refrigeration system's anomaly due to the anomaly duration being too short. Figure 7 This is a flowchart illustrating a data volume determination method provided in an embodiment of this application. The data volume determination method is applied to a control device. Figure 7 As shown, it includes the following steps:

[0090] S71, determine the amount of data corresponding to the target point data.

[0091] The method for determining the amount of data can be set according to actual needs.

[0092] S72, if the amount of data exceeds a preset data amount threshold, then the multiple target point data are input into a preset anomaly detection model to obtain multiple anomaly point data.

[0093] In one embodiment, the preset data volume threshold can be set according to actual needs; for example, the preset data volume threshold can be 100.

[0094] In this embodiment of the application, by monitoring the amount of data at multiple target points before inputting the data into the preset anomaly detection model, the problem of short anomaly duration can be avoided, thereby improving the efficiency and accuracy of anomaly analysis of the refrigeration system.

[0095] In one embodiment, anomaly data with higher importance are selected based on the recursive feature elimination algorithm and the performance evaluation model, and the cause of the refrigeration system anomaly is determined based on the selected anomaly data. Figure 8 This is a flowchart illustrating the process of determining the cause of an anomaly, as provided in an embodiment of this application. The method for determining the cause of an anomaly is applied to control equipment. Figure 8 As shown, it includes the following steps:

[0096] S81, normalize the multiple abnormal point data to obtain multiple normalized point data.

[0097] In one embodiment, for multiple outlier data points, normalization is used to map the outlier data points to a fixed range, resulting in multiple normalized data points. For example, multiple outlier data points can be mapped to [0,1]. The normalization method can include deviation normalization and z-score normalization, both of which are existing technologies and will not be elaborated here.

[0098] S82, process the multiple normalized point data according to the recursive feature elimination algorithm to obtain the importance of each normalized point data.

[0099] In one embodiment, multiple normalized point data are processed according to a recursive feature elimination algorithm to obtain the weight of each normalized point data. The weight of each normalized point data can reflect the impact of the normalized point data on the performance indicators of the refrigeration system. The greater the impact of the normalized point data, the higher its importance; the smaller the impact of the normalized point data, the lower its importance.

[0100] S83, select a preset number of normalized point data points ranked first in importance, and determine the cause of the refrigeration system's abnormality based on the selected normalized point data points.

[0101] In one embodiment, the preset number can be set according to actual needs; for example, the preset number can be three. The top three normalized data points based on importance are selected, and the corresponding refrigeration equipment within the refrigeration system is determined. This identifies refrigeration equipment with abnormalities within the refrigeration system, and the abnormal data points corresponding to these normalized data points are used as indicators of refrigeration equipment abnormalities. Anomaly analysis needs to be performed on the modules corresponding to these abnormal data points. For example, if the ultimately selected normalized data points correspond to the inlet and outlet water temperatures of the cooling module of refrigeration equipment A, then it can be determined that the cooling module of refrigeration equipment A within the refrigeration system is abnormal, and anomaly analysis of this cooling module is required.

[0102] Through testing of this embodiment, it was found that within one hour of an anomaly occurring in the refrigeration system's point data, the response rate of the control equipment improved by approximately 30%; the method of detecting anomalies in the coefficient of performance using the control equipment can quickly locate the cause of the anomaly, with a response speed approximately 80 times faster than manual troubleshooting; this embodiment of the application can achieve fully automated anomaly detection, alarm, and cause location, saving more than 67% of labor costs.

[0103] As can be seen, the embodiments of this application select high-importance outlier data as the causes of refrigeration system anomalies based on the recursive feature elimination algorithm and performance evaluation model, which can eliminate unimportant or redundant anomalies and improve the accuracy of refrigeration system anomaly determination.

[0104] Please see Figure 9 , Figure 9 This is a schematic diagram of a cooling coefficient anomaly detection device provided in an embodiment of this application. In some embodiments, the cooling coefficient anomaly detection device 20 may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the cooling coefficient anomaly detection device 20 may be stored in the memory of a control device and executed by at least one processor to perform (see details). Figure 2 (Description) Function for detecting abnormal coefficient of performance (COP).

[0105] In this embodiment, the cooling coefficient anomaly detection device 20 can be divided into multiple functional modules according to its functions. These functional modules may include: a first cooling coefficient determination module 201, a cooling mode determination module 202, a change point index determination module 203, a second cooling coefficient determination module 204, a target point data acquisition module 205, an abnormal point data determination module 206, and an anomaly cause selection module 207. The term "module" in this application refers to a series of computer program segments that can be executed by at least one processor and perform a fixed function, stored in memory. In this embodiment, the functions of each module will be detailed in subsequent embodiments.

[0106] The first cooling coefficient determination module 201 can be used to determine the first cooling coefficient corresponding to the cooling system at preset time intervals.

[0107] The cooling mode determination module 202 can be used to determine the cooling mode corresponding to the cooling system and the duration for which the cooling system is in the cooling mode.

[0108] The variable point index determination module 203 can be used to input multiple first cooling coefficients into a preset variable point detection model when the duration is greater than or equal to the target duration, so as to obtain multiple variable point indices corresponding to the cooling system.

[0109] The second cooling coefficient determination module 204 can be used to determine the second cooling coefficient corresponding to each variable point index.

[0110] The target point data acquisition module 205 can be used to acquire multiple target point data corresponding to each refrigeration device in the refrigeration system if an abnormality is detected in the refrigeration system based on the second refrigeration coefficient.

[0111] The abnormal location data determination module 206 can be used to input the multiple target location data into a preset abnormality detection model to obtain multiple abnormal location data.

[0112] The anomaly cause selection module 207 can be used to select anomaly point data with a greater importance than a preset importance from the multiple anomaly point data, and determine the anomaly cause of the refrigeration system based on the selected anomaly point data.

[0113] It is understood that the cooling coefficient abnormality detection device 20 and the remaining power correction method in the above embodiment belong to the same inventive concept. The specific implementation of each module of the cooling coefficient abnormality detection device 20 corresponds to each step of the remaining power correction method in the above embodiment, and will not be repeated here.

[0114] The module division described above is a logical functional division, and other division methods may be used in actual implementation. Furthermore, the functional modules in the various embodiments of this application can be integrated into the same processing unit, or each module can exist physically separately, or two or more modules can be integrated into the same unit. The integrated modules described above can be implemented in hardware or in a combination of hardware and software functional modules.

[0115] Figure 10 This is a schematic diagram of the structure of a refrigeration device provided in an embodiment of this application, such as... Figure 10 As shown, the control device 30 includes a memory 31, at least one controller 32, and at least one communication bus 33.

[0116] Figure 10 The structure of the refrigeration equipment shown does not constitute a limitation on the embodiments of this application. The control device 30 may also include more or fewer other hardware or software, or different component arrangements than shown. For example, the control device 30 may also include multiple interfaces.

[0117] In one embodiment of this application, the control device 30 may also be connected to a client device, which includes, but is not limited to, any electronic product that can interact with the user via a keyboard, mouse, remote control, touchpad or voice control device, such as a personal computer, tablet computer, smartphone, digital camera, etc.

[0118] It should be noted that the control device 30 is only an example. Other existing or future electronic products that are suitable for this application should also be included within the scope of protection of this application and are incorporated herein by reference.

[0119] In some embodiments, at least one communication bus 33 is configured to enable communication between the memory 31 and at least one controller 32, etc.

[0120] In some embodiments, the control device 30 may also be connected to a power management device (not shown), thereby enabling functions such as managing charging, discharging, and power consumption through the power management device. The control device 30 may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The control device 30 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here. Although not shown, the control device may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to at least one controller through the power management device, thereby enabling functions such as managing charging and discharging through the power management device. The control device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0121] In some embodiments, the memory stores a computer program that, when executed by at least one controller, implements all or part of the steps in the coefficient of performance (COP) anomaly detection method. The memory includes read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage devices, magnetic disk storage devices, magnetic tape storage devices, or any other computer-readable medium capable of carrying or storing data.

[0122] Furthermore, the computer-readable storage medium may mainly include a stored program area and a stored data area, wherein the stored program area may store the operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of the refrigeration coefficient anomaly detection device 1, etc.

[0123] In some embodiments, at least one controller is the control unit of the control device, connecting various components of the entire control device through various interfaces and lines. It executes programs or modules stored in memory and calls data stored in memory to perform various functions of the coefficient of performance (COP) anomaly detection device and process data. For example, when at least one controller executes a computer program stored in a storage device, it implements all or part of the steps of the COP anomaly detection method in this application embodiment; or it implements all or part of the functions of the control device. At least one controller may be composed of integrated circuits, such as a single-packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips.

[0124] The integrated unit implemented as a software functional module described above can be stored in a computer-readable storage medium. This software functional module, stored in a storage medium, includes several instructions to cause a processor to execute portions of the methods described in the various embodiments of this application.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0126] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0127] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0128] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that it can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within this application. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other elements or, and the singular does not exclude the plural. Multiple elements or devices recited in the specification may also be implemented by a single element or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any particular order.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the spirit and scope of the technical solutions of this application.

Claims

1. A method for detecting abnormal coefficient of performance (COP) in cooling systems, applied to control equipment, characterized in that: The control device and multiple refrigeration devices form a refrigeration system, and the method for detecting abnormal refrigeration coefficient includes: At preset time intervals, the first coefficient of performance (COP) of the refrigeration system is determined and collected. The first COP represents the power consumption of the refrigeration system per unit cooling capacity. Determine the cooling mode corresponding to the cooling system and the duration for which the cooling system is in the cooling mode; If the duration is greater than or equal to the target duration, then multiple first cooling coefficients are input into a preset change point detection model to obtain multiple change point indices corresponding to the cooling system. Determine the time axis corresponding to each of the variable point indices, determine the average of the multiple first cooling coefficients corresponding to the time axis, and obtain the second cooling coefficient; If an abnormality is detected in the refrigeration system based on the second refrigeration coefficient, then data on multiple target points corresponding to each refrigeration device in the refrigeration system are obtained. The multiple target point data are input into a preset anomaly detection model to obtain multiple anomaly point data; From the multiple abnormal point data, select the abnormal point data with an importance greater than a preset importance level, and determine the cause of the abnormality of the refrigeration system based on the selected abnormal point data.

2. The method for detecting abnormal coefficient of performance (COP) as described in claim 1, characterized in that, The process of determining and collecting the first coefficient of performance (COP) of the refrigeration system at each preset time interval includes: At each preset time interval, select the first total point data and the second total point data corresponding to the refrigeration system; The first refrigeration coefficient is determined and collected based on the first total point data and the second total point data; By combining multiple first refrigeration coefficients in order along the time axis, multiple first refrigeration coefficients corresponding to the refrigeration system are obtained.

3. The method for detecting abnormal coefficient of performance (COP) as described in claim 1, characterized in that, The method further includes: The maximum value among multiple second coefficients of performance is selected as the target second coefficient of performance. Determine the target time axis corresponding to the target second coefficient of performance; If the target time axis is on a preset time axis and the target second cooling coefficient is greater than a preset coefficient threshold, then it is determined that the cooling system is abnormal.

4. The method for detecting abnormal coefficient of performance (COP) as described in claim 3, characterized in that, The step of acquiring multiple target point data corresponding to each refrigeration device in the refrigeration system includes: Data from multiple initial points corresponding to each refrigeration device in the refrigeration system are collected according to a preset frequency. If there are missing data points among the multiple initial data points, the missing data points are filled in according to a preset interpolation method to obtain multiple first intermediate data points corresponding to each refrigeration device. The data of the multiple first intermediate points are aggregated at each preset time interval to obtain multiple second intermediate point data. The target point data is selected from the multiple second intermediate point data according to the target time axis.

5. The method for detecting abnormal coefficient of performance (COP) as described in claim 1, characterized in that, Before inputting the multiple target point data into a preset anomaly detection model to obtain multiple anomaly point data, the method further includes: Determine the amount of data corresponding to the target point data; If the amount of data exceeds a preset data threshold, then the process of inputting the multiple target point data into a preset anomaly detection model is executed to obtain multiple anomaly point data.

6. The method for detecting abnormal coefficient of performance (COP) as described in claim 4, characterized in that, The step of inputting the multiple target point data into a preset anomaly detection model to obtain multiple anomaly point data includes: The multiple target point data are input into the anomaly detection model to obtain the label corresponding to each target point data; Multiple target point data points with the label as the target label are selected as the multiple abnormal point data points.

7. The method for detecting abnormal coefficient of performance (COP) as described in claim 1, characterized in that, The step of selecting anomaly data points from the plurality of anomaly data points with an importance greater than a preset importance level, and determining the cause of the anomaly in the refrigeration system based on the selected anomaly data points, includes: The multiple outlier point data are normalized to obtain multiple normalized point data. The multiple normalized point data are processed according to the recursive feature elimination algorithm to obtain the importance of each normalized point data. A preset number of normalized data points ranked by importance are selected, and the cause of the refrigeration system's anomaly is determined based on the selected normalized data points.

8. A control device, characterized in that, The control device includes a controller and a memory, wherein the controller is used to execute a computer program stored in the memory to implement the coefficient of performance (COP) anomaly detection method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by the controller, implements the coefficient of performance (COP) anomaly detection method as described in any one of claims 1 to 7.

Images