A scroll compressor fault detection system

Abstract

The present invention discloses a fault detection system for a scroll compressor, comprising: a data acquisition module, a data processing module, and a fault detection module; relating to the technical field of compressor fault detection, and solving the technical problem that it is difficult to predict the fault state of a scroll compressor in the prior art, resulting in inaccurate and untimely detection of the faults of the scroll compressor; the present invention obtains change characteristic data based on the device standard score and the device actual score; calculates a fault evaluation coefficient based on the change characteristic data; inputs the fault evaluation coefficient into a compressor fault prediction model to obtain a fault evaluation prediction value; and detects the scroll compressor based on the fault evaluation coefficient prediction value and a fault evaluation range threshold to obtain a detection result. The present invention calculates the score change amount and the score change rate through the device standard score and the device actual score, can accurately reflect the current operating state of the scroll compressor, and is beneficial to improving the accuracy of subsequent fault detection of the scroll compressor.

Description

Technical Field

[0001] The present invention belongs to the technical field of compressor fault detection, and specifically relates to a scroll compressor fault detection system. Background Art

[0002] Scroll compressors have advantages such as high efficiency, low noise, and compact structure, and play a crucial role in multiple fields such as refrigeration, air conditioning, and gas transmission. However, especially under high-load and long-cycle operating conditions, scroll compressors face many potential fault risks. These faults not only affect the operating efficiency of the equipment but may also pose a serious threat to production safety. Therefore, how to achieve timely and accurate detection and diagnosis of scroll compressor faults has become an urgent technical problem to be solved.

[0003] The prior art sets several sensors on the scroll compressor to collect operating parameters such as noise, temperature, insulation resistance, etc. during operation, and judges whether the scroll compressor has a fault by comparing the operating parameters with corresponding preset thresholds. However, the detection method of the prior art requires the operating parameters to exceed the corresponding thresholds to determine that the scroll compressor has a fault, and it is difficult to predict the fault state of the scroll compressor, resulting in inaccurate and untimely detection of the scroll compressor faults, and further leading to high economic losses due to faults of the scroll compressor and the problem of short service life of the scroll compressor.

[0004] The present invention provides a scroll compressor fault detection system to solve the above technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a scroll compressor fault detection system for solving the technical problems that the detection method of the prior art requires the operating parameters to exceed the corresponding thresholds to determine that the scroll compressor has a fault, and it is difficult to predict the fault state of the scroll compressor, resulting in inaccurate and untimely detection of the scroll compressor faults, and further leading to high economic losses due to faults of the scroll compressor and the problem of short service life of the scroll compressor.

[0006] To achieve the above object, the first aspect of the present invention provides a scroll compressor fault detection system, including: a data acquisition module, a data processing module, and a fault detection module; wherein, the data acquisition module is in communication or electrically connected with the data processing module, and the data processing module is in communication or electrically connected with the fault detection module;

[0007] The data acquisition module: is used to obtain the reference data of the scroll compressor in the factory state; collect the original operation data of the scroll compressor for several consecutive cycles; wherein, both the reference data and the original operation data include the operation noise, housing temperature, lubricating oil consumption efficiency, insulation resistance and compression ratio of the scroll compressor; the lubricating oil consumption rate is obtained through the remaining amount of lubricating oil for several consecutive cycles; and,

[0008] Preprocess the original operation data to obtain operation data;

[0009] The data processing module: is used to calculate the equipment standard score based on the reference data; calculate the actual equipment score based on the operation data; obtain the change characteristic data based on the equipment standard score and the actual equipment score; calculate the fault evaluation coefficient based on the change characteristic data; input the fault evaluation coefficient into the compressor fault prediction model to obtain the fault evaluation prediction value; wherein, the change characteristic data includes the score change amount and the score change rate, and the compressor fault prediction model is obtained by training an artificial intelligence model;

[0010] The fault detection module: is used to detect the scroll compressor based on the fault evaluation coefficient prediction value and the fault evaluation range threshold to obtain a detection result; wherein, the detection result includes normal, minor fault and serious fault.

[0011] Preferably, the lubricating oil consumption rate is obtained through the remaining amount of lubricating oil for several consecutive cycles, including:

[0012] Fit the remaining amount of lubricating oil for several consecutive cycles to generate a change curve of the remaining amount of lubricating oil; take the derivative of the change curve of the remaining amount of lubricating oil to obtain the lubricating oil consumption rate.

[0013] Preferably, the fitting of the change curve of the remaining amount of lubricating oil based on the remaining amount of lubricating oil for several consecutive cycles includes:

[0014] Extract the remaining amount of lubricating oil for several consecutive cycles, and establish a change curve of the remaining amount of lubricating oil with time as the independent variable and the remaining amount of lubricating oil as the dependent variable.

[0015] In the present invention, the lubricating oil consumption rate is obtained by taking the derivative of the change curve of the remaining amount of lubricating oil, which is convenient for calculating the equipment standard score and the actual equipment score according to the lubricating oil consumption rate subsequently.

[0016] Preferably, the preprocessing of the original operation data includes:

[0017] A1: Extract the original operation data for several consecutive cycles;

[0018] A2: Eliminate the abnormal data in the original operation data;

[0019] A3: Fill in the missing data in the original running data;

[0020] A4: Divide the preprocessed raw running data into several groups of running data at equal time intervals, and mark the sequence number of each group of running data as t; where t=1, 2, ..., n, and n is the total number of groups of running data.

[0021] Preferably, the calculating of the equipment standard score based on the benchmark data comprises:

[0022] Extract the operating noise ZS0, housing temperature WD0, lubricating oil consumption rate RX0, insulation resistance JZ0 and compression ratio YB0 from the benchmark data, and use the formula The equipment standard score BP is calculated; wherein α, β, γ are preset constants greater than 0, and k1, k2, k3, k4, k5 are proportional coefficients greater than 0.

[0023] Preferably, the calculating the actual score of the device based on the operation data includes:

[0024] Extract the operating noise ZS from the tth group of operating data t , Shell temperature WD t , Lubricating oil consumption rate RX t , Insulation resistance JZ t and compression ratio YB t , through the formula Calculate the actual equipment score SP corresponding to the tth group of operating data t .

[0025] The present invention calculates the operating noise, shell temperature, lubricating oil consumption efficiency, insulation resistance and compression ratio of the scroll compressor through a formula to obtain the equipment standard score and the equipment actual score, which can accurately reflect the initial state and the current actual operating state of the scroll compressor when it leaves the factory, and is conducive to improving the accuracy of subsequent acquisition of change characteristic data.

[0026] Preferably, the step of obtaining the change characteristic data based on the equipment standard score and the equipment actual score includes:

[0027] By formula Calculate the score change BL t ; Through the formula Calculate the score change rate BS t ; Where T is the duration of a single time interval; the score change BL t and the score change rate BS t Integrate into change feature data.

[0028] The present invention calculates the score change amount and the score change rate through the equipment standard score and the actual equipment score, accurately reflecting the change of the current operating state of the scroll compressor compared with the factory state from two aspects of the change quantity and the change rate, which is beneficial to improving the accuracy of subsequent fault detection of the scroll compressor.

[0029] Preferably, calculating the fault evaluation coefficient based on the change feature data includes:

[0030] Extracting the score change amount BL t and the score change rate BS t from the change feature data, and calculating the fault evaluation coefficient GPX through the formula t ; where a and b are proportionality coefficients greater than 0.

[0031] The present invention calculates the fault evaluation coefficient by calculating the score change amount and the score change rate, which is beneficial to improving the accuracy of the fault state evaluation of the scroll compressor.

[0032] Preferably, the compressor fault prediction model is obtained by training an artificial intelligence model, and includes:

[0033] Extracting the fault evaluation coefficients of several consecutive periods and integrating them into several groups of training data and test data; using the training data to train the artificial intelligence model; using the test data to test the trained artificial intelligence model, and adjusting the artificial intelligence model according to the test results; finally obtaining a compressor fault prediction model with the input being the fault evaluation coefficients of the most recent several consecutive periods and the output being the fault evaluation coefficient of the prediction period; where the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0034] The present invention extracts the fault evaluation coefficient for training the artificial intelligence model, obtains the compressor fault prediction model after training, and inputs the fault evaluation coefficients of several consecutive periods into the compressor fault prediction model to obtain the fault evaluation coefficient of the prediction period, which is beneficial to predicting the fault state of the scroll compressor in advance and taking corresponding measures, thereby being beneficial to reducing the economic loss caused by the fault of the scroll compressor and being beneficial to improving the service life of the scroll compressor.

[0035] Preferably, detecting the scroll compressor based on the predicted value of the fault evaluation coefficient and the fault evaluation range threshold includes:

[0036] B1: Extracting the predicted value GPX of the fault evaluation coefficient 预 ;

[0037] B2: Setting the fault evaluation range threshold [PY1, PY2]; where PY1 < PY2;

[0038] B3: Determine whether the predicted value GPX of the fault evaluation coefficient 预 is within the fault evaluation range threshold [PY1, PY2]; if yes, mark the detection result as a minor fault; if no, jump to B4;

[0039] B4: Determine whether the predicted value GPX of the fault evaluation coefficient 预 is greater than the maximum value PY2 in the fault evaluation range threshold; if yes, mark the detection result as a serious fault; if no, mark the detection result as normal.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] 1. The present invention extracts the fault evaluation coefficient for training the artificial intelligence model. After the training is completed, a compressor fault prediction model is obtained. By inputting the fault evaluation coefficients of several consecutive cycles into the compressor fault prediction model, the fault evaluation coefficient of the prediction cycle is obtained, which is beneficial to predicting the fault state of the scroll compressor in advance and taking corresponding measures, thereby reducing the economic loss caused by the fault of the scroll compressor and improving the service life of the scroll compressor.

[0042] 2. The present invention calculates the equipment standard score and the equipment actual score by formulas using the running noise, shell temperature, lubricating oil consumption efficiency, insulation resistance, and compression ratio of the scroll compressor, which can accurately reflect the initial state at the time of leaving the factory and the current actual running state of the scroll compressor, and is beneficial to improving the accuracy of obtaining the change characteristic data subsequently. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0044] Figure 1 is a schematic diagram of the principle of the scroll compressor fault detection system of the present invention;

[0045] Figure 2 is the overall flowchart of the fault detection method of the scroll compressor of the present invention;

[0046] Figure 3 is the flowchart of preprocessing the original running data in the present invention;

[0047] Figure 4 is the flowchart of fault detection based on the predicted value of the fault evaluation coefficient in the present invention. Specific Embodiments

[0048] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] Please refer to Figures 1 - 4 , an embodiment of the first aspect of the present invention provides a scroll compressor fault detection system, including: a data acquisition module, a data processing module, and a fault detection module; wherein, the data acquisition module is communicatively or electrically connected to the data processing module, and the data processing module is communicatively or electrically connected to the fault detection module;

[0050] Data acquisition module: used to obtain the reference data of the scroll compressor in the factory state; collect the original operation data of the scroll compressor for several consecutive cycles; wherein, the reference data and the original operation data both include the operation noise, shell temperature, lubricating oil consumption efficiency, insulation resistance, and compression ratio of the scroll compressor; the lubricating oil consumption rate is obtained through the remaining amount of lubricating oil for several consecutive cycles; and,

[0051] Preprocess the original operation data to obtain operation data;

[0052] Data processing module: used to calculate the equipment standard score based on the reference data; calculate the equipment actual score based on the operation data; obtain the change characteristic data based on the equipment standard score and the equipment actual score; calculate the fault evaluation coefficient based on the change characteristic data; input the fault evaluation coefficient into the compressor fault prediction model to obtain the fault evaluation prediction value; wherein, the change characteristic data includes the score change amount and the score change rate, and the compressor fault prediction model is obtained by training with an artificial intelligence model;

[0053] Fault detection module: used to detect the scroll compressor based on the fault evaluation coefficient prediction value and the fault evaluation range threshold to obtain a detection result; wherein, the detection result includes normal, mild fault, and severe fault.

[0054] In this embodiment, the lubricating oil consumption rate is obtained through the remaining amount of lubricating oil for several consecutive cycles, including:

[0055] Fitting the remaining amount of lubricating oil for several consecutive cycles to generate a lubricating oil remaining amount change curve; taking the derivative of the lubricating oil remaining amount change curve to obtain the lubricating oil consumption rate.

[0056] In this embodiment, fitting the remaining amount of lubricating oil for several consecutive cycles to generate a lubricating oil remaining amount change curve includes:

[0057] Extract the remaining lubricating oil volume for several consecutive periods, and establish a curve of the change in the remaining lubricating oil volume with time as the independent variable and the remaining lubricating oil volume as the dependent variable.

[0058] In the present invention, the lubricating oil consumption rate is obtained by differentiating the curve of the change in the remaining lubricating oil volume, which is convenient for calculating the standard score and the actual score of the equipment according to the lubricating oil consumption rate subsequently.

[0059] In this embodiment, preprocessing is performed on the original operation data, including:

[0060] A1: Extract the original operation data for several consecutive periods;

[0061] A2: Eliminate the abnormal data in the original operation data;

[0062] A3: Fill in the missing data in the original operation data;

[0063] A4: Divide the preprocessed original operation data into several groups of operation data at equal time intervals, and mark the serial number of each group of operation data as t; where t = 1, 2,..., n, and n is the total number of groups of operation data.

[0064] Exemplarily, the original operation data of the scroll compressor is continuously collected for 150 minutes through the data acquisition module, and the preprocessed original operation data is divided into 150 groups of operation data at one-minute time intervals.

[0065] In this embodiment, calculating the standard score of the equipment based on the reference data includes:

[0066] Extract the running noise ZS0, shell temperature WD0, lubricating oil consumption rate RX0, insulation resistance JZ0, and compression ratio YB0 in the reference data, and calculate the standard score BP of the equipment through the formula where α, β, γ are preset constants greater than 0, and k1, k2, k3, k4, k5 are proportionality coefficients greater than 0.

[0067] Exemplarily, set α = 0.5 ml / s, β = 2 ln, γ = 3, k1 = 0.2, k2 = 0.15, k3 = 10, k4 = 4, k5 = 7, running noise ZS0 = 60 dB, shell temperature WD0 = 123 °C, lubricating oil consumption rate RX0 = 0.53 ml / s, insulation resistance JZ0 = 1.9 ln, and compression ratio YB0 = 3.2; calculate the standard score BP = 32.55 through the formula;

[0068] In this embodiment, calculating the actual score of the equipment based on the operation data includes:

[0069] Extract the running noise ZS in the t-th group of operation data t, the housing temperature WD t , the lubricating oil consumption rate RX t , the insulation resistance JZ t and the compression ratio YB t , through the formula Calculate the actual equipment score SP corresponding to the t-th group of operating data t .

[0070] Exemplarily, set the operating noise ZS in the t-th group of operating data t = 70 dB, the housing temperature WD t = 130 °C, the lubricating oil consumption rate RX t = 0.87 ml / s, the insulation resistance JZ t = 1.2 MΩ and the compression ratio YB t = 4.1; Calculate the actual equipment score SP corresponding to the t-th group of operating data through the formula t = 48.1.

[0071] The present invention calculates the equipment standard score and the actual equipment score through the operating noise, housing temperature, lubricating oil consumption efficiency, insulation resistance and compression ratio of the scroll compressor, which can accurately reflect the initial state and the current actual operating state of the scroll compressor at the factory state, and is beneficial to improving the accuracy of obtaining the change characteristic data subsequently.

[0072] In this embodiment, obtaining the change characteristic data based on the equipment standard score and the actual equipment score includes:[[]]

[0073] Through the formula Calculate the score change amount BL t ; Through the formula Calculate the score change rate BS t ; where T is the duration of a single time interval; Integrate the score change amount BL t and the score change rate BS t into the change characteristic data.

[0074] Exemplarily, set the duration of a single time interval T = 1 min, the score change amount BL t = 15.55, the score change amount BL t-1 = 16.23, the score change rate BS t = 0.68 / min;

[0075] The present invention calculates the score change amount and the score change rate through the equipment standard score and the actual equipment score, and accurately reflects the change of the current operating state of the scroll compressor compared with the factory state from two aspects of the change quantity and the change rate.

[0076] In this embodiment, calculating a fault evaluation coefficient based on the change feature data includes:

[0077] Extracting the score change amount BL in the change feature data t and the score change rate BS t , and calculating the fault evaluation coefficient GPX through the formula ; where a and b are proportionality coefficients greater than 0. t

[0078] Exemplarily, set a = 0.3 and b = 4, substitute the score change amount BL t = 15.55 and the score change rate BS t = 0.68 / min into the formula to calculate the fault evaluation coefficient GPX t = 7.385.

[0079] The present invention calculates the fault evaluation coefficient by calculating the score change amount and the score change rate, which is beneficial to improving the accuracy of the fault state evaluation of the scroll compressor.

[0080] In this embodiment, the compressor fault prediction model is obtained by training an artificial intelligence model, including:

[0081] Extracting the fault evaluation coefficients of several consecutive periods and integrating them into several groups of original data, taking 80% of the original data as training data and 20% as test data; training the artificial intelligence model with the training data; testing the trained artificial intelligence model with the test data, and adjusting the artificial intelligence model according to the test results; finally obtaining a compressor fault prediction model with the input being the fault evaluation coefficients of the most recent several consecutive periods and the output being the fault evaluation coefficient of the prediction period; where the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0082] The present invention extracts the fault evaluation coefficient for training the artificial intelligence model, obtains the compressor fault prediction model after training, and inputs the fault evaluation coefficients of several consecutive periods into the compressor fault prediction model to obtain the fault evaluation coefficient of the prediction period, which is beneficial to predicting the fault state of the scroll compressor in advance and taking corresponding measures, thereby being beneficial to reducing the economic loss caused by the fault of the scroll compressor and being beneficial to improving the service life of the scroll compressor.

[0083] In this embodiment, detecting the scroll compressor based on the predicted value of the fault evaluation coefficient and the fault evaluation range threshold includes:

[0084] B1: Extracting the predicted value GPX of the fault evaluation coefficient 预 ;

[0085] B2: Set the fault evaluation range threshold [PY1, PY2]; where PY1 < PY2;

[0086] B3: Determine whether the predicted value GPX of the fault evaluation coefficient 预 is within the fault evaluation range threshold [PY1, PY2]; if yes, mark the detection result as a minor fault; if no, jump to B4;

[0087] B4: Determine whether the predicted value GPX of the fault evaluation coefficient 预 is greater than the maximum value PY2 in the fault evaluation range threshold; if yes, mark the detection result as a severe fault; if no, mark the detection result as normal.

[0088] Exemplarily, set the predicted value GPX of the fault evaluation coefficient 预 = 7.5; set PY1 = 6.5, PY2 = 7.8, then the fault evaluation range threshold [PY1, PY2] = [6.5, 7.8]; since the predicted value GPX of the fault evaluation coefficient 预 is within the fault evaluation range threshold [PY1, PY2], the detection result is marked as a minor fault.

[0089] For some data in the above formula, the dimension is removed and only the numerical value is calculated. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the actual situation; the preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0090] The working principle of the present invention:

[0091] The present invention obtains the reference data of the scroll compressor in the factory state; collects the original operation data of the scroll compressor for several consecutive cycles; preprocesses the original operation data to obtain the operation data; calculates the standard score of the equipment based on the reference data; calculates the actual score of the equipment based on the operation data; obtains the change characteristic data based on the standard score and the actual score of the equipment; calculates the fault evaluation coefficient based on the change characteristic data; inputs the fault evaluation coefficient into the compressor fault prediction model to obtain the predicted value of the fault evaluation; and performs detection on the scroll compressor based on the predicted value of the fault evaluation coefficient and the fault evaluation range threshold to obtain the detection result.

[0092] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A scroll compressor fault detection system, comprising: Data acquisition module, data processing module and fault detection module; characterized in that the data acquisition module communicates with or is electrically connected to the data processing module, and the data processing module communicates with or is electrically connected to the fault detection module; The data acquisition module is used to obtain the reference data of the scroll compressor in the factory state; collect the original operation data of the scroll compressor for several consecutive cycles; wherein the reference data and the original operation data both include the operation noise, shell temperature, lubricating oil consumption efficiency, insulation resistance and compression ratio of the scroll compressor; the lubricating oil consumption rate is obtained by the remaining amount of lubricating oil for several consecutive cycles; and, Preprocessing the original operation data to obtain operation data; The data processing module is used to calculate the equipment standard score based on the benchmark data; calculate the equipment actual score based on the operation data; obtain the change characteristic data based on the equipment standard score and the equipment actual score; calculate the fault assessment coefficient based on the change characteristic data; input the fault assessment coefficient into the compressor fault prediction model to obtain the fault assessment prediction value; wherein the change characteristic data includes the score change amount and the score change rate, and the compressor fault prediction model is obtained by training the artificial intelligence model; The fault detection module is used to detect the scroll compressor based on the fault assessment coefficient prediction value and the fault assessment range threshold value to obtain a detection result; wherein the detection result includes normal, slight fault and severe fault; The compressor fault prediction model is obtained through artificial intelligence model training, including: Extract fault assessment coefficients of several consecutive cycles and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain a compressor fault prediction model whose input is the fault assessment coefficient of the most recent several consecutive cycles and whose output is the fault assessment coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

2. The vortex compressor fault detection system according to claim 1, wherein The lubricating oil consumption rate is obtained through the remaining amount of lubricating oil in several consecutive cycles, including: A lubricating oil remaining amount variation curve is generated based on the lubricating oil remaining amount fitting of several continuous cycles; and the lubricating oil consumption rate is obtained by derivation of the lubricating oil remaining amount variation curve.

3. The vortex compressor fault detection system according to claim 2, wherein, The generating of the lubricating oil remaining amount variation curve based on the lubricating oil remaining amount fitting of several consecutive cycles includes: The remaining lubricating oil amount of several consecutive cycles is extracted, and the remaining lubricating oil amount variation curve is established with time as the independent variable and the remaining lubricating oil amount as the dependent variable.

4. A scroll compressor fault detection system according to claim 1, wherein, The preprocessing of the running raw data includes: A1: Extract the raw running data of several consecutive cycles; A2: Eliminate abnormal data from the original running data; A3: Fill in the missing data in the original running data; A4: Divide the preprocessed raw running data into several groups of running data at equal time intervals, and mark the sequence number of each group of running data as t; where t=1, 2, ..., n, and n is the total number of groups of running data.

5. The fault detection system for a scroll compressor according to claim 4, wherein The device standard score is calculated based on the benchmark data, including: Extract the running noise ZS0, housing temperature WD0, lubricating oil consumption rate RX0, insulation resistance JZ0, and compression ratio YB0 from the reference data, and through the formula calculate the standard score BP of the equipment; where α, β, and γ are preset constants greater than 0, and k1, k2, k3, k4, and k5 are proportionality coefficients greater than 0.

6. The vortex compressor fault detection system according to claim 5, wherein, Calculating the actual score of the device based on the operation data, including: Extract the operating noise ZS from the t-th group of operating data t , the housing temperature WD t , the lubricating oil consumption rate RX t , the insulation resistance JZ t and the compression ratio YB t , and calculate the actual equipment score SP corresponding to the t-th group of operating data through the formula t .​ 7. The vortex compressor fault detection system according to claim 6, characterized in that, Obtaining the change characteristic data based on the standard score and the actual score of the device, including: Calculate the score change amount BL through the formula ; Calculate the score change rate BS through the formula t ; where T is the duration of a single time interval; Integrate the score change amount BL and the score change rate BS t into change feature data t ; t ​ 8. A scroll compressor fault detection system according to claim 7, characterized in that, Calculating the fault evaluation coefficient based on the change characteristic data, including: Extract the score change amount BL t and the score change rate BS t , and calculate the fault evaluation coefficient GPX through the formula t ; where a and b are proportionality coefficients greater than 0.

9. The fault detection system for a scroll compressor according to claim 1, wherein Detecting the scroll compressor based on the predicted value of the fault evaluation coefficient and the fault evaluation range threshold, including: B1: Extract the predicted value GPX of the fault evaluation coefficient 预 ; B2: Setting the fault evaluation range threshold [PY1, PY2]; where PY1 < PY2; B3: Determine whether the predicted value GPX of the fault evaluation coefficient 预 is within the fault evaluation range threshold [PY1, PY2]; if yes, mark the detection result as a minor fault; if no, jump to B4; B4: Determine the predicted value GPX of the fault assessment coefficient 预 Check whether it is greater than the maximum value PY2 of the fault assessment range threshold; if yes, mark the detection result as a serious fault; if not, mark the detection result as normal.

Images