Fault detection method and system for digital equipment in building materials processing based on the Internet of Things

Abstract

The present invention relates to the field of fault detection technology, and in particular to a method and system for fault detection of digital equipment for building materials processing based on the Internet of Things. The method for fault detection of digital equipment for building materials processing based on the Internet of Things includes the following steps: S1: obtaining equipment operation parameter data, hydraulic equipment operating environment parameter data and hydraulic equipment operation detection data of hydraulic equipment; S2: obtaining the flow rate change of hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operating environment parameter data and establishing a coordinate system. The present invention establishes a coordinate system in the hydraulic equipment by combining the change of hydraulic oil under cold working conditions with the frequent start and stop of hydraulic equipment and the rapid change of load, and determines the first fault interval according to the degree of disorder of the coordinate system; then, according to the comparison between the detection data of the hydraulic equipment and the detection data of the first fault interval, the second fault interval is determined according to the simulation results of the fault diagnosis model, and finally the fault area is determined according to the first fault interval and the second fault interval.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and in particular to a method and system for fault detection of digital equipment for building material processing based on the Internet of Things. Background Art

[0002] In recent years, with the continuous development of my country's industrialization level, the scale of building materials digital equipment has gradually expanded, the demand for hydraulic equipment has also gradually increased, and the pressure on fault detection of hydraulic equipment has also gradually increased. Hydraulic equipment plays an important role in many fields such as construction, road construction, aerospace, and agricultural machinery due to its wide range of operating areas, strong site adaptability, and strong versatility. As an industrial mechanical equipment, hydraulic equipment will cause major safety accidents regardless of any failure due to the complexity of the working environment of hydraulic equipment and the particularity of its own equipment.

[0003] From the perspective of hydraulic equipment fault detection methods, today's hydraulic equipment fault detection methods are usually based on various sensors for analysis to perform fault diagnosis and system maintenance. Today's hydraulic equipment fault detection methods not only require accurate mathematical model support, but also have many problems such as slow detection speed, poor flexibility, and poor versatility. From the perspective of hydraulic equipment maintenance methods, today's hydraulic equipment maintenance is generally managed manually. In the context of the era of rapid development of intelligence, hydraulic equipment is gradually developing in the direction of complexity and intelligence. For maintenance management, the workload of manual maintenance, maintenance response capabilities, and the comprehensive skill requirements of maintenance personnel all face major challenges. Summary of the invention

[0004] In order to overcome the shortcomings of frequent start and stop of hydraulic equipment under cold working conditions and large fault detection hysteresis and error under rapid load change operations, the present invention provides a fault detection method and system for digital equipment for building materials processing based on the Internet of Things.

[0005] The technical implementation scheme of the present invention is: a method for fault detection of building materials processing digital equipment based on the Internet of Things, comprising the following steps:

[0006] S1: Acquire equipment operation parameter data of hydraulic equipment, hydraulic equipment operating environment parameter data, and hydraulic equipment operation detection data;

[0007] S2: Obtain the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establish a coordinate system;

[0008] S3: Obtaining the first fault interval according to the degree of disorder reflected by the coordinate system and the operation cycle data of the hydraulic equipment;

[0009] S4: obtaining a first influencing factor sequence according to the hydraulic equipment operation detection data; obtaining a second influencing factor sequence according to the first fault interval;

[0010] S5: Obtain a second fault interval according to the first influencing factor sequence and the second influencing factor sequence;

[0011] S6: Obtaining a fault area of ​​the hydraulic equipment according to the first fault area and the second fault area.

[0012] Preferably, the method of obtaining the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establishing a coordinate system includes: obtaining the viscosity change trend of the hydraulic oil with temperature in the hydraulic equipment preheating operation, and obtaining the estimated flow rate of the hydraulic oil according to the viscosity change trend of the hydraulic oil with temperature;

[0013] Among them, the viscosity change trend of the hydraulic oil with temperature is obtained according to the viscosity model equation. The viscosity model equation formula is: ,in,( ) represents the kinematic viscosity of the liquid (usually in cSt, i.e. centistokes), (T) is the absolute temperature (in Kelvin K), (A, B, C, D) are constants determined according to the characteristics of the specific liquid, ( ) is a reference temperature, a standard test temperature for the liquid is selected.

[0014] Preferably, obtaining the estimated flow rate of the hydraulic oil according to the viscosity variation trend of the hydraulic oil with temperature comprises: obtaining the actual flow rate of the hydraulic oil according to the estimated flow rate of the hydraulic oil in combination with the operating parameter data of the hydraulic equipment, obtaining the flow rate of the hydraulic oil in the hydraulic equipment according to the actual flow rate of the hydraulic oil, and calculating the flow rate change of the hydraulic equipment in the startup stage, stable working condition, variable working condition and after shutdown cooling according to the flow rate of the hydraulic oil in the hydraulic equipment;

[0015] Among them, the expected flow rate, equipment geometry, oil properties and hydraulic equipment operating parameters are input into the flow rate change model to perform flow field analysis to predict the actual flow rate distribution of the hydraulic oil in the system; combined with the actual operation data of the hydraulic equipment, a prediction model is established using a support vector machine to more accurately predict the actual flow rate; system identification technology is applied, combined with online sensor data, and model parameters are adjusted in real time to achieve online estimation of the actual flow rate of the hydraulic oil; the Kalman filter is used to fuse the model prediction and measured data to improve the accuracy and real-time performance of the flow rate estimation; dynamic simulation is implemented in the simulation software to simulate the flow rate changes of the hydraulic equipment during the startup phase, stable operating conditions, variable operating conditions and shutdown cooling.

[0016] Preferably, the calculating of the flow rate change of the hydraulic equipment in the startup stage, stable working condition, variable working condition and after shutdown and cooling according to the flow rate of the hydraulic oil in the hydraulic equipment comprises: establishing a coordinate system in the hydraulic equipment according to the flow rate change, reflecting the degree of dispersion of the flow rate change in the hydraulic equipment according to the coordinate system, and inferring the fault according to the degree of dispersion of the flow rate change;

[0017] Among them, a time-space coordinate system is established based on the physical structure of the hydraulic equipment and the fluid flow path; whether the coordinate system is a two-dimensional plane coordinate system or a three-dimensional space coordinate system is determined according to the complexity of the hydraulic system; each coordinate point represents the position and time information of a monitoring point in the system; the flow rate data of each monitoring point is mapped to the above coordinate system to form a flow rate distribution diagram; a statistical method is used to quantify the degree of dispersion of flow rate changes and identify the characteristics of flow rate fluctuations; a fault classification model is established using support vector machine SVM; the training data set should contain the flow rate change characteristics under normal operation and known fault conditions; when it is detected that the dispersion of flow rate changes exceeds the preset threshold, a fault warning is automatically triggered, and maintenance personnel are notified in time to conduct inspection and intervention.

[0018] Preferably, the obtaining of the first fault interval according to the degree of disorder reflected by the coordinate system in combination with the operating cycle data of the hydraulic equipment comprises: inferring the fault according to the degree of disorder and obtaining the first fault interval in combination with the operating cycle, frequency and load change data of the hydraulic equipment;

[0019] First, the LSTM model is constructed, and then the trained LSTM model is used to extract features from the data. The output layer of the LSTM is used as input to the isolation forest model. Based on the anomalies identified by the isolation forest model, the dense area where faults occur is identified in the time series, which is the fault interval. The output layer formula of the LSTM is:

[0020]

[0021] ,in,( ) is the sigmoid activation function, ( ), ) is the weight matrix, ( ) is the bias term, ( ) is the current input, ( ) is the hidden state of the previous time step, and the forget gate formula is

[0022]

[0023] , the cell state formula is

[0024]

[0025] ,here,( ) represents the element-wise product, ( ) is the hyperbolic tangent activation function, which is used to generate new candidate memories. The output gate formula is

[0026]

[0027] , the hidden state formula is

[0028]

[0029] ,LSTM can learn features from historical operating cycle, frequency, and load change sequence data, and then these features can be passed as input to the subsequent isolation forest model for anomaly detection, thereby identifying the fault interval.

[0030] Preferably, the method of obtaining a first influencing factor sequence based on the hydraulic equipment operation detection data and obtaining a second influencing factor sequence based on the first fault interval includes: obtaining abnormal data detected in the hydraulic equipment based on the hydraulic equipment operation detection data, and sorting the abnormal data according to the abnormality degree to obtain the first influencing factor sequence; obtaining abnormal data of hydraulic equipment faults based on the first fault interval, and sorting the abnormal data according to the abnormality degree to obtain the second influencing factor sequence.

[0031] Preferably, the method of obtaining a first influencing factor sequence according to hydraulic equipment operation detection data; and obtaining a second influencing factor sequence according to the first fault interval includes: obtaining common influencing factor data in the first influencing factor sequence and the second influencing factor sequence according to the first influencing factor sequence and the second influencing factor sequence, sorting according to the common influencing factor data, and obtaining multiple influencing factor sequences according to the sorting results.

[0032] Preferably, obtaining multiple influencing factor sequences according to the sorting results includes: placing the multiple influencing factor sequences into a fault diagnosis model for correction, recording the correction gain of each influencing factor, dividing the correction gain into positive gain and negative gain according to the correction result, and obtaining the maximum influencing path after all sequences are corrected.

[0033] Preferably, obtaining the second fault interval according to the first impact factor sequence and the second impact factor sequence includes: after all sequences are corrected and the maximum impact path is obtained, the second fault interval is determined, and according to the first fault interval and the second fault interval, a faulty device in the first fault interval and a faulty device in the second fault interval are obtained;

[0034] Among them, the method of obtaining the fault area of ​​the hydraulic equipment according to the first fault interval and the second fault interval includes: determining the first fault area according to the intersection area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, determining the second fault area according to the difference area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, and determining the fault troubleshooting sequence and maintenance suggestions according to the first fault area and the second fault area.

[0035] The digital equipment fault detection system for building materials processing based on the Internet of Things includes:

[0036] Data acquisition and preprocessing module: used to obtain equipment operation parameter data of hydraulic equipment, hydraulic equipment operating environment parameter data and hydraulic equipment operation detection data and perform preprocessing, then normalize and divide the normalized data into training set and test set;

[0037] Temperature change trend module: used to obtain the trend of hydraulic oil in hydraulic equipment with temperature change;

[0038] Hydraulic oil estimated flow rate module: used to obtain the estimated flow rate of hydraulic oil according to the trend of hydraulic oil changing with temperature;

[0039] Support vector machine prediction model building module: used to combine the actual operation data of hydraulic equipment and use support vector machine to build a prediction model, so as to more accurately predict the actual flow rate;

[0040] Kalman filter fusion building block: used to use Kalman filter fusion model prediction and measured data to improve the accuracy and real-time performance of flow velocity estimation;

[0041] Simulation module: used to implement dynamic simulation in the simulation software to simulate the flow rate changes of hydraulic equipment during the startup phase, stable working conditions, variable working conditions and shutdown cooling;

[0042] Support vector machine fault classification module: used to classify faults;

[0043] LSTM model building module: used to analyze and process time series;

[0044] Isolation forest model construction module: Then use the trained LSTM model to extract features from the data, and use the output layer of LSTM as input to the isolation forest model to identify abnormal points based on the isolation forest model;

[0045] Fault diagnosis model building module: used to put multiple influencing factor sequences into the fault diagnosis model for correction, record the correction gain of each influencing factor, divide the correction gain into positive gain and negative gain according to the correction result, and obtain the maximum influencing path after all sequences are corrected.

[0046] Beneficial effects:

[0047] The present invention establishes a coordinate system in the hydraulic equipment by studying the viscosity variation trend of the hydraulic oil of the hydraulic equipment under cold working conditions with temperature, combining the frequent start and stop of the hydraulic equipment and the load working state, and determines the first fault interval according to the abnormal flow velocity state reflected by the coordinate system; at the same time, the second fault interval is determined in the correction model in combination with the hydraulic equipment detection data and different influencing factors determined by the first fault interval; the fault area is determined according to the intersection of the first fault interval and the second fault interval; in each fault detection process, multiple algorithms are used to realize accurate identification and processing of data and faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The present invention is a flow chart of a method and system for fault detection of digital equipment for building materials processing based on the Internet of Things. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Embodiment 1: A method for detecting faults in digital equipment for building material processing based on the Internet of Things, such as Figure 1 As shown, the following steps are included:

[0051] S1: Acquire equipment operation parameter data of hydraulic equipment, hydraulic equipment working environment parameter data and hydraulic equipment operation detection data;

[0052] S2: Obtain the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establish a coordinate system;

[0053] S3: Obtaining the first fault interval according to the degree of disorder reflected by the coordinate system and the operation cycle data of the hydraulic equipment;

[0054] S4: obtaining a first influencing factor sequence according to the hydraulic equipment operation detection data; obtaining a second influencing factor sequence according to the first fault interval;

[0055] S5: Obtain a second fault interval according to the first influencing factor sequence and the second influencing factor sequence;

[0056] S6: Obtaining a fault area of ​​the hydraulic equipment according to the first fault area and the second fault area.

[0057] Obtain the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establish a coordinate system, including: obtaining the viscosity change trend of the hydraulic oil with temperature in the hydraulic equipment preheating operation, and obtaining the estimated flow rate of the hydraulic oil according to the viscosity change trend of the hydraulic oil with temperature;

[0058] Among them, the viscosity change trend of the hydraulic oil with temperature is obtained according to the viscosity model equation. The viscosity model equation formula is: ,in,( ) represents the kinematic viscosity of the liquid (usually in cSt, i.e. centistokes), (T) is the absolute temperature (in Kelvin K), (A, B, C, D) are constants determined according to the characteristics of the specific liquid, ( ) is a reference temperature, a standard test temperature for the liquid is selected.

[0059] Obtaining an estimated flow rate of the hydraulic oil according to the viscosity variation trend of the hydraulic oil with temperature, including: obtaining an actual flow rate of the hydraulic oil according to the estimated flow rate of the hydraulic oil combined with the operating parameter data of the hydraulic equipment, obtaining the flow rate of the hydraulic oil in the hydraulic equipment according to the actual flow rate of the hydraulic oil, and calculating the flow rate change of the hydraulic equipment in the startup stage, stable working condition, variable working condition and after shutdown cooling according to the flow rate of the hydraulic oil in the hydraulic equipment;

[0060] Among them, the expected flow rate, equipment geometry, oil properties and hydraulic equipment operating parameters are input into the flow rate change model to perform flow field analysis to predict the actual flow rate distribution of the hydraulic oil in the system; combined with the actual operation data of the hydraulic equipment, a prediction model is established using a support vector machine to more accurately predict the actual flow rate; system identification technology is applied, combined with online sensor data, and model parameters are adjusted in real time to achieve online estimation of the actual flow rate of the hydraulic oil; the Kalman filter is used to fuse the model prediction and measured data to improve the accuracy and real-time performance of the flow rate estimation; dynamic simulation is implemented in the simulation software to simulate the flow rate changes of the hydraulic equipment during the startup phase, stable operating conditions, variable operating conditions and shutdown cooling.

[0061] Calculating the flow velocity changes of the hydraulic equipment in the startup stage, stable working condition, variable working condition and after shutdown cooling according to the flow velocity of the hydraulic oil in the hydraulic equipment, including: establishing a coordinate system in the hydraulic equipment according to the flow velocity changes, reflecting the degree of dispersion of the flow velocity changes in the hydraulic equipment according to the coordinate system, and inferring the fault according to the degree of dispersion of the flow velocity changes;

[0062] Among them, a time-space coordinate system is established based on the physical structure of the hydraulic equipment and the fluid flow path; whether the coordinate system is a two-dimensional plane coordinate system or a three-dimensional space coordinate system is determined according to the complexity of the hydraulic system; each coordinate point represents the position and time information of a monitoring point in the system; the flow rate data of each monitoring point is mapped to the above coordinate system to form a flow rate distribution diagram; a statistical method is used to quantify the degree of dispersion of flow rate changes and identify the characteristics of flow rate fluctuations; a fault classification model is established using support vector machine SVM; the training data set should contain the flow rate change characteristics under normal operation and known fault conditions; when it is detected that the dispersion of flow rate changes exceeds the preset threshold, a fault warning is automatically triggered, and maintenance personnel are notified in time to conduct inspection and intervention.

[0063] Obtaining a first fault interval according to the degree of disorder reflected by the coordinate system combined with the operating cycle data of the hydraulic equipment, including: inferring the fault according to the degree of disorder combined with the operating cycle, frequency and load change data of the hydraulic equipment to obtain the first fault interval;

[0064] First, the LSTM model is constructed, and then the trained LSTM model is used to extract features from the data. The output layer of the LSTM is used as input to the isolation forest model. Based on the anomalies identified by the isolation forest model, the dense area where faults occur is identified in the time series, which is the fault interval. The output layer formula of the LSTM is:

[0065]

[0066] ,in,( ) is the sigmoid activation function, ( ), ) is the weight matrix, ( ) is the bias term, ( ) is the current input, ( ) is the hidden state of the previous time step, and the forget gate formula is

[0067]

[0068] , the cell state formula is

[0069]

[0070] ,here,( ) represents the element-wise product, ( ) is the hyperbolic tangent activation function, which is used to generate new candidate memories. The output gate formula is

[0071]

[0072] , the hidden state formula is

[0073]

[0074] ,LSTM can learn features from historical operating cycle, frequency, and load change sequence data, and then these features can be passed as input to the subsequent isolation forest model for anomaly detection, thereby identifying the fault interval.

[0075] Acquiring a first influencing factor sequence according to the hydraulic equipment operation detection data; acquiring a second influencing factor sequence according to the first fault interval, including: acquiring abnormal data detected in the hydraulic equipment according to the hydraulic equipment operation detection data, and sorting the abnormal data according to the abnormal degree to obtain the first influencing factor sequence; acquiring abnormal data of hydraulic equipment fault according to the first fault interval, and sorting the abnormal data according to the abnormal degree to obtain the second influencing factor sequence;

[0076] Among them, the first influencing factor sequence is the missing data or lagged data in the hydraulic equipment operation detection data, for example, the missing of hydraulic equipment temperature data or the lag in the detection of hydraulic equipment temperature data; the second influencing factor sequence is the fault area inferred from the first fault interval.

[0077] Acquire a first influencing factor sequence according to the hydraulic equipment operation detection data; acquire a second influencing factor sequence according to the first fault interval, including: acquiring common influencing factor data in the first influencing factor sequence and the second influencing factor sequence according to the first influencing factor sequence and the second influencing factor sequence, sorting according to the common influencing factor data, and acquiring multiple influencing factor sequences according to the sorting result;

[0078] The common influencing factor data are data showing abnormalities in the first influencing factor sequence and data showing abnormalities in the second influencing factor data, and are sorted according to the common influencing factors. For example, if among three data, the first data is abnormal in the first influencing factor sequence, and the second and third data are normal, and the second data is abnormal in the second influencing factor sequence, and the first and third data are normal, then the common influencing factor data are the first data and the second data. If the first data is abnormal at the same time, the first data is the common influencing factor data. The abnormal data in the first influencing factor sequence and the second influencing factor sequence are unioned and then the abnormal influencing factors are calculated. All influencing factors are sorted, including sorting according to the degree of data abnormality. For example, if the abnormality degree of the first data in the first sequence is 50%, the abnormality degree of the second data is 30%, and the abnormality degree of the third data is 60%; the abnormality degree of the first data in the second sequence is 30%, the abnormality degree of the second data is 50%, and the abnormality degree of the third data is 20%, the abnormality degree in each sequence is added, wherein the total abnormality degree in the first sequence is 140%, and the abnormality degree in the second sequence is 100%, and the abnormality degree is only used for sorting.

[0079] Obtain multiple influencing factor sequences according to the sorting results, including: putting the multiple influencing factor sequences into the fault diagnosis model for correction, recording the correction gain of each influencing factor, dividing the correction gain into positive gain and negative gain according to the correction result, and obtaining the maximum influencing path after all sequences are corrected;

[0080] Among them, the positive gain and negative gain are: correcting the data in the sequence one by one, for example, in a sequence A, B, C, if A is abnormal data, then correct A, then simulate and run, then continue to correct B, then simulate and run, and finally correct C, then simulate and run, and record the recovery degree of the hydraulic equipment after each simulated run. If the hydraulic equipment recovers, it is a positive gain, otherwise it is a negative gain; the maximum impact path is after all data are corrected, if the correction range of A is 20%, the correction range of B is 30%, and the correction range of C is 10% under a certain path; if the correction range of A is -20%, the correction range of B is 30%, and the correction range of C is 10% under a certain path, the simulation operation effect of the hydraulic equipment under this path is the best, then it is the maximum impact path.

[0081] Obtaining the second fault interval according to the first influencing factor sequence and the second influencing factor sequence, including: after all sequences are corrected and the maximum influencing path is obtained, determining the second fault interval, and obtaining the faulty device in the first fault interval and the faulty device in the second fault interval according to the first fault interval and the second fault interval;

[0082] Among them, the method of obtaining the fault area of ​​the hydraulic equipment according to the first fault interval and the second fault interval includes: determining the first fault area according to the intersection area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, determining the second fault area according to the difference area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, and determining the fault troubleshooting sequence and maintenance suggestions according to the first fault area and the second fault area.

[0083] Embodiment 2: Based on Embodiment 1, a digital equipment fault detection system for building materials processing based on the Internet of Things includes:

[0084] The data acquisition and preprocessing module is used to obtain the equipment operation parameter data of the hydraulic equipment, the hydraulic equipment working environment parameter data, and the hydraulic equipment operation detection data and perform preprocessing, and then perform normalization, and divide the normalized data into a training set and a test set;

[0085] Temperature change trend module, used to obtain the trend of hydraulic oil in hydraulic equipment with temperature change;

[0086] The hydraulic oil estimated flow rate module is used to obtain the estimated flow rate of the hydraulic oil according to the trend of the hydraulic oil changing with temperature;

[0087] The support vector machine prediction model building module is used to combine the actual operation data of hydraulic equipment and use the support vector machine to build a prediction model, thereby more accurately predicting the actual flow rate;

[0088] Kalman filter fusion building block, used to use Kalman filter fusion model prediction and measured data to improve the accuracy and real-time performance of flow velocity estimation;

[0089] The simulation module is used to implement dynamic simulation in the simulation software to simulate the flow rate changes of hydraulic equipment during the startup phase, stable working conditions, variable working conditions and shutdown cooling;

[0090] Support vector machine fault classification module, used to classify faults;

[0091] LSTM model building module, used to analyze and process time series;

[0092] The isolation forest model builds a module, then uses the trained LSTM model to extract features from the data, and uses the output layer of the LSTM as input to the isolation forest model to identify anomalies based on the isolation forest model;

[0093] The fault diagnosis model building module is used to put multiple influencing factor sequences into the fault diagnosis model for correction, record the correction gain of each influencing factor, divide the correction gain into positive gain and negative gain according to the correction result, and obtain the maximum influencing path after all sequences are corrected.

[0094] The above is a detailed introduction to the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for fault detection of digital equipment for building materials processing based on the Internet of Things, characterized in that: The following steps are involved: S1: Acquire equipment operation parameter data of hydraulic equipment, hydraulic equipment working environment parameter data and hydraulic equipment operation detection data; S2: Obtain the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establish a coordinate system; S3: Obtaining the first fault interval according to the degree of disorder reflected by the coordinate system and the operation cycle data of the hydraulic equipment; S4: obtaining a first influencing factor sequence according to the hydraulic equipment operation detection data; obtaining a second influencing factor sequence according to the first fault interval; S5: Obtain a second fault interval according to the first influencing factor sequence and the second influencing factor sequence; S6: Obtaining a fault area of ​​the hydraulic equipment according to the first fault area and the second fault area.

2. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 1 is characterized in that: The method of obtaining the flow rate change of the hydraulic oil in the equipment according to the equipment operation parameter data and the equipment operation environment parameter data and establishing a coordinate system includes: obtaining the viscosity change trend of the hydraulic oil with temperature during the preheating operation of the hydraulic equipment, and obtaining the estimated flow rate of the hydraulic oil according to the viscosity change trend of the hydraulic oil with temperature; Among them, the viscosity change trend of the hydraulic oil with temperature is obtained according to the viscosity model equation. The viscosity model equation formula is: ,in, Indicates the kinematic viscosity of the liquid, the unit is usually cSt, i.e. centistokes, T is the absolute temperature, the unit is Kelvin K, A, B, C, D are constants determined according to the characteristics of the specific liquid, It is a reference temperature, a standard test temperature for the liquid is selected.

3. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 2 is characterized in that: Further including: The actual flow rate of the hydraulic oil is obtained based on the predicted flow rate of the hydraulic oil and the operating parameter data of the hydraulic equipment. The flow rate of the hydraulic oil in the hydraulic equipment is obtained based on the actual flow rate of the hydraulic oil. The flow rate changes of the hydraulic equipment in the startup stage, stable working conditions, variable working conditions and after shutdown and cooling are calculated based on the flow rate of the hydraulic oil in the hydraulic equipment. The expected flow rate, equipment geometry, oil properties and hydraulic equipment operating parameters are input into the flow rate variation model to perform flow field analysis to predict the actual flow rate distribution of the hydraulic oil in the system. Combined with the actual operating data of the hydraulic equipment, a prediction model is established using a support vector machine to more accurately predict the actual flow rate. Apply system identification technology, combine with online sensor data, adjust model parameters in real time, and realize online estimation of the actual flow rate of hydraulic oil. Use Kalman filter to fuse model prediction and measured data to improve the accuracy and real-time performance of flow rate estimation. Implement dynamic simulation in simulation software to simulate the flow rate changes of hydraulic equipment during startup, stable working conditions, variable working conditions, and shutdown cooling.

4. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 3 is characterized in that: The method of calculating the flow rate change of the hydraulic equipment during the startup phase, stable working condition, variable working condition and after shutdown and cooling according to the flow rate of the hydraulic oil in the hydraulic equipment comprises: establishing a coordinate system in the hydraulic equipment according to the flow rate change, reflecting the degree of dispersion of the flow rate change in the hydraulic equipment according to the coordinate system, and inferring the fault according to the degree of dispersion of the flow rate change; Among them, a time-space coordinate system is established based on the physical structure of the hydraulic equipment and the fluid flow path; whether the coordinate system is a two-dimensional plane coordinate system or a three-dimensional space coordinate system is determined according to the complexity of the hydraulic system; each coordinate point represents the position and time information of a monitoring point in the system; the flow rate data of each monitoring point is mapped to the above coordinate system to form a flow rate distribution diagram; a statistical method is used to quantify the degree of dispersion of flow rate changes and identify the characteristics of flow rate fluctuations; a fault classification model is established using support vector machine SVM; the training data set should contain the flow rate change characteristics under normal operation and known fault conditions; when it is detected that the dispersion of flow rate changes exceeds the preset threshold, a fault warning is automatically triggered, and maintenance personnel are notified in time to conduct inspection and intervention.

5. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 4 is characterized in that: The obtaining of the first fault interval according to the degree of disorder reflected by the coordinate system combined with the operating cycle data of the hydraulic equipment includes: inferring the fault according to the degree of disorder combined with the operating cycle, frequency and load change data of the hydraulic equipment to obtain the first fault interval; First, the LSTM model is constructed, and then the trained LSTM model is used to extract features from the data. The output layer of the LSTM is used as input to the isolation forest model. Based on the anomalies identified by the isolation forest model, the dense area where faults occur is identified in the time series, which is the fault interval. The output layer formula of the LSTM is: ,in, is the sigmoid activation function, , is the weight matrix, is the bias term, is the current input, is the hidden state of the previous time step, and the forget gate formula is , the cell state formula is ,here, represents the element-wise product, is the hyperbolic tangent activation function, which is used to generate new candidate memories. The output gate formula is , the hidden state formula is ,LSTM can learn features from historical operating cycle, frequency, and load change sequence data, and then these features can be passed as input to the subsequent isolation forest model for anomaly detection, thereby identifying the fault interval.

6. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 1 is characterized in that: The first influencing factor sequence is obtained according to the hydraulic equipment operation detection data; Acquiring a second influencing factor sequence according to the first fault interval includes: acquiring abnormal data detected in the hydraulic equipment according to the hydraulic equipment operation detection data, and sorting the abnormal data according to the abnormality degree to obtain the first influencing factor sequence; According to the first fault interval, abnormal data of hydraulic equipment fault is obtained, and the abnormal data is sorted according to the abnormal degree to obtain a second influencing factor sequence.

7. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 1 is characterized in that: The first influencing factor sequence is obtained according to the hydraulic equipment operation detection data; Obtaining a second influencing factor sequence according to the first fault interval includes: obtaining common influencing factor data in the first influencing factor sequence and the second influencing factor sequence according to the first influencing factor sequence and the second influencing factor sequence, sorting according to the common influencing factor data, and obtaining multiple influencing factor sequences according to the sorting result.

8. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 7 is characterized in that: The method of obtaining multiple influencing factor sequences according to the sorting results includes: placing the multiple influencing factor sequences into a fault diagnosis model for correction, recording the correction gain of each influencing factor, dividing the correction gain into positive gain and negative gain according to the correction result, and obtaining the maximum influencing path after all sequences are corrected.

9. The method for fault detection of building materials processing digital equipment based on the Internet of Things according to claim 8, characterized in that: The obtaining of the second fault interval according to the first impact factor sequence and the second impact factor sequence includes: after all sequences are corrected and the maximum impact path is obtained, the second fault interval is determined, and according to the first fault interval and the second fault interval, the faulty device in the first fault interval and the faulty device in the second fault interval are obtained; Among them, the method of obtaining the fault area of ​​the hydraulic equipment according to the first fault interval and the second fault interval includes: determining the first fault area according to the intersection area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, determining the second fault area according to the difference area of ​​the fault equipment in the first fault interval and the fault equipment in the second fault interval, and determining the fault troubleshooting sequence and maintenance suggestions according to the first fault area and the second fault area.

10. The digital equipment fault detection system for building materials processing based on the Internet of Things is characterized by: Also includes: Data acquisition and preprocessing module: used to obtain equipment operation parameter data of hydraulic equipment, hydraulic equipment operating environment parameter data and hydraulic equipment operation detection data and perform preprocessing, then normalize and divide the normalized data into training set and test set; Temperature change trend module: used to obtain the trend of hydraulic oil in hydraulic equipment with temperature change; Hydraulic oil estimated flow rate module: used to obtain the estimated flow rate of hydraulic oil according to the trend of hydraulic oil changing with temperature; Support vector machine prediction model building module: used to combine the actual operation data of hydraulic equipment and use support vector machine to build a prediction model, so as to more accurately predict the actual flow rate; Kalman filter fusion building block: used to use Kalman filter fusion model prediction and measured data to improve the accuracy and real-time performance of flow velocity estimation; Simulation module: used to implement dynamic simulation in the simulation software to simulate the flow rate changes of hydraulic equipment during the startup phase, stable working conditions, variable working conditions and shutdown cooling; Support vector machine fault classification module: used to classify faults; LSTM model building module: used to analyze and process time series; Isolation forest model construction module: Then use the trained LSTM model to extract features from the data, and use the output layer of LSTM as input to the isolation forest model to identify abnormal points based on the isolation forest model; Fault diagnosis model building module: used to put multiple influencing factor sequences into the fault diagnosis model for correction, record the correction gain of each influencing factor, divide the correction gain into positive gain and negative gain according to the correction result, and obtain the maximum influencing path after all sequences are corrected.

Images