A data parallel processing system and method for electric wheel reducer of mine car

Abstract

The present invention provides a data parallel processing system for an electric wheel reducer of a mine car, comprising a field-level device connected to the mine car reducer, a working condition-level device connected to the field-level device via a network cable, and a cloud processing device connected to the working condition-level device via a local area network; the field-level device comprises an electric wheel temperature-vibration integrated sensor and a main generator temperature-vibration integrated sensor arranged on the mine car, a data acquisition cabinet connected to the sensor via wireless communication transmission, and an online monitoring and diagnosis platform connected to the data acquisition cabinet via a network cable; the working condition-level device comprises a number of data processing servers connected to the data acquisition cabinet and the online monitoring and diagnosis platform via a local area network; the cloud processing device comprises a cloud platform connected to the online monitoring and diagnosis platform and the data processing server. The present invention can reasonably distribute the data collected by each monitoring point, timely process suspicious data, and more effectively discover the possible fault risks in the operation process of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of electric wheel reducer fault diagnosis data processing, and in particular to a data parallel processing system and method for an electric wheel reducer of a mine car. Background Art

[0002] As one of the most important assemblies of a mine car, the electric wheel has a high cost and a high cost for major repairs. The electric wheel reducer is currently monitored by offline regular testing, which has a large gap period. If the major repair is not timely, it is easy to cause serious mechanical accidents such as bearing damage and tooth breaking, resulting in large losses. For safety reasons, the major repair cycle is shortened, which increases costs.

[0003] Parallel processing is a computing method in which two or more processes can be performed simultaneously in a computer system. Parallel processing allows working on different aspects of the same program at the same time. The main purpose of parallel processing is to save time in solving large and complex problems. In order to use parallel processing, the program must first be parallelized, that is, the work is divided into different processing processes (threads). Parallel processing cannot be achieved automatically because of interrelated problems. In addition, parallelism does not guarantee speedup.

[0004] Application number CN202410029485.0 discloses a cloud-based elevator data parallel processing method and system, the method comprising: collecting elevator data of an elevator; screening valid data in the elevator data, analyzing the priority of the valid data, and converting the valid data into a data queue using the priority; constructing a pre-trained parallel elevator warning model for the data queue, calculating the warning throughput of the elevator warning value, and training the pre-trained parallel elevator warning model according to the warning throughput to obtain a trained parallel elevator warning model; transmitting the data queue in parallel to the cloud, decomposing the trained parallel elevator warning model into multiple parallel data processing structures in the cloud, and storing the data queue in a preset high-speed cache based on the multiple parallel data processing structures; performing data parallel processing on the data queue in the cache using multiple parallel data processing structures to obtain ultra-high-speed cloud processing results of the elevator data. Summary of the invention

[0005] The present invention aims to overcome the technical problem in the prior art that the amount of data processing in the online monitoring system is too large, resulting in untimely data processing and thus affecting the efficiency of equipment fault monitoring. A data parallel processing system and method for the electric wheel reducer of a mine car is provided, which reasonably distributes the data collected at each monitoring point and processes suspicious data in a timely manner, so as to more effectively discover the possible fault risks during the operation of the equipment.

[0006] The present invention provides a data parallel processing system for a mine car electric wheel reducer, comprising a field-level device connected to the mine car reducer, a working condition-level device connected to the field-level device via a network cable, and a cloud processing device connected to the working condition-level device via a local area network;

[0007] The field-level equipment includes an electric wheel temperature-vibration integrated sensor and a main generator temperature-vibration integrated sensor installed on the mine car, a data acquisition cabinet connected to the electric wheel temperature-vibration integrated sensor and the main generator temperature-vibration integrated sensor via wireless communication transmission, and an online monitoring and diagnosis platform connected to the data acquisition cabinet via a network cable;

[0008] The working condition level equipment includes several data processing servers connected to the data acquisition cabinet and the online monitoring and diagnosis platform through a local area network;

[0009] The cloud processing equipment includes a cloud platform connected to the online monitoring and diagnosis platform and the data processing server, and a remote monitoring center and an intelligent operation and maintenance center connected to the cloud platform;

[0010] The field-level device is used to collect raw data A, obtain raw data B after filtering raw data A, obtain operation status data C after data processing raw data B, and transmit the status data C to the working condition-level device, which is used to receive the processed data D transmitted by the working condition-level device, and display the operation status data of the electric wheel reducer according to the processed data D, and transmit the operation status data to the cloud processing device;

[0011] The working condition level equipment is used to aggregate, classify and process the status data C to obtain the processed data D, and then transmit the processed data D to the online monitoring and diagnosis platform and the cloud processing equipment;

[0012] The cloud processing device is used to aggregate and store the processed data D and the operating status data, to perform secondary processing on the processed data D to obtain processed data E, to return the processed data E to the corresponding working condition level device, and to distribute the processed data D that needs to be secondary processed to the idle working condition level device;

[0013] For each of the original data A, original data B, state data C, data processing D and data processing E, a data doubt value and a fault prediction value are attached;

[0014] The online monitoring platform processes the original data A as follows:

[0015] Analyze the data doubt value of the original data A in real time, adjust the data processing priority of the original data A with the doubt value according to the size of the data doubt value, and directly send the original data A with a priority level exceeding the set level to the cloud processing device for data processing;

[0016] When the fault prediction value of the original data A is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0017] When the fault prediction value and the data doubt value of the original data A are both less than the set threshold, the original data A is filtered to obtain the original data B, and then the original data B is processed to obtain the operating status data C, and the operating status data C is transmitted to the working condition level equipment;

[0018] If the fault prediction value of the original data B or the operating status data C is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0019] If the data doubt value of the original data B or the operation status data C is greater than the set threshold, the processing priority of the corresponding original data B or the operation status data C is adjusted, and the original data B or the operation status data C with a priority level exceeding the set level is directly sent to the cloud processing device for data processing;

[0020] The processing of the operating status data C by the working condition level equipment includes:

[0021] According to the category of the running status data C, a data processing model is selected to process the data to obtain processed data D;

[0022] If the fault prediction value of the processed data D is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0023] If the data doubt value of the processed data D is greater than the set threshold, the processing priority of the corresponding processed data D is adjusted, and the processed data D with a priority level exceeding the set level is directly sent to the cloud processing device for data processing;

[0024] When the fault prediction value and data doubt value of the original data A are both less than the set threshold, the processed data D is transmitted to the online monitoring and diagnosis platform and the cloud processing device;

[0025] If the running status data C to be processed exceeds the set threshold, the running status data C is forwarded to the adjacent working condition level device or cloud processing platform;

[0026] The data processing process of the cloud processing platform includes:

[0027] Used to summarize the original data A, original data B, operating status data C, and processed data D transmitted by each working level equipment and online monitoring platform;

[0028] For the processed data D that needs secondary processing, the corresponding data processing model is retrieved according to the data category to process the processed data D to obtain the processed data E;

[0029] For original data A, original data B, operating status data C, processed data D and processed data E whose data doubt values ​​are greater than the set threshold, the corresponding data processing model is called according to the data category, and fault diagnosis is performed on the corresponding data. If the diagnosis result is normal, the data doubt value of the corresponding data is adjusted. If the diagnosis result is a fault, an alarm is issued and the alarm information is transmitted to the online monitoring platform of the data source.

[0030] The data parallel processing system for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the original data A includes the speed data, vibration data and ambient temperature data of the electric wheel of the mine car and the time data corresponding to the above data, the speed data includes the real-time speed, the deviation value between the real-time speed and the set speed, the detection point stroke, and the detection point stroke deviation value, the vibration data includes the vibration frequency, amplitude, phase, time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, and energy spectrum, the time domain characteristics include variance, mean square value, effective value, skewness and kurtosis, peak index, margin index, waveform index, and pulse index, and the frequency domain characteristics include FFT spectrum, power spectrum, and envelope spectrum.

[0031] The present invention provides a data parallel processing method for a mine car electric wheel reducer, comprising the following steps:

[0032] S1, field-level equipment collects raw data A;

[0033] S2, the online monitoring and diagnosis platform processes the original data A, and determines whether the fault prediction value of the original data A is greater than the set threshold value. If it is determined to be yes, the process proceeds to step S11, and if it is determined to be no, the process proceeds to step S3;

[0034] S3, determine whether the data doubt value of the original data A is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined not, proceed to step S4;

[0035] S4. The online monitoring and diagnosis platform filters the original data to obtain the original data B, processes the original data B to obtain the operation status data C, and transmits the operation status data C to the working condition level equipment;

[0036] S5, the working condition level equipment sorts the state data C from large to small according to the data doubt value, processes the state data C in the sorting order, obtains the processed data D, and determines whether the fault prediction value of the processed data D is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S6;

[0037] S6, determine whether the data in doubt value of the processed data D is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined to be no, proceed to step S7;

[0038] S7, the cloud processing platform receives the processed data D, and sorts the processed data D that needs to be processed again from large to small according to the data doubt value, processes the processed data D in the sorting order, obtains the processed data E, and determines whether the fault prediction value of the processed data E is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S8;

[0039] S8, determine whether the data doubt value of the processed data E is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined to be no, proceed to step S9;

[0040] S9, the data parallel processing system operates normally;

[0041] S10, the original data A / processed data D / processed data E are transmitted to the cloud processing platform for fault diagnosis. If the diagnosis result is normal, the process returns to step S4; if the diagnosis result is an alarm, the process goes to step S11;

[0042] S11, fault alarm.

[0043] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the data doubt value S:

[0044] S = f(t) × f(β, E)

[0045] Among them, f(t) is the time constant curve, f(β,E) is the function related to the measured data, E is the measured data value, and β is the deviation coefficient between the measured data and the theoretical data.

[0046] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred mode, in step S5, if the number of queued status data C to be processed exceeds a set threshold, the working condition level device forwards the operating status data C to an adjacent working condition level device or a cloud processing platform;

[0047] After receiving the forwarded operating status data C, the working condition-level device adds it to the local sorting sequence of pending status data C according to its data doubt value;

[0048] After receiving the forwarded operating status data C, the cloud processing platform forwards it to the idle working condition level device for data processing.

[0049] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred method, original data A, original data B, operating status data C, processed data D and processed data E whose data doubt value is less than a set threshold are used to train the corresponding data processing model.

[0050] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the original data B is the original data A excluding abnormal data points:

[0051] B=A-α

[0052] Among them, α is the abnormal data point data;

[0053] The operating status data C includes the temperature effect correction value and time effect correction value of the same type of data in the original data B:

[0054] C = f(T, t)B

[0055] Where f(T,t) is a function related to temperature T and time t.

[0056] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the processed data D includes associated data values ​​with different types of operating status data C:

[0057] D=D1∪D2

[0058] D1=d1×C

[0059]

[0060] Among them, D1 is the processed data linearly related to the running status data C, D2 is the processed data associated with different categories of data in the running status data C, and C i , C j is a subset of the operating status data C, and d1 and d2 are associated parameters.

[0061] The data parallel processing method for the electric wheel reducer of a mine car described in the present invention is, as a preferred embodiment, the processed data E includes the associated data values ​​with the processed data D of different monitoring devices:

[0062] E=E1∪E2

[0063] E1=e1×D

[0064]

[0065] Among them, E1 is the processed data linearly related to the processed data D, E2 is the processed data associated with the data collected by different monitoring devices in the processed data D, and D i , D j is a subset of the processed data D, and e1 and e2 are associated parameters.

[0066] (1) The present invention assigns data doubt values ​​to all data (including original collected data and data processed in the intermediate process), assigns different priorities according to different data doubt values, and processes data in descending order of priority, thereby being able to discover the risk of equipment operation failure as early as possible.

[0067] (2) The present invention trains different data models and processes suspicious data in parallel, which can find the risk points of equipment operation in the fastest way and improve the efficiency of fault prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a composition diagram of a data parallel processing system for electric wheel reducer of a mine car;

[0069] Figure 2 A schematic diagram of the location of sensors installed on the outer side of the electric wheel of a data parallel processing system for an electric wheel reducer of a mine car;

[0070] Figure 3 A schematic diagram of the position of the sensor installed on the inner side of the electric wheel of a data parallel processing system for an electric wheel reducer of a mine car;

[0071] Figure 4 A schematic diagram of the sensor location for the main generator of a data parallel processing system for an electric wheel reducer of a mine car;

[0072] Figure 5 The present invention is a flow chart of a data parallel processing method for an electric wheel reducer of a mine car. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0074] Example 1

[0075] like Figure 1 As shown, a data parallel processing system for a mine car electric wheel reducer includes a field-level device connected to the mine car reducer, a working condition-level device connected to the field-level device via a network cable, and a cloud processing device connected to the working condition-level device via a local area network;

[0076] The field-level equipment includes an electric wheel temperature-vibration integrated sensor and a main generator temperature-vibration integrated sensor installed on the mine car, a data acquisition cabinet connected to the electric wheel temperature-vibration integrated sensor and the main generator temperature-vibration integrated sensor via wireless communication, and an online monitoring and diagnosis platform connected to the data acquisition cabinet via a network cable; Figures 2 to 4As shown, each mining car includes at least 3 collection points, which are respectively arranged on the outside of the motor, the inside of the motor (on the gear box) and the main generator;

[0077] The working condition level equipment includes several data processing servers connected to the data acquisition cabinet and the online monitoring and diagnosis platform through a local area network;

[0078] The cloud processing equipment includes a cloud platform connected to the online monitoring and diagnosis platform and the data processing server, and a remote monitoring center and an intelligent operation and maintenance center connected to the cloud platform;

[0079] The field-level equipment is used to collect the original data A, filter the original data A to obtain the original data B, and then process the original data B to obtain the operating status data C, and transmit the status data C to the working condition-level equipment, which is used to receive the processed data D transmitted by the working condition-level equipment, and display the operating status data of the electric wheel reducer according to the processed data D, and transmit the operating status data to the cloud processing equipment; the original data A includes the speed data, vibration data and ambient temperature data of the electric wheel of the mine car and the time data corresponding to the above data. The speed data includes the real-time speed, the deviation value between the real-time speed and the set speed, the detection point stroke, and the detection point stroke deviation value. The vibration data includes vibration frequency, amplitude, phase, time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, and energy spectrum. The time domain characteristics include variance, mean square value, effective value, skewness and kurtosis, peak index, margin index, waveform index, and pulse index. The frequency domain characteristics include FFT spectrum, power spectrum, and envelope spectrum.

[0080] The original data B is the original data A with the abnormal data points excluded:

[0081] B=A-α

[0082] Among them, α is the abnormal data point data;

[0083] The operating status data C includes the temperature effect correction value and time effect correction value of the same type of data in the original data B:

[0084] C = f(T, t)B

[0085] Where f(T,t) is a function related to temperature T and time t;

[0086] The processing data D includes data values ​​associated with different types of operating status data C:

[0087] D=D1∪D2

[0088] D1=d1×C

[0089]

[0090] Among them, D1 is the processed data linearly related to the running status data C, D2 is the processed data associated with different categories of data in the running status data C, and C i , C j is a subset of the running status data C, d1 and d2 are associated parameters;

[0091] The working condition level equipment is used to aggregate, classify and process the status data C to obtain the processed data D, and then transmit the processed data D to the online monitoring and diagnosis platform and the cloud processing equipment;

[0092] The cloud processing device is used to aggregate and store the processed data D and the operating status data, to perform secondary processing on the processed data D to obtain processed data E, to return the processed data E to the corresponding working condition level device, and to distribute the processed data D that needs to be secondary processed to the idle working condition level device;

[0093] Processing data E includes data values ​​associated with processing data D of different monitoring devices:

[0094] E=E1∪E2

[0095] E1=e1×D

[0096]

[0097] Among them, E1 is the processed data linearly related to the processed data D, E2 is the processed data associated with the data collected by different monitoring devices in the processed data D, and D i , D j is a subset of the processed data D, e1 and e2 are associated parameters;

[0098] For each of the original data A, original data B, state data C, data processing D and data processing E, a data doubt value and a fault prediction value are attached;

[0099] Data doubt value S:

[0100] S = f(t) × f(β, E)

[0101] Where, f(t) is the time constant curve, f(β,E) is the function related to the measured data, E is the measured data value, and β is the deviation coefficient between the measured data and the theoretical data;

[0102] The original data A, original data B, operation status data C, processed data D and processed data E whose data doubt value is less than the set threshold are used to train the corresponding data processing model;

[0103] The online monitoring platform processes the original data A as follows:

[0104] Analyze the data doubt value of the original data A in real time, adjust the data processing priority of the original data A with the doubt value according to the size of the data doubt value, and directly send the original data A with a priority level exceeding the set level to the cloud processing device for data processing;

[0105] When the fault prediction value of the original data A is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0106] When the fault prediction value and the data doubt value of the original data A are both less than the set threshold, the original data A is filtered to obtain the original data B, and then the original data B is processed to obtain the operating status data C, and the operating status data C is transmitted to the working condition level equipment;

[0107] If the fault prediction value of the original data B or the operating status data C is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0108] If the data doubt value of the original data B or the operation status data C is greater than the set threshold, the processing priority of the corresponding original data B or the operation status data C is adjusted, and the original data B or the operation status data C with a priority level exceeding the set level is directly sent to the cloud processing device for data processing;

[0109] The processing of the operating status data C by the working condition level equipment includes:

[0110] According to the category of the running status data C, a data processing model is selected to process the data to obtain processed data D;

[0111] If the fault prediction value of the processed data D is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device;

[0112] If the data doubt value of the processed data D is greater than the set threshold, the processing priority of the corresponding processed data D is adjusted, and the processed data D with a priority level exceeding the set level is directly sent to the cloud processing device for data processing;

[0113] When the fault prediction value and data doubt value of the original data A are both less than the set threshold, the processed data D is transmitted to the online monitoring and diagnosis platform and the cloud processing device;

[0114] If the running status data C to be processed exceeds the set threshold, the running status data C is forwarded to the adjacent working condition level device or cloud processing platform;

[0115] The data processing process of the cloud processing platform includes:

[0116] Used to summarize the original data A, original data B, operating status data C, and processed data D transmitted by each working level equipment and online monitoring platform;

[0117] For the processed data D that needs secondary processing, the corresponding data processing model is retrieved according to the data category to process the processed data D to obtain the processed data E;

[0118] For original data A, original data B, operating status data C, processed data D and processed data E whose data doubt values ​​are greater than the set threshold, the corresponding data processing model is called according to the data category, and fault diagnosis is performed on the corresponding data. If the diagnosis result is normal, the data doubt value of the corresponding data is adjusted. If the diagnosis result is a fault, an alarm is issued and the alarm information is transmitted to the online monitoring platform of the data source.

[0119] The data parallel processing method for the electric wheel reducer of a mine car in this embodiment includes the following steps:

[0120] S1, field-level equipment collects raw data A;

[0121] S2, the online monitoring and diagnosis platform processes the original data A, and determines whether the fault prediction value of the original data A is greater than the set threshold value. If it is determined to be yes, the process proceeds to step S11, and if it is determined to be no, the process proceeds to step S3;

[0122] S3, determine whether the data doubt value of the original data A is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined not, proceed to step S4;

[0123] S4. The online monitoring and diagnosis platform filters the original data to obtain the original data B, processes the original data B to obtain the operation status data C, and transmits the operation status data C to the working condition level equipment;

[0124] S5, the working condition level equipment sorts the state data C from large to small according to the data doubt value, processes the state data C in the sorting order, obtains the processed data D, and determines whether the fault prediction value of the processed data D is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S6;

[0125] S6, determine whether the data in doubt value of the processed data D is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined to be no, proceed to step S7;

[0126] S7, the cloud processing platform receives the processed data D, and sorts the processed data D that needs to be processed again from large to small according to the data doubt value, processes the processed data D in the sorting order, obtains the processed data E, and determines whether the fault prediction value of the processed data E is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S8;

[0127] S8, determine whether the data doubt value of the processed data E is greater than the set threshold value, if it is determined to be yes, proceed to step S10, if it is determined to be no, proceed to step S9;

[0128] S9, the data parallel processing system operates normally;

[0129] S10, the original data A / processed data D / processed data E are transmitted to the cloud processing platform for fault diagnosis. If the diagnosis result is normal, the process returns to step S4; if the diagnosis result is an alarm, the process goes to step S11;

[0130] S11, fault alarm.

[0131] Example 2

[0132] During the data processing in step S5, if the number of status data C to be processed in the queue exceeds the set threshold, the operating state data C is forwarded by the operating state level device to the adjacent operating state level device or the cloud processing platform;

[0133] After receiving the forwarded operating status data C, the working condition-level device adds it to the local sorting sequence of pending status data C according to its data doubt value;

[0134] After receiving the forwarded operating status data C, the cloud processing platform forwards it to the idle working condition level device for data processing.

[0135] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A data parallel processing system for a mine car electric wheel reducer, characterized in that: include: A field-level device connected to the mine car reducer, a working condition-level device connected to the field-level device via a network cable, and a cloud processing device connected to the working condition-level device via a local area network; The field-level equipment includes an electric wheel temperature-vibration integrated sensor and a main generator temperature-vibration integrated sensor arranged on the mine car, a data acquisition cabinet connected to the electric wheel temperature-vibration integrated sensor and the main generator temperature-vibration integrated sensor via wireless communication transmission, and an online monitoring and diagnosis platform connected to the data acquisition cabinet via a network cable; The working condition level equipment includes a number of data processing servers connected to the data acquisition cabinet and the online monitoring and diagnosis platform via a local area network; The cloud processing device includes a cloud platform connected to the online monitoring and diagnosis platform and the data processing server, and a remote monitoring center and an intelligent operation and maintenance center connected to the cloud platform; The field-level device is used to collect raw data A, obtain raw data B after filtering the raw data A, obtain operation status data C after data processing the raw data B, and transmit the status data C to the working condition-level device, and is used to receive the processed data D transmitted by the working condition-level device, and display the operation status data of the electric wheel reducer according to the processed data D, and is used to transmit the operation status data to the cloud processing device; The working condition level device is used to aggregate, classify and process the status data C to obtain processed data D, and then transmit the processed data D to the online monitoring and diagnosis platform and the cloud processing device; The cloud processing device is used to aggregate and store the processed data D and the operating status data, to perform secondary processing on the processed data D to obtain processed data E, to return the processed data E to the corresponding working condition level device, and to distribute the processed data D that needs to be secondary processed to the idle working condition level device; A data doubt value and a fault prediction value are attached to each of the original data A, the original data B, the state data C, the data processing D and the data processing E; The processing of the raw data A by the online monitoring platform includes: Analyze the data doubt value of the original data A in real time, adjust the data processing priority of the original data A with the data doubt value according to the size of the data doubt value, and directly send the original data A with a priority level exceeding the set level to the cloud processing device for data processing; When the fault prediction value of the original data A is greater than a set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device; When the fault prediction value and the data doubt value of the original data A are both less than a set threshold, the original data A is filtered to obtain the original data B, and then the original data B is processed to obtain the operating status data C, and the operating status data C is transmitted to the working condition level device; If the fault prediction value of the original data B or the operating status data C is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device; If the data doubt value of the original data B or the operating status data C is greater than a set threshold, the processing priority of the corresponding original data B or the operating status data C is adjusted, and the original data B or the operating status data C whose priority level exceeds the set level is directly sent to the cloud processing device for data processing; The processing of the operating status data C by the operating condition level device includes: Selecting a data processing model according to the category of the running status data C to process the data, and obtaining the processed data D; If the fault prediction value of the processed data D is greater than the set threshold, no subsequent data processing is performed, and an alarm is directly issued and the alarm information is transmitted to the cloud processing device; If the data doubt value of the processed data D is greater than a set threshold, the processing priority of the corresponding processed data D is adjusted, and the processed data D with a priority level exceeding the set level is directly sent to the cloud processing device for data processing; When the fault prediction value and the data doubt value of the original data A are both less than a set threshold, the processed data D is transmitted to the online monitoring and diagnosis platform and the cloud processing device; If the running status data C to be processed exceeds a set threshold, forwarding the running status data C to the adjacent working condition level device or the cloud processing platform; The data processing process of the cloud processing platform includes: Used to summarize the original data A, the original data B, the operating status data C, and the processed data D transmitted by each of the working condition level devices and the online monitoring platform; For the processed data D that needs secondary processing, the corresponding data processing model is retrieved according to the data category to perform data processing on the processed data D to obtain the processed data E; For the original data A, the original data B, the operating status data C, the processed data D and the processed data E whose data doubt values ​​are greater than the set threshold, the corresponding data processing model is called according to the data category, and fault diagnosis is performed on the corresponding data. If the diagnosis result is normal, the data doubt value of the corresponding data is adjusted. If the diagnosis result is a fault, an alarm is issued and the alarm information is transmitted to the online monitoring platform of the data source.

2. A data parallel processing system for a mine car electric wheel reducer according to claim 1, characterized in that: The original data A includes the speed data, vibration data and ambient temperature data of the electric wheel of the mine car and the time data corresponding to the above data. The speed data includes the real-time speed, the deviation value between the real-time speed and the set speed, the detection point stroke, and the detection point stroke deviation value. The vibration data includes vibration frequency, amplitude, phase, time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, and energy spectrum. The time domain characteristics include variance, mean square value, effective value, skewness and kurtosis, peak index, margin index, waveform index, and pulse index. The frequency domain characteristics include FFT spectrum, power spectrum, and envelope spectrum.

3. A data parallel processing method for a mine car electric wheel reducer according to any one of claims 1 to 2, characterized in that: The following steps are involved: S1, the field-level device collects the original data A; S2, the online monitoring and diagnosis platform processes the original data A to determine whether the fault prediction value of the original data A is greater than a set threshold, if it is determined to be yes, proceed to step S11, if it is determined to be no, proceed to step S3; S3, determining whether the data doubt value of the original data A is greater than a set threshold value, if it is determined to be yes, proceeding to step S10, if it is determined to be no, proceeding to step S4; S4, the online monitoring and diagnosis platform filters the original data to obtain the original data B, processes the original data B to obtain the operating status data C, and transmits the operating status data C to the working condition level device; S5, the working condition level device sorts the state data C from large to small according to the data doubt value, processes the state data C in the sorting order, obtains the processed data D, and determines whether the fault prediction value of the processed data D is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S6; S6, determining whether the data doubt value of the processed data D is greater than a set threshold, if it is determined to be yes, proceeding to step S10, if it is determined to be no, proceeding to step S7; S7, the cloud processing platform receives the processed data D, and sorts the processed data D that needs to be processed again from large to small according to the data doubt value, processes the processed data D in the sorting order, obtains the processed data E, and determines whether the fault prediction value of the processed data E is greater than the set threshold value, if it is determined to be yes, proceeds to step S11, if it is determined to be no, proceeds to step S8; S8, determining whether the data doubt value of the processed data E is greater than a set threshold value, if it is determined to be yes, proceeding to step S10, if it is determined to be no, proceeding to step S9; S9, the data parallel processing system operates normally; S10, the original data A / the processed data D / the processed data E are transmitted to the cloud processing platform for fault diagnosis. If the diagnosis result is normal, the process returns to step S4; if the diagnosis result is an alarm, the process goes to step S11; S11, fault alarm.

4. A data parallel processing method for a mine car electric wheel reducer according to claim 3, characterized in that: The data in doubt value S: S = f(t) × f(β, E) Among them, f(t) is the time constant curve, f(β,E) is the function related to the measured data, E is the measured data value, and β is the deviation coefficient between the measured data and the theoretical data.

5. The data parallel processing method for the electric wheel reducer of a mine car according to claim 4, characterized in that: In step S5, if the number of the status data C to be processed in the queue exceeds a set threshold, the operating state level device forwards the operating state data C to an adjacent operating state level device or the cloud processing platform; After receiving the forwarded operation status data C, the working condition level device adds the data to the local sorting sequence of the status data C to be processed according to the data doubt value; After receiving the forwarded operating status data C, the cloud processing platform forwards it to the idle working condition-level device for data processing.

6. A data parallel processing method for a mine car electric wheel reducer according to claim 4, characterized in that: The original data A, the original data B, the operating status data C, the processed data D and the processed data E whose data doubt values ​​are less than a set threshold are used to train the corresponding data processing models.

7. A data parallel processing method for a mine car electric wheel reducer according to claim 4, characterized in that: The original data B is the original data A excluding abnormal data points: B=A-α Among them, α is the abnormal data point data; The operating status data C includes the temperature effect correction value and the time effect correction value of the same type of data in the original data B: C = f(T, t)B Where f(T,t) is a function related to temperature T and time t.

8. A data parallel processing method for a mine car electric wheel reducer according to claim 7, characterized in that: The processing data D includes data values ​​associated with different types of the operating status data C: D=D1∪D2 D1=d1×C Wherein, D1 is the processed data linearly related to the running status data C, D2 is the processed data associated with different categories of data in the running status data C, and C i , C j is a subset of the operating status data C, and d1 and d2 are associated parameters.

9. A data parallel processing method for a mine car electric wheel reducer according to claim 8, characterized in that: The processing data E includes data values ​​associated with the processing data D of different monitoring devices: E=E1∪E2 E1=e1×D Wherein, E1 is the processed data linearly related to the processed data D, E2 is the processed data associated with the data collected by different monitoring devices in the processed data D, and D i , D j is a subset of the processed data D, and e1 and e2 are associated parameters.

Images