Instrument and equipment fault detection system based on big data technology

Abstract

The present invention discloses an instrument equipment fault detection system based on big data technology, which relates to the technical field of intelligent management, including an acquisition module, an analysis module and an early warning module, which performs a first analysis on the information of a sensor sub-device to obtain a first fault, performs a second analysis on the information of a transmission sub-device to obtain a second fault, performs a third analysis on the information of a power sub-device to obtain a third fault, performs a fourth analysis on the information of an execution sub-device to obtain a fourth fault, constructs a transfer probability matrix, and predicts and warns the fault of the instrument equipment. The present invention uses the acquisition module to timely discover potential faults, conducts detailed analysis on various sub-devices, ensures the accuracy of fault discrimination, constructs a transfer probability matrix, uses historical data to predict future faults, and improves the reliability of the equipment. The early warning module sends out multi-level early warning signals, which significantly improves the intelligence and automation level of equipment fault detection.

Description

Technical Field

[0001] The present invention relates to a technical field, and in particular to an instrument equipment fault detection system based on big data technology. Background Art

[0002] In recent years, instrument equipment fault detection is developing towards higher intelligence and automation. More and more instrument equipment has integrated intelligent sensors and self-diagnosis functions, which can monitor the status in real time. Big data analysis and machine learning technologies are used to improve the accuracy and response speed of fault prediction. The application of cloud computing technology makes remote monitoring and maintenance possible, reducing labor costs.

[0003] At present, in a Chinese invention patent with publication number CN 115700802 A, an instrument equipment fault detection method, device, equipment and storage medium are disclosed. The method obtains the current pointer angle value by performing angle prediction on the current instrument pointer image obtained in real time, so that the instrument pointer value can be obtained in real time with higher accuracy. The accuracy of judging whether the equipment has a fault based on the obtained pointer angle value is higher. However, the related technology does not analyze the multi-dimensional parameters of the instrument equipment, lacks the comprehensiveness and comprehensiveness of fault detection, does not predict future faults based on historical data, and lacks the foresight and intelligence of fault detection. Summary of the invention

[0004] The technical problem solved by the present invention is that the related technology does not analyze the multi-dimensional parameters of the instrument equipment, lacks the comprehensiveness and comprehensiveness of fault detection, does not predict future faults based on historical data, and lacks the foresight and intelligence of fault detection.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an instrument equipment fault detection system based on big data technology, including a collection module, an analysis module and an early warning module;

[0006] The acquisition module sets a time point of a first historical time period, performs a first encoding on the time point, and obtains information of a sub-device at the time point, wherein the sub-device includes a sensor sub-device, a transmission sub-device, a power sub-device, and an execution sub-device;

[0007] The analysis module performs a first analysis on the information of the sensor sub-device, determines the fault of the sensor sub-device according to the first analysis result, which is recorded as the first fault, performs a second analysis on the information of the transmission sub-device, determines the fault of the transmission sub-device according to the second analysis result, which is recorded as the second fault, performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, which is recorded as the third fault, performs a fourth analysis on the information of the execution sub-device, determines the fault of the execution sub-device according to the fourth analysis result, which is recorded as the fourth fault, sets labels for the first fault, the second fault, the third fault, and the fourth fault, and constructs a transition probability matrix, and predicts the fault of the instrument equipment according to the transition probability matrix;

[0008] The early warning module sends out a first early warning signal, a second early warning signal, a third early warning signal and a fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively, and sends out a fifth early warning signal according to the predicted fault of the instrument and equipment.

[0009] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the acquisition module sets a time point of the first historical time period, and performs a first encoding on the time point, and the first encoding is represented by X M , where M is a natural number, obtaining information of sub-devices at a time point, wherein the first historical time period includes failures of various types of sub-devices;

[0010] The sub-devices include a sensing sub-device, a transmission sub-device, a power sub-device and an execution sub-device. The information of the sensing sub-device is the output signal waveform, the information of the transmission sub-device is the packet loss rate, the bit error rate and the transmission speed, the information of the power sub-device includes the voltage and current, and the information of the execution sub-device includes the response time, the execution speed, the temperature and the noise intensity.

[0011] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the analysis module performs a first analysis on the information of the sensor sub-equipment, and determines the fault of the sensor sub-equipment according to the first analysis result, which is recorded as a first fault. The first fault includes sensor failure and sensor drift. The setting logic of the first fault includes:

[0012] Output signal waveform, obtain the shape and value of the waveform, when the shape of the waveform is a horizontal straight line and the value is 0, set the first fault as sensor failure, when the shape of the waveform contains a small range of sawtooth waves and the general trend is a smooth curve, set the first fault as sensor drift, the horizontal straight line is represented by a straight line parallel to the horizontal axis, and the small range is represented by the starting coordinate difference of the sawtooth wave being less than half of the starting coordinate difference of the smooth curve.

[0013] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the analysis module performs a second analysis on the information of the transmission sub-equipment, determines the fault of the transmission sub-equipment according to the second analysis result, and records it as a second fault. The second fault includes signal loss, signal interference and signal delay. The setting logic of the second fault includes: setting the first value as the packet loss rate threshold, setting the second value as the bit error rate threshold, setting the third value as the transmission speed threshold, obtaining the packet loss rate, the bit error rate and the transmission speed, comparing the packet loss rate with the first value, when the packet loss rate is greater than or equal to the first value, setting the second fault as signal loss, comparing the bit error rate with the second value, when the bit error rate is greater than or equal to the second value, setting the second fault as signal interference, comparing the transmission speed with the third value, when the transmission speed is less than the third value, setting the second fault as signal delay.

[0014] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the analysis module performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, and records it as the third fault. The third fault includes short circuit and open circuit. The analysis logic of the third analysis includes: obtaining voltage, current and power supply model, retrieving the power supply database, inputting the power supply model into the power supply database, matching the rated current according to the power supply model, comparing the current with the rated current, when the current is greater than the rated current, setting the third fault to a short circuit, and when the voltage value is 0, setting the third fault to an open circuit.

[0015] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the analysis module performs a fourth analysis on the information of the execution sub-device, determines the fault of the execution sub-device according to the fourth analysis result, and records it as a fourth fault. The fourth fault includes execution failure, execution stagnation, execution overheating and execution wear. The setting logic of the fourth fault includes: obtaining response time, execution speed, temperature and noise intensity, setting the fourth value as the response time threshold, setting the fifth value as the temperature threshold, setting the sixth value as the noise intensity threshold, setting the seventh value as the execution speed threshold, comparing the response time with the fourth value, when the response time is greater than the fourth value, setting the fourth fault as execution failure, comparing the temperature with the fifth value, when the temperature is greater than the fifth value, setting the fourth fault as execution overheating, comparing the noise intensity with the sixth value, when the noise intensity is greater than the sixth value, setting the fourth fault as execution wear, comparing the execution speed with the seventh value, when the execution speed is less than the seventh value, setting the fourth fault as execution stagnation.

[0016] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the analysis module sets labels for the first fault, the second fault, the third fault and the fourth fault, assigns sensor failure to A1, sets sensor drift to A2, assigns signal loss to B1, assigns signal interference to B2, assigns signal delay to B3, assigns short circuit to C1, assigns open circuit to C2, assigns execution failure to D1, assigns execution jam to D2, assigns execution overheating to D3, assigns execution wear to D4, sets the label of each sub-equipment in normal operation to E, and constructs a transition probability matrix, and predicts the fault of the instrument equipment according to the transition probability matrix.

[0017] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, the construction logic of the transition probability matrix includes:

[0018] According to the permutation and combination method, the first fault, the second fault, the third fault, the fourth fault and the normal operation state after the label is set are permuted and combined to obtain sub-combinations. The number of the sub-combinations is 180. The sub-combinations are second-coded, and the second coding corresponds to the first coding. The second coding is represented by X MN , where N is a natural number, and N is distributed between [1,180]. Obtain the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point. Obtain the second code to which the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point belong, recorded as the state code. Sort the state codes in the order of the time points, obtain adjacent state codes, combine adjacent state codes to obtain migration pairs, merge the same migration pairs, count the number of occurrences of each migration pair, use each migration pair as the column and row of the statistical table, input the number of occurrences of the migration pair into the statistical table, and calculate the sum of each row. Calculate the quotient of the number of occurrences of the migration pair and the sum of each row, and use the quotient as the transition probability. Adjacent state codes are represented as the two state codes before and after the time point sequence.

[0019] When predicting the migration pair at the next time point, obtain the migration pair at the previous time point, and obtain the corresponding transfer probability through the statistical table. Through the last status code of the migration pair at the next time point, obtain the first fault, second fault, third fault, fourth fault or normal operation status at the next time point.

[0020] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, the early warning module sends out a first early warning signal, a second early warning signal, a third early warning signal and a fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively. When in normal operation state, no early warning signal is sent, and a fifth early warning signal is sent according to the predicted fault of the instrument equipment.

[0021] As a preferred solution of the instrument equipment fault detection system based on big data technology described in the present invention, wherein: the first warning signal, the second warning signal, the third warning signal and the fourth warning signal respectively light up warning signal lights of different colors and brightness, and display the current fault type through the OLED display screen; the fifth warning signal flashes the current fault type, predicted fault type, current time point, predicted time point and migration probability through the OLED display screen, and the predicted time point is obtained by weighting the time period between the current time point and the time point.

[0022] The beneficial effects of the present invention are as follows: through the acquisition module, the status of sub-devices can be obtained in real time and potential faults can be discovered in time; the analysis module performs detailed analysis on various sub-devices to ensure the accuracy of fault identification, constructs a transfer probability matrix, uses historical data to predict future faults, and improves the reliability of equipment; the early warning module sends out multi-level early warning signals to help maintenance personnel take timely measures to reduce the impact of faults, significantly improving the intelligence and automation level of equipment fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of the basic flow of an instrument equipment fault detection system based on big data technology provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0025] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides an instrument equipment fault detection system based on big data technology, including a collection module, an analysis module and an early warning module;

[0026] The acquisition module sets a time point of a first historical time period, performs a first encoding on the time point, and obtains information of a sub-device at the time point, wherein the sub-device includes a sensor sub-device, a transmission sub-device, a power sub-device, and an execution sub-device;

[0027] The analysis module performs a first analysis on the information of the sensor sub-device, determines the fault of the sensor sub-device according to the first analysis result, which is recorded as the first fault, performs a second analysis on the information of the transmission sub-device, determines the fault of the transmission sub-device according to the second analysis result, which is recorded as the second fault, performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, which is recorded as the third fault, performs a fourth analysis on the information of the execution sub-device, determines the fault of the execution sub-device according to the fourth analysis result, which is recorded as the fourth fault, sets labels for the first fault, the second fault, the third fault, and the fourth fault, and constructs a transition probability matrix, and predicts the fault of the instrument equipment according to the transition probability matrix;

[0028] The early warning module sends out a first early warning signal, a second early warning signal, a third early warning signal and a fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively, and sends out a fifth early warning signal according to the predicted fault of the instrument and equipment.

[0029] The present invention can obtain the status of sub-devices in real time and discover potential faults in time through the acquisition module. The analysis module conducts detailed analysis on various sub-devices to ensure the accuracy of fault identification, constructs a transfer probability matrix, and uses historical data to predict future faults to improve the reliability of equipment. The early warning module sends out multi-level early warning signals to help maintenance personnel take timely measures to reduce the impact of faults, thereby significantly improving the intelligence and automation level of equipment fault detection.

[0030] The acquisition module sets a time point of a first historical time period and performs a first encoding on the time point, wherein the first encoding is represented by X M , where M is a natural number, obtaining information of sub-devices at a time point, wherein the first historical time period includes failures of various types of sub-devices;

[0031] The sub-devices include a sensing sub-device, a transmission sub-device, a power sub-device and an execution sub-device. The information of the sensing sub-device is the output signal waveform, the information of the transmission sub-device is the packet loss rate, the bit error rate and the transmission speed, the information of the power sub-device includes the voltage and current, and the information of the execution sub-device includes the response time, the execution speed, the temperature and the noise intensity.

[0032] In the specific implementation, through the comprehensive collection of information from different sub-devices, key information is ensured to facilitate fault diagnosis. Various sub-devices, including sensing, transmission, power supply and execution, provide rich data dimensions to make fault detection more accurate. Through output signal waveform, packet loss rate, bit error rate and other indicators, equipment performance can be monitored in real time and anomalies can be quickly identified. Voltage and current monitoring of power sub-devices, response time and noise intensity of execution sub-devices and other information help to analyze the cause of the fault in more detail. The various data collected provide a solid foundation for subsequent analysis and decision-making, and enhance the reliability of the system.

[0033] The analysis module performs a first analysis on the information of the sensor sub-device, and determines a fault of the sensor sub-device according to the first analysis result, which is recorded as a first fault. The first fault includes sensor failure and sensor drift. The setting logic of the first fault includes:

[0034] Output signal waveform, obtain the shape and value of the waveform, when the shape of the waveform is a horizontal straight line and the value is 0, set the first fault as sensor failure, when the shape of the waveform contains a small range of sawtooth waves and the general trend is a smooth curve, set the first fault as sensor drift, the horizontal straight line is represented by a straight line parallel to the horizontal axis, and the small range is represented by the starting coordinate difference of the sawtooth wave being less than half of the starting coordinate difference of the smooth curve.

[0035] In specific implementation, by analyzing the shape and value of the output signal waveform, the specific fault type of the sensor sub-equipment, sensor failure and sensor drift can be accurately judged, and a clear fault judgment logic can be set to make the fault detection process more systematic, reduce misjudgment, and improve the reliability of detection. The real-time analysis of waveform changes enables fault identification to respond quickly, which helps to take maintenance measures in time and reduce equipment downtime. By conducting a detailed analysis of the waveform shape, such as horizontal straight lines and sawtooth waves, we can have a deeper understanding of the working status of the sensor, improve the accuracy of fault diagnosis, accurately judge the fault type, and help with targeted maintenance, avoid unnecessary waste of resources, and reduce overall maintenance costs.

[0036] The analysis module performs a second analysis on the information of the transmission sub-device, and determines the fault of the transmission sub-device according to the second analysis result, which is recorded as the second fault. The second fault includes signal loss, signal interference and signal delay. The setting logic of the second fault includes: setting the first value as the packet loss rate threshold, setting the second value as the bit error rate threshold, setting the third value as the transmission speed threshold, obtaining the packet loss rate, the bit error rate and the transmission speed, comparing the packet loss rate with the first value, when the packet loss rate is greater than or equal to the first value, setting the second fault as signal loss, comparing the bit error rate with the second value, when the bit error rate is greater than or equal to the second value, setting the second fault as signal interference, comparing the transmission speed with the third value, when the transmission speed is less than the third value, setting the second fault as signal delay.

[0037] In specific implementation, by analyzing the packet loss rate, bit error rate and transmission speed, different types of faults, signal loss, signal interference and signal delay can be carefully identified, and clear threshold standards can be set to standardize the fault judgment process, reduce the deviation of human judgment, and be able to obtain and compare key indicators in real time to ensure timely detection and response to signal anomalies. Through the analysis of specific indicators, the source of the fault can be located more accurately, providing clear guidance for subsequent maintenance.

[0038] The analysis module performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, and records it as the third fault. The third fault includes short circuit and open circuit. The analysis logic of the third analysis includes: obtaining voltage, current and power model, retrieving the power database, entering the power model into the power database, matching the rated current according to the power model, comparing the current with the rated current, when the current is greater than the rated current, setting the third fault to a short circuit, and when the voltage value is 0, setting the third fault to an open circuit.

[0039] In specific implementation, by analyzing voltage and current, it is possible to accurately identify fault types such as short circuit and open circuit, ensure the accuracy of detection, and use the power supply database to match the rated current to improve the basis for fault judgment and the intelligence level of the system. The clear analysis logic provides a standardized operating procedure for fault detection, improves work efficiency, and can prevent potential equipment damage and extend the service life of the equipment by timely identifying power supply faults.

[0040] The analysis module performs a fourth analysis on the information of the execution sub-device, and determines the fault of the execution sub-device according to the fourth analysis result, which is recorded as a fourth fault. The fourth fault includes execution failure, execution jam, execution overheating and execution wear. The setting logic of the fourth fault includes: obtaining response time, execution speed, temperature and noise intensity, setting the fourth value as a response time threshold, setting the fifth value as a temperature threshold, setting the sixth value as a noise intensity threshold, setting the seventh value as an execution speed threshold, comparing the response time with the fourth value, and when the response time is greater than the fourth value, setting the fourth fault as execution failure, comparing the temperature with the fifth value, and when the temperature is greater than the fifth value, setting the fourth fault as execution overheating, comparing the noise intensity with the sixth value, and when the noise intensity is greater than the sixth value, setting the fourth fault as execution wear, comparing the execution speed with the seventh value, and when the execution speed is less than the seventh value, setting the fourth fault as execution jam.

[0041] In the specific implementation, by analyzing the response time, execution speed, temperature and noise intensity, various fault types such as execution failure, execution jam, execution overheating and execution wear are comprehensively identified, and analysis is performed based on specific measurement data, which improves the scientificity and reliability of fault diagnosis.

[0042] The analysis module sets labels for the first fault, the second fault, the third fault and the fourth fault, assigns sensor failure to A1, sets sensor drift to A2, assigns signal loss to B1, assigns signal interference to B2, assigns signal delay to B3, assigns short circuit to C1, assigns open circuit to C2, assigns execution failure to D1, assigns execution jam to D2, assigns execution overheating to D3, assigns execution wear to D4, sets the label of each sub-device in normal operation to E, constructs a transition probability matrix, and predicts the faults of the instrument and equipment according to the transition probability matrix.

[0043] In the specific implementation, a unique label is set for each fault type to facilitate rapid identification and classification, improve the efficiency of fault management, and build a transition probability matrix to model the transition relationship between various fault states to enhance the understanding of the dynamic changes of faults. The transition probability matrix is ​​used to predict future faults based on the current state, identify potential risks in advance, and reduce the possibility of sudden failures.

[0044] The construction logic of the transition probability matrix includes:

[0045] According to the permutation and combination method, the first fault, the second fault, the third fault, the fourth fault and the normal operation state after the label is set are permuted and combined to obtain sub-combinations. The number of the sub-combinations is 180. The sub-combinations are second-coded, and the second coding corresponds to the first coding. The second coding is represented by XMN , where N is a natural number, and N is distributed between [1,180]. Obtain the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point. Obtain the second code to which the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point belong, recorded as the state code. Sort the state codes in the order of the time points, obtain adjacent state codes, combine adjacent state codes to obtain migration pairs, merge the same migration pairs, count the number of occurrences of each migration pair, use each migration pair as the column and row of the statistical table, input the number of occurrences of the migration pair into the statistical table, and calculate the sum of each row. Calculate the quotient of the number of occurrences of the migration pair and the sum of each row, and use the quotient as the transition probability. Adjacent state codes are represented as the two state codes before and after the time point sequence.

[0046] When predicting the migration pair at the next time point, obtain the migration pair at the previous time point, and obtain the corresponding transfer probability through the statistical table. Through the last status code of the migration pair at the next time point, obtain the first fault, second fault, third fault, fourth fault or normal operation status at the next time point.

[0047] In the specific implementation, up to 180 states can be generated by permuting and combining fault and normal states, ensuring comprehensive coverage of the system operation state. The second code is used to correspond to the first code, which simplifies the state representation and facilitates analysis and processing. By sorting the state codes in the time series, the changing law of the fault state can be captured, providing a basis for dynamic analysis. By combining adjacent state codes and counting their occurrence frequencies, the most common state transitions can be effectively identified, providing data support for prediction, and calculating the probability of occurrence of migration pairs, so that fault prediction has a stronger mathematical foundation and enhances the accuracy and reliability of the prediction. Based on the migration pairs and transition probabilities at the previous time point, the state at the next time point can be scientifically predicted, and potential faults can be identified in advance.

[0048] The early warning module sends out the first early warning signal, the second early warning signal, the third early warning signal and the fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively. When in normal operation state, no early warning signal is sent out. According to the predicted fault of the instrument and equipment, a fifth early warning signal is sent out.

[0049] The first warning signal, the second warning signal, the third warning signal and the fourth warning signal respectively light up warning signal lights of different colors and brightness, and display the current fault type through the OLED display screen. The fifth warning signal flashes the current fault type, predicted fault type, current time point, predicted time point and migration probability through the OLED display screen. The predicted time point is obtained by weighting the time period between the current time point and the time point.

[0050] In specific implementation, warning signal lights of different colors and brightness can quickly attract the attention of operators and facilitate rapid identification of fault types. The OLED display screen displays the current fault type, enhances the visualization of the system, and helps users better understand the equipment status. The flashing display function of the fifth warning signal provides a predicted fault type and time point, enhancing the intelligence level of the warning system. The predicted time point is calculated by weighted calculation of the current time point and the time interval, which improves the accuracy of the prediction and makes the warning more forward-looking.

[0051] The present invention can obtain the status of sub-devices in real time and discover potential faults in time through the acquisition module. The analysis module conducts detailed analysis on various sub-devices to ensure the accuracy of fault identification, constructs a transfer probability matrix, and uses historical data to predict future faults to improve the reliability of equipment. The early warning module sends out multi-level early warning signals to help maintenance personnel take timely measures to reduce the impact of faults, thereby significantly improving the intelligence and automation level of equipment fault detection.

[0052] It should be understood by those skilled in the art that the embodiments of the present invention can be provided as methods, systems or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0053] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. The instrument equipment fault detection system based on big data technology is characterized by: It includes collection module, analysis module and early warning module; The acquisition module sets a time point of a first historical time period, performs a first encoding on the time point, and obtains information of a sub-device at the time point, wherein the sub-device includes a sensor sub-device, a transmission sub-device, a power sub-device, and an execution sub-device; The analysis module performs a first analysis on the information of the sensor sub-device, determines the fault of the sensor sub-device according to the first analysis result, which is recorded as the first fault, performs a second analysis on the information of the transmission sub-device, determines the fault of the transmission sub-device according to the second analysis result, which is recorded as the second fault, performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, which is recorded as the third fault, performs a fourth analysis on the information of the execution sub-device, determines the fault of the execution sub-device according to the fourth analysis result, which is recorded as the fourth fault, sets labels for the first fault, the second fault, the third fault, and the fourth fault, and constructs a transition probability matrix, and predicts the fault of the instrument equipment according to the transition probability matrix; The early warning module issues a first early warning signal, a second early warning signal, a third early warning signal and a fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively, and issues a fifth early warning signal according to the predicted fault of the instrument and equipment; The analysis module performs a first analysis on the information of the sensor sub-device, and determines a fault of the sensor sub-device according to the first analysis result, which is recorded as a first fault. The first fault includes sensor failure and sensor drift. The setting logic of the first fault includes: Output signal waveform, obtain the shape and value of the waveform, when the shape of the waveform is a horizontal straight line and the value is 0, set the first fault as sensor failure, when the shape of the waveform contains a small range of sawtooth waves and the general trend is a smooth curve, set the first fault as sensor drift, the horizontal straight line is represented by a straight line parallel to the horizontal axis, and the small range is represented by the starting coordinate difference of the sawtooth wave being less than half of the starting coordinate difference of the smooth curve; The construction logic of the transition probability matrix includes: According to the permutation and combination method, the first fault, the second fault, the third fault, the fourth fault and the normal operation state after the label is set are permuted and combined to obtain sub-combinations. The number of the sub-combinations is 180. The sub-combinations are second-coded, and the second coding corresponds to the first coding. The second coding is represented by X MN , where N is a natural number, and N is distributed between [1,180]. Obtain the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point. Obtain the second code to which the labels of the first fault, the second fault, the third fault, and the fourth fault corresponding to the time point belong, recorded as the state code. Sort the state codes in the order of the time points, obtain adjacent state codes, combine adjacent state codes to obtain migration pairs, merge the same migration pairs, count the number of occurrences of each migration pair, use each migration pair as the column and row of the statistical table, input the number of occurrences of the migration pair into the statistical table, and calculate the sum of each row. Calculate the quotient of the number of occurrences of the migration pair and the sum of each row, and use the quotient as the transition probability. Adjacent state codes are represented as the two state codes before and after the time point sequence. When predicting the migration pair at the next time point, obtain the migration pair at the previous time point, and obtain the corresponding transfer probability through the statistical table. Through the last status code of the migration pair at the next time point, obtain the first fault, second fault, third fault, fourth fault or normal operation status at the next time point.

2. The instrument equipment fault detection system based on big data technology as claimed in claim 1, characterized in that: The acquisition module sets a time point of a first historical time period and performs a first encoding on the time point, wherein the first encoding is represented by X M , where M is a natural number, obtaining information of sub-devices at a time point, wherein the first historical time period includes failures of various types of sub-devices; The sub-devices include a sensing sub-device, a transmission sub-device, a power sub-device and an execution sub-device. The information of the sensing sub-device is the output signal waveform, the information of the transmission sub-device is the packet loss rate, the bit error rate and the transmission speed, the information of the power sub-device includes the voltage and current, and the information of the execution sub-device includes the response time, the execution speed, the temperature and the noise intensity.

3. The instrument equipment fault detection system based on big data technology as claimed in claim 1, characterized in that: The analysis module performs a second analysis on the information of the transmission sub-device, and determines the fault of the transmission sub-device according to the second analysis result, which is recorded as the second fault. The second fault includes signal loss, signal interference and signal delay. The setting logic of the second fault includes: setting the first value as the packet loss rate threshold, setting the second value as the bit error rate threshold, setting the third value as the transmission speed threshold, obtaining the packet loss rate, the bit error rate and the transmission speed, comparing the packet loss rate with the first value, when the packet loss rate is greater than or equal to the first value, setting the second fault as signal loss, comparing the bit error rate with the second value, when the bit error rate is greater than or equal to the second value, setting the second fault as signal interference, comparing the transmission speed with the third value, when the transmission speed is less than the third value, setting the second fault as signal delay.

4. The instrument equipment fault detection system based on big data technology as claimed in claim 3, characterized in that: The analysis module performs a third analysis on the information of the power sub-device, determines the fault of the power sub-device according to the third analysis result, and records it as the third fault. The third fault includes short circuit and open circuit. The analysis logic of the third analysis includes: obtaining voltage, current and power model, retrieving the power database, entering the power model into the power database, matching the rated current according to the power model, comparing the current with the rated current, when the current is greater than the rated current, setting the third fault to a short circuit, and when the voltage value is 0, setting the third fault to an open circuit.

5. The instrument equipment fault detection system based on big data technology as claimed in claim 4, characterized in that: The analysis module performs a fourth analysis on the information of the execution sub-device, and determines the fault of the execution sub-device according to the fourth analysis result, which is recorded as a fourth fault. The fourth fault includes execution failure, execution jam, execution overheating and execution wear. The setting logic of the fourth fault includes: obtaining response time, execution speed, temperature and noise intensity, setting the fourth value as a response time threshold, setting the fifth value as a temperature threshold, setting the sixth value as a noise intensity threshold, setting the seventh value as an execution speed threshold, comparing the response time with the fourth value, and when the response time is greater than the fourth value, setting the fourth fault as execution failure, comparing the temperature with the fifth value, and when the temperature is greater than the fifth value, setting the fourth fault as execution overheating, comparing the noise intensity with the sixth value, and when the noise intensity is greater than the sixth value, setting the fourth fault as execution wear, comparing the execution speed with the seventh value, and when the execution speed is less than the seventh value, setting the fourth fault as execution jam.

6. The instrument equipment fault detection system based on big data technology as claimed in claim 5, characterized in that: The analysis module sets labels for the first fault, the second fault, the third fault and the fourth fault, assigns sensor failure to A1, sets sensor drift to A2, assigns signal loss to B1, assigns signal interference to B2, assigns signal delay to B3, assigns short circuit to C1, assigns open circuit to C2, assigns execution failure to D1, assigns execution jam to D2, assigns execution overheating to D3, assigns execution wear to D4, sets the label of each sub-device in normal operation to E, constructs a transition probability matrix, and predicts the faults of the instrument and equipment according to the transition probability matrix.

7. The instrument equipment fault detection system based on big data technology as claimed in claim 1, characterized in that: The early warning module sends out the first early warning signal, the second early warning signal, the third early warning signal and the fourth early warning signal according to the first fault, the second fault, the third fault and the fourth fault respectively. When in normal operation state, no early warning signal is sent out. According to the predicted fault of the instrument and equipment, a fifth early warning signal is sent out.

8. The instrument equipment fault detection system based on big data technology as claimed in claim 1, characterized in that: The first warning signal, the second warning signal, the third warning signal and the fourth warning signal respectively light up warning signal lights of different colors and brightness, and display the current fault type through the OLED display screen. The fifth warning signal flashes the current fault type, predicted fault type, current time point, predicted time point and migration probability through the OLED display screen. The predicted time point is obtained by weighting the time period between the current time point and the time point.

Images