Fault real-time detection method and device, equipment and medium

Abstract

The invention provides a fault real-time detection method and device, equipment and a medium, and the method comprises the steps: obtaining the operation state data of at least one piece of steel mill equipment, and carrying out the standardization processing, and generating standardized data, dividing the standardized data into four operation stages of raw material preparation, smelting processing, forming processing and shutdown according to the operation stages; four data queues are constructed on cache equipment, data flow fluctuation coefficients are calculated according to the standardized data cache inflow rates of the four data queues and the residual capacity of the cache equipment, and cache strategies of the four data queues are adjusted according to the data flow fluctuation coefficients; and respectively inputting the standardized data of the four data queues into corresponding equipment fault judgment models for processing to obtain fault data, wherein each equipment fault judgment model comprises a first rule layer and a second rule layer. And the real-time fault detection effect is improved through caching in stages and dual judgment of a static threshold value and a dynamic threshold value.

Description

Technical Field

[0001] The present disclosure relates to the technical field of fault detection, and particularly to a method, device, equipment and medium for real-time fault detection. Background Art

[0002] In modern steelmaking production, the normal operation of equipment is the key to ensuring production efficiency and product quality. Traditional equipment maintenance methods often rely on regular inspections and manual judgments, making it difficult to predict and handle equipment failures in a timely manner, resulting in increased equipment downtime and rising maintenance costs.

[0003] With the development of big data applications, in order to process big data more intelligently, the collection, transmission, and control of data content have become increasingly important. However, in the prior art, there are many types of steel plant equipment and a large amount of data is generated, and the types of equipment failures are relatively complex. Traditional processing systems cannot process real-time data, and only rely on a single parameter threshold for judgment, resulting in poor fault prediction effects and unable to accurately and efficiently detect abnormal data in the real-time data of steel plant equipment. Summary of the Invention

[0004] The present disclosure provides a method, device, equipment and medium for real-time fault detection to solve the problems in the prior art that real-time data processing of steel plant equipment cannot be performed and early warning cannot be accurately and efficiently carried out.

[0005] In a first aspect, the present disclosure provides a method for real-time fault detection, the method comprising: Obtain the operation status data of at least one steel plant equipment, perform standardization processing on the operation status data to generate standardized data with time stamp alignment, and divide the standardized data into standardized data of four operation stages: raw material preparation, smelting and processing, forming and processing, and shutdown according to the operation stages; Construct four data queues on a cache device for receiving the standardized data of the four operation stages, calculate the data flow fluctuation coefficients of the four data queues respectively according to the cache inflow rate of the standardized data of the four data queues and the remaining capacity of the cache device, and adjust the cache policies of the four data queues according to the data flow fluctuation coefficients; Input the standardized data of the four data queues into corresponding equipment fault judgment models for processing to obtain fault data; wherein, different data queues correspond to different equipment fault judgment models, and the equipment fault judgment model includes a first rule layer and a second rule layer, the first rule layer is used for performing static threshold judgment on the standardized data, and the second rule layer is used for performing dynamic threshold judgment on the standardized data; Send a warning message that a steel plant equipment has a fault according to the fault data.

[0006] According to a real-time fault detection method provided by the present disclosure, the standardized data of the four data queues are respectively input into corresponding device fault judgment models for processing to obtain fault data, which specifically includes: for any data queue, obtaining multi-source data corresponding to the device model from the standardized data with aligned timestamps according to the device model; inputting the multi-source data into the first rule layer for static judgment of parameter thresholds, and outputting a first fault detection result, where the first fault detection result includes: unmarked multi-source data and / or first-level marked fault data; inputting the multi-source data into the second rule layer for dynamic judgment of parameter thresholds, and outputting a second fault detection result, where the second fault detection result includes: unmarked multi-source data and / or second-level marked fault data; determining the fault data corresponding to the multi-source data according to the first fault detection result and the second fault detection result, and the fault data includes fault types and marking information.

[0007] According to a real-time fault detection method provided by the present disclosure, the first rule layer includes static judgment rules in a multi-level key-value form constructed according to device type, device model, fault type, and parameter combination; among them, the determination of the thresholds of the parameters in the parameter combination of the first rule layer includes at least one of the following: determining a first static threshold of a parameter according to a basic threshold set in a device manufacturing specification and the service life of the device; or, determining a second static threshold of a parameter according to historical fault data of the device and the device health; or, determining a third static threshold according to the first static threshold and the second static threshold.

[0008] According to a real-time fault detection method provided by the present disclosure, the second rule layer includes dynamic judgment rules in a multi-level key-value form constructed according to device type, device model, fault type, and parameter combination, and the determination of the thresholds of the parameters in the parameter combination of the second rule layer includes: obtaining the thresholds of the parameters in the parameter combination determined by the first rule layer; calculating the coupling between the parameters in the parameter combination according to the historical fault data of the device to obtain a parameter correlation weight matrix; calculating the dynamic thresholds of the parameters according to the parameter correlation matrix and the thresholds of the parameters.

[0009] According to a real-time fault detection method provided by the present disclosure, calculating the data flow fluctuation coefficients of the four data queues according to the cache inflow rate of the standardized data of the four data queues and the remaining capacity of the cache device respectively includes: using a flow monitoring sensor at the data transmission interface of the cache device to monitor the cache inflow rate of the standardized data of the four data queues; using a capacity monitor integrated inside the cache device to monitor the remaining capacity of the cache device, and the remaining capacity of the cache device includes the remaining capacity of each data queue; calculating the data flow fluctuation coefficients of each data queue according to the cache inflow rate and the remaining capacity of the standardized data of each data queue.

[0010] According to a real-time fault detection method provided by the present disclosure, calculating the data flow fluctuation coefficients of each data queue based on the cache inflow rate and remaining capacity of the standardized data of each data queue includes: calculating the data flow fluctuation coefficients of each data queue according to a first formula, and the first formula is:

[0011] Wherein, the standard deviation of the data inflow rate is the standard deviation of the data inflow rate within a preset time period, the average data inflow rate is the average inflow rate of the data inflow rate within a preset time period, α is a capacity weight coefficient, "1 - remaining capacity percentage" is a cache capacity adjustment factor, and the remaining capacity percentage is the ratio between the remaining capacity and the allocated capacity of each data queue.

[0012] According to a real-time fault detection method provided by the present disclosure, adjusting the cache policies of the four data queues according to the data flow fluctuation coefficients includes: classifying the data cache inflows of each data queue into a stable state, a violently fluctuating state, and an extremely fluctuating state according to the data flow fluctuation coefficients and fluctuation thresholds of each data queue; when the data cache inflow of any data queue is in the extremely fluctuating state, temporarily storing the standardized data to be cached in this data queue in a cloud cache device through a high-speed data transmission link; when the data cache inflow of any data queue is in the violently fluctuating state and there is at least one data queue with a stable data cache inflow among the four data queues, increasing the allocation amount of the data queue with a violently fluctuating data cache inflow and decreasing the allocation amount of the data queue with a stable data cache inflow; when the data cache inflow of any data queue is in the violently fluctuating state and there is no data queue with a stable data cache inflow among the four data queues, temporarily storing the standardized data to be cached in this data queue in a cloud cache device through a high-speed data transmission link; when the data cache inflow of any data queue is in the stable state, cleaning the data that has not been accessed for a long time and organizing the storage fragments of the local cache device.

[0013] In a second aspect, the present disclosure also provides a real-time fault detection device, and the device includes: a data acquisition module, configured to acquire the operation state data of at least one steel plant device, perform standardized processing on the operation state data to generate standardized data with time stamps aligned, and divide the standardized data into standardized data of four operation stages: raw material preparation, smelting and processing, forming and processing, and shutdown according to the operation stages. A data caching module, configured to build four data queues on a caching device for receiving the standardized data of the four operation stages, calculate the data flow fluctuation coefficients of the four data queues respectively according to the standardized data caching inflow rates and the remaining capacity of the caching device, and adjust the caching policies of the four data queues according to the data flow fluctuation coefficients; A processing module, configured to respectively input the standardized data of the four data queues into corresponding device fault judgment models for processing to obtain fault data; wherein, different data queues correspond to different device fault judgment models, and the device fault judgment models include a first rule layer and a second rule layer, the first rule layer is configured to perform static threshold judgment on the standardized data, and the second rule layer is configured to perform dynamic threshold judgment on the standardized data; An early warning module, configured to send early warning information about faults occurring in steel plant equipment according to the fault data.

[0014] In a third aspect, the present disclosure further provides an electronic device, which includes: a processor and a memory, the memory stores machine-readable instructions executable by the processor, and the processor is configured to read the machine-readable instructions from the memory and execute the machine-readable instructions to implement the fault real-time detection method provided by the embodiments of the present disclosure.

[0015] In a fourth aspect, the present disclosure further provides a computer-readable storage medium storing a computer program, and the computer program is configured to execute the fault real-time detection method provided by the embodiments of the present disclosure.

[0016] In summary, the fault real-time detection method, device, equipment and medium provided by the present disclosure improve the data quality of operation status data by performing standardization processing and timestamp alignment on real-time operation status data; effectively improve the effect of device fault detection by dividing the standardized data into operation stages and setting different fault judgment models for different operation stages; ensure the stability of real-time data processing by adaptively caching the cached data of each data queue on the caching device; realize real-time, accurate and efficient detection of various device faults by performing dual judgments of static thresholds and dynamic thresholds on the standardized data of different operation stages, and at the same time realize early warning of faults through dynamic threshold judgment, and give early warning of faults before they occur, effectively improving the fault detection effect. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the present disclosure or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present disclosure. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of a method for real-time fault detection provided by the present disclosure; Figure 2 It is a schematic flow chart of a method for processing a fault judgment model provided by the present disclosure; Figure 3 It is a schematic structural diagram of a device for real-time fault detection provided by the present disclosure; Figure 4 It is a schematic structural diagram of an electronic device provided by the present disclosure. Detailed implementation manners

[0019] To make the objectives, technical solutions, and advantages of the present disclosure clearer, the following will clearly and completely describe the technical solutions in the present disclosure in conjunction with the accompanying drawings in the present disclosure. Obviously, the described embodiments are some embodiments of the present disclosure, rather than all embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present disclosure.

[0020] Figure 1 It is a schematic flow chart of a method for real-time fault detection provided by the present disclosure. Referring to Figure 1 , the method includes: Step S11, obtaining the operation status data of at least one steel plant device, performing standardization processing on the operation status data to generate standardized data with time stamps aligned, and dividing the standardized data into standardized data of four operation stages: raw material preparation, smelting and processing, forming and processing, and shutdown according to the operation stage.

[0021] Specifically, it can be understood that the steel plant device refers to some devices used in the steel production process, such as iron-making devices like blast furnaces and hot blast stoves, steel-making devices like converters and electric furnaces, and rolling devices like rolling mills. The operation status data refers to the device operation data collected in real time for each steel plant device. To ensure the safe and stable operation of the steel plant devices and the smooth progress of the production process, it is necessary to obtain the operation status data of the devices in real time so as to timely give early warnings of device faults and take corresponding measures. The operation status data obtained in real time includes but is not limited to temperature, vibration, pressure, current, sound, device operation parameters, etc.

[0022] Among them, temperature is a key factor affecting the performance of steel plant equipment. Since steel plant equipment is in a high-temperature environment for a long time, abnormal temperature may imply wear of equipment parts, etc. For example, high temperature will exacerbate the thermal expansion of bearings, and may even cause abnormal vibration of the room temperature of the shafting. Therefore, real-time monitoring of the temperature change of the equipment can timely give early warnings of equipment failures.

[0023] Vibration is an important basis for equipment fault diagnosis. For example, vibration data of key parts such as bearings and gears of steel plant equipment are monitored in real time to timely detect abnormal vibration in key parts, and then give early warnings in a timely manner.

[0024] Pressure stability is an important foundation for the safe operation of equipment. Overpressure may cause an explosion, and underpressure may cause the equipment to fail to work and then damage the equipment; at the same time, pressure can also reflect whether the seal of the equipment fails and whether there are problems such as liquid leakage in the equipment.

[0025] Current is closely related to the load condition of the equipment. Abnormal current may indicate short circuit or overload faults in the equipment; to a certain extent, sound can also reflect the operating state of the equipment. When the sound of the equipment during normal operation is relatively stable, when abnormal noises such as knocking sounds and friction sounds appear, it may imply problems such as loosening or failure of internal parts.

[0026] In some embodiments, the data acquisition of parameters such as temperature, vibration, pressure, current, and sound of steel plant equipment needs to rely on sensors for acquisition. At this time, distributed sensor arrays can be used for data acquisition, that is, in the steel plant environment, one or more of temperature sensors, vibration sensors, pressure sensors, vibration sensors, and current sensors are deployed at key points of each equipment. For example, temperature sensors are installed at positions such as the furnace wall of the blast furnace, the furnace body of the converter, and the mold of the continuous caster to monitor the temperature change of the equipment; pressure sensors are installed at positions such as the cooling system of the blast furnace and the cooling system of the continuous caster to monitor the pressure of the cooling water; pressure sensors are installed in the oxygen lance system of the converter to monitor the pressure of oxygen; vibration sensors are installed at positions such as the tapping hole and slag tapping hole of the blast furnace, the tilting mechanism of the converter, and the straightening machine of the continuous caster to monitor the vibration state of the equipment.

[0027] Specifically, obtaining the operating state data of at least one steel plant equipment includes: obtaining the operating state data of at least one steel plant equipment collected in real time by the distributed sensor array.

[0028] In some other embodiments, some steel plant equipment itself is built-in with monitoring instruments, and some parameters such as temperature and vibration of the steel plant equipment can be collected in real time through the built-in monitoring instruments.

[0029] Specifically, obtaining the operation status data of at least one steel mill device further includes: obtaining the operation status data of the steel mill device collected in real time by the built-in monitoring instruments of at least one steel mill device.

[0030] It should be noted that the sensors deployed in the steel mill can also be other types of sensors, such as acceleration sensors, speed sensors, etc. The specifically deployed sensors can be adjusted according to the actual data collection requirements, and the embodiments of the present disclosure do not make special limitations on this. In addition, at key points of some devices, dual sensors of the same type can be deployed to avoid single-point failure.

[0031] Specifically, it can also be understood that since there are many and complex devices in the steel mill, a large amount of data will be continuously generated during the operation of these devices, including the operation parameters of the devices, process parameters, and monitoring data collected in real time by sensors. In order to achieve timely prediction and handling of equipment failures, it is often necessary to monitor multi-source data of at least one device collected in real time by multiple sensors. The formats of the parameters collected by different sensors are also different, and the production environment of the steel mill is complex, with interference factors such as high temperature, dust, and vibration. There may also be certain noise in the equipment data. Therefore, after obtaining the operation status data collected in real time, it is also necessary to perform standardized processing such as data cleaning and format conversion on the data.

[0032] In some embodiments, performing standardized processing on the operation status data to generate standardized data with timestamp alignment includes: performing data cleaning on the operation status data to obtain denoised data; performing format conversion on the denoised data to obtain standard format data; and generating standardized data with timestamp alignment according to the collection time of the standard format data.

[0033] Among them, the data cleaning can be one of the common denoising methods in the field of big data processing, and the embodiments of the present disclosure do not make limitations on this. The noise in the operation status data collected in real time is removed through data cleaning. The format conversion refers to integrating and normalizing the denoised data after cleaning to generate a unified data format to ensure that the same type of data maintains a unified data format. The collection time refers to the time node when each sensor collects the corresponding parameters after various parameters are sequentially generated by the steel mill device. Since the operation status data of the steel mill device has obvious temporal characteristics, further time alignment of the standard format data is convenient for subsequent prediction and handling of equipment failures, etc.

[0034] Specifically, it can also be understood that since the response sensitivity of the device to parameters is different in different stages. In order to more accurately detect equipment failures, the operation status data collected in real time is further divided into operation status data of multiple operation stages according to the equipment process flow.

[0035] In some embodiments, the operation status data collected in real time can be further divided into operation status data of multiple operation stages according to the device process flow. The standardized data with aligned timestamps after standardization processing can be divided into standardized data of four operation stages: raw material preparation, smelting and processing, shaping and processing, and shutdown according to the operation stages. Quality monitoring or other operation stages can also be added according to the actual working environment of the steel plant, or each operation stage can be further subdivided according to the actual working environment of the steel plant. The embodiments of the present disclosure do not make special limitations on this.

[0036] The operation stage of raw material preparation can be pre-treatment operations such as crushing, screening, and mixing of raw materials. For example, a crusher crushes large raw materials into appropriate particle sizes, and a vibrating screen screens the crushed materials; it can also be the start-up preheating stage before the equipment operates. The real-time fault detection for the operation stage of raw material preparation can be the detection of crusher motor faults, overloading of crushers, damage to crusher bearings, loosening of crusher anchor bolts, loosening of vibrating screen box components, and damage to vibrating screen meshes; it can also be for equipment in the smelting and processing or shaping and processing stages, and determine whether the equipment has faults in the start-up preheating stage by detecting parameters such as temperature change, lubrication, and sealing. For example, thermal stress cracks in the hearth and furnace wall of a blast furnace, cracking of the lining of a converter, etc.

[0037] The operation stage of smelting and processing refers to converting raw materials into molten steel through processes such as high-temperature melting. For example, a blast furnace reduces iron ore to molten iron, a hot blast stove provides hot air for the blast furnace, and a converter or an electric furnace further refines the molten iron. The real-time fault detection for the operation stage of raw material preparation can be the detection of hearth burn-through of a blast furnace, hot blast stove faults of a blast furnace, tilting mechanism faults of a converter, and erosion of the converter lining.

[0038] The operation stage of shaping and processing refers to processing the molten steel after smelting into steel products of different shapes and specifications through various shaping equipment to meet the needs of different users. Common shaping equipment such as continuous casting machines and rolling mills. The operation stage of shaping and processing can be that a continuous casting machine solidifies molten steel into billets, and the operation stage of shaping and processing can also be that a rolling mill rolls the billets into various steel products such as plates, pipes, and wire rods. The real-time fault detection for the operation stage of shaping and processing can be the detection of continuous casting machine mold faults, wear of continuous casting machine segment rollers, rolling mill roll faults, rolling mill motor faults, etc.

[0039] The described shutdown operation stage refers to when the equipment shuts down. When the equipment shuts down, the states of various components gradually return to their initial states, and the data changes during this process can also reflect some performance and potential problems of the equipment. For the real-time detection of faults in the described shutdown operation stage, for example, the temperature at key parts such as the tuyeres of blast furnaces and the lining of converters should drop according to a certain law when shutting down, which may indicate equipment faults. Another example is that during the shutdown process of a rolling mill, the vibration sensor detects that the vibration amplitude always remains at a high level and is accompanied by abnormal low-frequency vibrations. After detection, it is found that the bearing of the rolling mill is damaged, resulting in unstable rotation of the rolling mill and causing vibrations.

[0040] In the above method, by performing standardized processing and timestamp alignment on the real-time collected operation status data, the data quality, data accuracy, and reliability of the collected operation status data are improved; by dividing the standardized data into operation stages according to the technological process of the steel plant equipment, grouped processing of the standardized data is realized. On the one hand, the difficulty of real-time data processing is reduced. On the other hand, since there are differences in the parameter types, thresholds, etc. that need to be referred to when detecting equipment faults in different operation stages, dividing the standardized data by stage can effectively improve the effect of equipment fault detection.

[0041] Step S12: Build four data queues on the caching device for receiving the standardized data of the four operation stages. Calculate the data flow fluctuation coefficients of the four data queues respectively according to the inflow rate of the standardized data of the four data queues and the remaining capacity of the caching device, and adjust the caching strategies of the four data queues according to the data flow fluctuation coefficients.

[0042] Among them, the caching device is a local caching device, which can receive the real-time collected standardized data normally in the case of network disconnection or weak network. Establish connection channels between the caching device and each steel plant equipment and each sensor. The connection channels are wired connection channels or wireless connection channels. Through the connection channels, the operation status data collected by each steel plant equipment or each sensor are transmitted to the caching device for caching, avoiding directly performing fault judgment on the standardized data with a large amount of data, reducing the pressure of data reading and data processing for fault judgment, improving the response speed, and maintaining the stability of data transmission and processing.

[0043] The initial lengths of the four data queues may be the same, and the initial length of each data queue is calculated by evenly dividing according to the cache device capacity; the initial lengths of the four data queues may also be different. Different weight values are assigned to each data queue according to the operating stages of the standardized data cached by each data queue, and then the initial length of each data queue is calculated according to the cache device capacity and the weight values of each data queue. For example, since the amount of real-time data collected during the shutdown operation stage is relatively small, while the amount of real-time data collected during the smelting and processing and forming processing operation stages is relatively large, when assigning weight values, the weight sizes of the data queues for caching the standardized data of each operation stage can be in the order of smelting and processing > forming processing > raw material preparation > shutdown.

[0044] Specifically, it can be understood that the amount of real-time data collected by monitoring the operation of steel plant equipment at different operation stages is closely related to the operation stage of the steel plant equipment. For example, in the initial start-up stage of the equipment, the stage of smelting raw materials into molten iron, and the stage of transferring molten iron in a blast furnace, the amount of data in each stage will change accordingly. Especially in the stage of transferring molten iron, when the molten iron flows out, multiple sensors such as temperature sensors and flow sensors need to work at high frequency simultaneously, and the amount of data may reach nearly 20 times that of normal times. Another example is that when collecting the vibration data of a rolling mill, the sampling frequency may be 100Hz when the rolling mill is operating normally, while it may reach 10kHz during high-speed rolling. Therefore, in order to ensure that the cache device can cache the operation status data of steel plant equipment collected in real time quickly, accurately, and stably, the data flow fluctuation coefficient can be calculated during caching, and the cache strategy of each data queue can be adaptively adjusted through the data flow fluctuation coefficient.

[0045] In some embodiments, calculating the data flow fluctuation coefficients of the four data queues according to the standardized data cache inflow rate and the remaining capacity of the cache device respectively includes: Step S121, monitoring the cache inflow rate of the standardized data of the four data queues by using a flow monitoring sensor at the data transmission interface of the cache device.

[0046] Step S122, monitoring the remaining capacity of the cache device by using a capacity monitor integrated inside the cache device, where the remaining capacity of the cache device includes the remaining capacity of each data queue.

[0047] Step S123, calculating the data flow fluctuation coefficients of each data queue according to the cache inflow rate and the remaining capacity of the standardized data of each data queue.

[0048] Specifically, steps S121 and S122 monitor the standardized data cache inflow rate and remaining capacity for each data queue. When constructing four data queues on the cache device, the initial length of each data queue has been specified. Based on the initial length, the allocated capacity of each data queue can be determined. Then, according to the actually cached data volume in each data queue, the remaining capacity of each data queue can be calculated. Then, the data flow fluctuation coefficient of each data queue is calculated according to the first formula, and the first formula is:

[0049] Among them, the standard deviation of the data inflow rate is the standard deviation of the data inflow rate within a preset time period, and the average data inflow rate is the average inflow rate of the data inflow rate within a preset time period. The preset time period is the past few hours, the past day, or the past week, and can be specifically set according to the actual situation. The present disclosure does not make any limitation thereto.

[0050] α is the capacity weight coefficient, which is used to control the influence degree of the cache device capacity on the overall fluctuation coefficient. The value range of α is 0 to 1. Generally, the initial value of α can be set to 0.5.

[0051] "1 - remaining capacity percentage" is the cache capacity adjustment factor, and the remaining capacity percentage is the ratio between the remaining capacity and the allocated capacity of each data queue. When the remaining capacity of the data queue is relatively large, the remaining capacity percentage is close to 1, that is, the cache capacity adjustment factor is close to 0. At this time, the calculated data flow fluctuation coefficient is completely based on the fluctuation situation within the preset time period; while when the remaining capacity of the data queue is relatively small, the cache capacity adjustment factor increases, and at this time, the calculated data flow fluctuation coefficient will also increase further, increasing the sensitivity of the data flow fluctuation coefficient to the data flow fluctuation.

[0052] In some embodiments, adjusting the cache policies of the four data queues according to the data flow fluctuation coefficient includes: Step S124, classifying the data cache inflow of each data queue into a stable state, a violently fluctuating state, and an extremely fluctuating state according to the data flow fluctuation coefficient and the fluctuation threshold of each data queue.

[0053] Among them, the fluctuation thresholds at least include a first fluctuation threshold and a second fluctuation threshold, and the fluctuation thresholds of each data queue are different, that is, each data queue is separately provided with two fluctuation thresholds; for any data queue, when the data traffic fluctuation coefficient is less than or equal to the second threshold, the data cache inflow of this data queue is divided into a stable state; when the data traffic fluctuation coefficient is greater than the second threshold and less than the first threshold, the data cache inflow of this data queue is divided into a severe fluctuation state; when the data traffic fluctuation coefficient is greater than or equal to the first threshold, the data cache inflow of this data queue is divided into an extreme fluctuation state.

[0054] Step S125, when the data cache inflow of any data queue is in the extreme fluctuation state, use the high-speed data transmission link to temporarily store the standardized data to be cached in this data queue in the cloud cache device.

[0055] Step S126, when the data cache inflow of any data queue is in the severe fluctuation state and there is at least one data queue with a stable data cache inflow among the four data queues, increase the allocation of the data queue with the severe fluctuation state of the data cache inflow and reduce the allocation of the data queue with the stable data cache inflow.

[0056] Specifically, when there is a queue with a severe fluctuation state in the data cache inflow among the four data queues, the remaining capacity can be readjusted with the data queue in the stable state. For example, if there is one data queue in the severe fluctuation state, expand the capacity of the data queue in the severe fluctuation state according to the data queue in the stable state to reduce the cache pressure of this data queue. If there are multiple data queues in the severe fluctuation state, allocate and expand the capacity of the multiple data queues in the severe fluctuation state according to the data queue in the stable state. The specific expansion capacity of each data queue can be allocated according to the data traffic fluctuation coefficient of the multiple data queues in the severe fluctuation state, or can be allocated according to the respective remaining capacities of the multiple data queues in the severe fluctuation state. The embodiments of the present disclosure do not limit this.

[0057] Step S127, when the data cache inflow of any data queue is in the severe fluctuation state and there is no data queue with a stable data cache inflow among the four data queues, use the high-speed data transmission link to temporarily store the standardized data to be cached in this data queue in the cloud cache device; Specifically, when there is a severe fluctuation state among the four data queues and there is no data queue with a stable data cache inflow to provide the expansion condition, it is also necessary to temporarily cache in the cloud to ensure that the received acquisition data can be cached in real time, thereby ensuring the timeliness of the data.

[0058] Step S128: When the data cache inflow of any data queue is in a steady state, clean the data that has not been accessed for a long time and organize the storage fragments of the local cache device.

[0059] Specifically, when the data cache inflow of the data queue is relatively steady, the cache device can be cleaned to clear more space for allocation to each data queue.

[0060] In the above method, by caching the standardized data of the four operation stages in the cache device, it is ensured that the real-time collected data can be cached in the cache device, guaranteeing the timeliness of the data obtained during fault judgment. At the same time, the access pressure on each sensor is reduced, and the real-time data is directly read from the cache device for fault judgment, improving the stability of real-time data processing. By calculating the data flow fluctuation coefficient in combination with the cache capacity adjustment factor, the volatility of the data cache inflow rate and the ability of the current cache device to withstand fluctuations can be quickly understood, and the data flow fluctuation coefficients of each data queue are further used to achieve adaptive adjustment of the data queue cache strategy. By adopting cache adjustment strategies such as temporarily storing in the cloud for extremely fluctuating cache inflow states, expanding the capacity for severely fluctuating cache inflow states, and cleaning some data when the cache inflow state is relatively steady, the normal reception and processing of data are ensured, avoiding the overload of the local cache device, releasing the cache space, thereby improving the utilization rate of the cache device, and further guaranteeing the stability of real-time data processing.

[0061] Step S13: Input the standardized data of the four data queues into the corresponding device fault judgment models for processing to obtain fault data.

[0062] Among them, different data queues correspond to different device fault judgment models, and the device fault judgment model includes a first rule layer and a second rule layer. The first rule layer is used for static threshold judgment of the standardized data, and the second rule layer is used for dynamic threshold judgment of the standardized data.

[0063] Specifically, it can be understood that the parameter judgment rules for whether a device fails in different operation stages are also different. Therefore, before processing the standardized data of the cache device, it is necessary to pre-train the corresponding device fault judgment models for the four operation stages of raw material preparation, smelting and processing, forming and processing, and shutdown respectively. That is, different data queues correspond to different device fault judgment models.

[0064] Specifically, it can also be understood that due to the numerous types of steel mill equipment, there are also many corresponding fault types. The fault detection for steel mill equipment can be through comparing the threshold of a single parameter to detect equipment faults. For example, for the bearing damage of a crusher, it is necessary to monitor whether the vibration amplitude significantly increases or abnormal vibration frequency appears; for the spring fracture of a vibrating screen, it is necessary to monitor whether the amplitude of the vibrating screen is too large; for the mold fault of a continuous caster, it is necessary to monitor the vibration frequency of the mold, etc. The fault detection for steel mill equipment can also be through the combined analysis of multiple parameters. For example, for the abnormal furnace condition of a blast furnace, it can be comprehensively judged by combining the increase in top temperature and the fluctuation of static pressure in the furnace body; for the tuyere burn-through of a blast furnace, it is necessary to monitor the temperature difference change of tuyere cooling water and the fluctuation of top pressure, etc.; for the oxygen lance nodulation of a converter, it needs to be combined with oxygen lance current, blower blowing time, etc. Whether it is the fault detection of a single parameter or multiple parameters, it is to rely on the real-time detected data to judge whether a fault has occurred when the fault occurs, and it is impossible to predict equipment faults in advance. Therefore, in order to quickly and accurately predict equipment faults through the operation status data collected in real time, an equipment fault judgment model including a first rule layer and a second rule layer is constructed. On the one hand, at least two parameter thresholds can be set for the fault judgment rules of each fault type to improve the fault detection accuracy; on the other hand, a static threshold for real-time detection and a dynamic threshold for advance prediction can be set, so that while performing real-time fault detection, the fault can also be detected in advance, further improving the detection effect.

[0065] In some embodiments, the first rule layer includes a static judgment rule in the form of multi-level key values constructed according to equipment type, equipment model, fault type, and parameter combination. The thresholds of each parameter in the parameter combination of the first rule layer are static thresholds, and the static thresholds can be one-sided thresholds or two-sided thresholds. The present disclosure provides three methods for specifically determining the static thresholds. Among them, the determination of the thresholds of each parameter in the parameter combination of the first rule layer includes at least one of the following: a1. Determine the first static threshold of the parameter according to the basic threshold set in the equipment manufacturing specification and the service life of the equipment; or, a2. Determine the second static threshold of the parameter according to the historical fault data and equipment health of the equipment; or, a3. Determine the third static threshold according to the first static threshold and the second static threshold.

[0066] Specifically, for the first method a1 of determining the threshold of the parameter in the first rule layer, since the equipment is more likely to have faults after being used for a long time, it is necessary to calculate the product of the threshold corresponding to the fault type given at the time of equipment factory and the depreciation coefficient determined according to the service life of the equipment as the static threshold, so as to perform more accurate fault detection.

[0067] Specifically, for the second method a2 of determining the threshold of the parameters of the first rule layer, when the frequency of equipment failure is high, the equipment is more likely to fail again, while for equipment that has never failed, the probability of failure is relatively low. Therefore, the health of the equipment can be considered to be determined according to the number of past failures of the equipment. For example, when the equipment has never failed, the health of the equipment is set to 1, and when the number of failures is greater than 10, the health of the equipment is set to 0.5. The present disclosure does not make special limitations on this.

[0068] Specifically, for the third method a3 of determining the threshold of the parameters of the first rule layer, the threshold of the parameters corresponding to the fault type is comprehensively determined by combining the thresholds determined by the above two methods, enhancing the credibility of the threshold. For example, the average value of the first static threshold and the second static threshold is directly calculated as the third static threshold, or the weighted average value of the first static threshold and the second static threshold can be calculated as the third static threshold.

[0069] In some embodiments, the second rule layer includes a dynamic judgment rule in a multi-level key-value form constructed according to the equipment type, equipment model, fault type, and parameter combination. The thresholds of the parameters in the parameter combination of the second rule layer are dynamic thresholds, and the type of the dynamic threshold is the same as that of the static threshold. When the static threshold is a unilateral threshold, the dynamic threshold is also a unilateral threshold but with different numerical values. When the static threshold is a bilateral threshold, the dynamic threshold is also a bilateral threshold but with different upper and lower limit values of the bilateral threshold. The determination of the thresholds of the parameters in the parameter combination of the second rule layer includes: Step b1, obtain the thresholds of the parameters in the parameter combination determined by the first rule layer.

[0070] Specifically, when determining the thresholds of the parameters of the second rule layer for the first fault type, first obtain the thresholds of the parameters of the first rule layer determined by at least one of the above methods a1, a2, or a3 for the first fault type.

[0071] Step b2, calculate the coupling between the parameters in the parameter combination according to the historical fault data of the equipment to obtain a parameter correlation weight matrix.

[0072] Specifically, due to the strong coupling between the operating state data of steel mill equipment. For example, oxygen lance nodulation can cause fluctuations in oxygen pressure, and the oxygen pressure fluctuations can in turn affect the reactions in the furnace. The furnace reactions lead to changes in furnace temperature, and the furnace temperature changes can further exacerbate nodulation. Therefore, for the detection of each fault type, multiple parameters are selected for comprehensive analysis to avoid misjudgment of equipment faults. Moreover, in a period of time before a fault occurs, the same numerical combinations may also appear for the parameters corresponding to the fault. Therefore, in order to perform more accurate fault detection, the embodiments of the present disclosure calculate the dynamic thresholds for early warning corresponding to the occurrence of a fault through parameter coupling analysis, and use them to achieve early warning of faults and improve the effect of fault detection.

[0073] Assume that the static thresholds of the parameters of the first fault type obtained according to step b1 are: the static threshold of the current parameter is [TA1, TA2], and the static threshold of the temperature parameter is [TW1, TW2]. Then, calculate the coupling between the parameters in the parameter combination according to the historical fault data of the equipment, that is, calculate the coupling between the current parameter A and the temperature parameter B according to the historical fault data of the equipment. For example: obtain the first fault type that has occurred from the historical fault data, perform spatio-temporal alignment with the time of the first fault type, and extract the time series data of the current parameter and the temperature parameter N minutes before the first fault type occurs, and use this as the training data set; then, use the training data set to train and correct the initial weight matrix between the current parameter A and the temperature parameter B to obtain the final associated weight matrix W = [[W_AA, W_AB], [W_BA, W_BB]].

[0074] It should be noted that: the number of rows and columns of the parameter association weight matrix changes according to the parameters. That is, when the parameters corresponding to the first fault type involve three parameters: current, temperature, and pressure, the finally calculated associated weight matrix according to the historical fault data is a 3*3 matrix. When calculating the coupling between the parameters, machine learning algorithms such as supervised learning or semi-supervised learning can be used to train and obtain the parameter association weight matrix, and the present disclosure does not limit this.

[0075] Step b3, calculate the dynamic thresholds of the parameters according to the parameter association matrix and the thresholds of each parameter.

[0076] Specifically, first normalize the parameter association matrix, and then calculate the upper limit value or the lower limit value of the threshold of each parameter in combination with the coupling influence amount between the parameters. Taking the association weight matrix W of the first fault type as an example, step b3 specifically includes the following steps: Step b31, normalize the association weight matrix to obtain a normalized matrix w = [[w_aa, w_ab], [w_ba, w_bb]]; Step b32: Calculate the dynamic lower limit value of the current threshold according to the second formula, and calculate the dynamic upper limit value of the current threshold according to the third formula. The second formula is:

[0077] The third formula is:

[0078] Step b33: Calculate the dynamic lower limit value of the temperature threshold according to the fourth formula, and calculate the dynamic upper limit value of the temperature threshold according to the fifth formula. The fourth formula is:

[0079] The fifth formula is:

[0080] Where, and are dynamic coefficients. The value range of is 0.3 - 0.5, generally taking 0.3. The value range of is 0.8 - 1.2, generally taking 0.8.

[0081]

[0082] It should be noted that in the above embodiments, taking the two parameters of temperature and current of the first fault type as examples, steps b1 - b3 are described in detail. Similarly, according to the principle of determining the dynamic thresholds of the two parameters in steps b1 - b3, the dynamic thresholds of other two different parameters can be determined, and the dynamic thresholds of more than two parameters can also be determined. In addition, the parameter thresholds in the first rule layer and the parameter thresholds in the second rule layer need to be updated in real time according to the time process and the fault data occurring in real time of the device to ensure that the subsequent real-time fault detection has a high detection effect.

[0083] In the above method, a static threshold is determined by combining the aging degree or the health degree of the device in the first rule layer, improving the real-time detection accuracy of faults; a dynamic threshold is determined according to the correlation intensity of the associated parameters in the historical fault data in the second rule layer, realizing early warning of device faults; through the set first rule layer and second rule layer, double judgment of the static threshold and the dynamic threshold is carried out, realizing early warning of device faults while detecting faulty devices in real time, and improving the fault detection effect.

[0084] Step S14: Send a warning message indicating that a steel plant device has failed according to the fault data.

[0085] Specifically, it can be understood that the fault data includes the fault detection results of the first rule layer and the fault detection results of the second rule layer. Since the first rule layer is used to perform static threshold judgment on the standardized data and the second rule layer is used to perform dynamic threshold judgment on the standardized data, that is, the fault detection results of the first rule layer are used to indicate whether a device has failed, and the fault detection results of the second rule layer are used to indicate whether a device is about to fail. Therefore, when sending the warning message, different warning messages need to be sent according to the specific detection results to more comprehensively and accurately display the fault data to the staff and improve the warning effect.

[0086] In some embodiments, the fault data includes at least one of the fault data marked at the first level and the fault data marked at the second level. The fault data marked at the first level is the fault detection result of the first rule layer, and the fault data marked at the second level is the fault detection result of the second rule layer; the warning message includes a first warning message and a second warning message. The first warning message is used to indicate that a certain device has a certain type of fault, and the second warning message is used to indicate that a certain device is about to have a certain type of fault. At this time, step S14 "Send a warning message indicating that a steel plant device has failed according to the fault data" at least includes: Step S141: If the fault data is the fault data marked at the first level, send the first warning message; Step S142: If the fault data is the fault data marked at the second level, send the second warning message; Step S143: If the fault data is the fault data marked at the first level and the fault data is also the fault data marked at the second level, send the first warning message and the second warning message simultaneously.

[0087] Specifically, it can also be understood that the fault data that can be determined in step S13 includes various types of fault data determined through the first rule layer and various types of fault data determined through the second rule layer. That is, the finally obtained fault data includes the fault data determined according to two methods. It is possible that a certain fault of the device is detected by the second rule layer although no fault is detected by the first rule layer, which provides great convenience for equipment maintenance and greatly improves the fault detection result.

[0088] In some embodiments, when there are multiple fault types simultaneously, the sending priority of warning information can be set according to the fault type. For example, according to the degree of influence of the fault type on the device, there are three types of faults: high-level, medium-level, and low-level. When sending the warning information about the fault of the steel plant equipment, the warning information corresponding to the high-level fault type is preferentially sent and directly sent to the on-site operation large screen to be displayed to the staff in the fastest and most intuitive way. At the same time, it can also be sent to the electronic devices of the associated staff. For the warning of medium-level and low-level faults, it can be only sent to the electronic devices of the associated staff, or different warning methods can be set. The present disclosure does not make any limitation on this.

[0089] In the above method, by sending different warning information according to the marked type of the fault data, the fault that has occurred or is about to occur is warned more accurately, improving the warning effect. By combining the priority of the fault type to send warning information, the "high-level fault" with a high priority can be preferentially sent to the equipment of the staff or other relevant equipment, so that the staff can give priority to paying attention to such faults and perform rapid fault handling or equipment maintenance, enriching the fault warning method.

[0090] It should be noted that after sending the warning information about the fault of the steel plant equipment according to the fault data, the fault real-time detection method further includes: Step S15, generating index data according to the first fault detection result and the second fault detection result, where the index data is unmarked multi-source data, and generating an equipment status report according to the index data and the fault data.

[0091] Specifically, the equipment status report includes information such as the equipment health status, fault prediction result, maintenance record, etc. The equipment status report can be stored in the form of a chart or a dashboard, etc., which is convenient for users to clearly and intuitively view the operation status and fault trend of the equipment.

[0092] In the above embodiments, the data quality of the real-time operation status data is improved by performing normalization processing and timestamp alignment on the real-time operation status data; the effect of equipment fault detection is effectively improved by dividing the operation stages of the normalized data and setting different fault judgment models for different operation stages; the stability of real-time data processing is ensured by adaptively caching the data in each data queue on the cache device; the real-time, accurate and efficient detection of various equipment faults is achieved by performing dual judgments of static thresholds and dynamic thresholds on the normalized data in different operation stages, and at the same time, early warning of faults is realized through dynamic threshold judgment, and early warning is carried out before the faults occur, effectively improving the fault detection effect.

[0093] Figure 2 FIG. is a schematic flowchart of a processing method of a fault judgment model provided by the present disclosure. The fault judgment model includes a first rule layer and a second rule layer. The first rule layer is used to perform static threshold judgment on the normalized data, and the second rule layer is used to perform dynamic threshold judgment on the normalized data. The processing method of the fault judgment model refers to the method of processing the normalized data of the four data queues, that is, referring to Figure 2 In step S13, "inputting the normalized data of the four data queues into the corresponding equipment fault judgment model for processing to obtain fault data" specifically includes: Step S21, for any one of the data queues, obtain the multi-source data corresponding to the equipment model from the normalized data with timestamp alignment according to the equipment model.

[0094] Among them, the obtained multi-source data corresponding to the equipment model may be the multi-source data at the current moment or within a current period of time, such as 30s, 1min, 5min; the multi-source data refers to the obtained parameters of at least multiple types including temperature, current, vibration, pressure, etc. regarding the operation status of the same equipment.

[0095] Step S22, input the multi-source data into the first rule layer for static judgment of parameter thresholds, and output a first fault detection result.

[0096] Among them, the first fault data includes: unmarked multi-source data and / or first-level marked fault data. The first-level marked fault data refers to that fault marking and first-level marking are performed on abnormal data. The fault marking is the corresponding fault type, and the first-level marking is the corresponding fault determination method for the first rule layer.

[0097] Specifically, it can be understood that when using the first rule layer to judge multi-source data, if the parameters in the multi-source data do not meet the parameter thresholds set by the first rule layer, the fault judgment result "first fault data" is the multi-source data without marking; if the parameters in the multi-source data all meet the parameter thresholds set by the first rule layer, the fault judgment result "first fault data" is the fault data marked at the first level.

[0098] Specifically, it can also be understood that step S22 specifically includes the following steps: according to the device model, read the parameter combinations corresponding to various fault types under the device model from the first rule layer, and determine the static thresholds of the parameters in the parameter combinations; perform fault judgment based on the multi-source data corresponding to the device model and the static thresholds of the parameters, and generate a first-level fault mark.

[0099] In some embodiments, the multi-source data corresponding to the device model are parameters such as temperature, vibration, and pressure of a certain type of rolling mill. At this time, it is necessary to read the parameters to be monitored and their thresholds for each fault type under this device model from the first rule layer, and then judge between the data values of multiple parameters and the parameter thresholds according to the data type to determine whether the rolling mill has a fault. When making a specific judgment, multi-threaded concurrent technology can be used to simultaneously process the judgment between the data values of multiple parameters and the parameter thresholds to improve the processing speed, or other technologies can be used for processing. The present disclosure does not make special limitations on this.

[0100] Step S23, input the multi-source data into the second rule layer for dynamic judgment of parameter thresholds, and output a second fault detection result.

[0101] Among them, the second fault detection result includes: unmarked multi-source data and / or fault data marked at the second level. The fault data marked at the second level refers to that the abnormal data has been marked with a fault and a second-level mark. The fault mark is the corresponding fault type, and the second-level mark is the corresponding fault determination method for the second rule layer.

[0102] Specifically, it can be understood that when using the second rule layer to judge multi-source data, if the parameters in the multi-source data do not meet the parameter thresholds set by the second rule layer, the fault judgment result "second fault data" is the unmarked multi-source data; if the parameters in the multi-source data all meet the parameter thresholds set by the second rule layer, the fault judgment result "second fault data" is the fault data marked at the second level.

[0103] In some embodiments, step S23 specifically includes the following steps: According to the device model, read the parameter combinations corresponding to various fault types under the device model from the second rule layer, and determine the dynamic thresholds of the parameters in the parameter combination; perform a fault judgment based on the multi-source data corresponding to the device model and the dynamic thresholds of the parameters, and generate a secondary fault flag.

[0104] Specifically, it can also be understood that the first rule layer and the second rule layer are two independent layers. The multi-source data can be input into the first rule layer and the second rule layer simultaneously for static threshold judgment and dynamic threshold judgment, or the multi-source data can be input into the first rule layer and the second rule layer successively for static threshold judgment and dynamic threshold judgment. The present disclosure does not make any special limitations in this regard.

[0105] It should be noted that in the process of "performing a fault judgment based on the multi-source data corresponding to the device model and the static thresholds of the parameters to generate a primary fault flag" or "performing a fault judgment based on the multi-source data corresponding to the device model and the dynamic thresholds of the parameters", a decision rule number can be used for matching judgment, or other common methods can be used for matching judgment. The present disclosure does not make any limitations in this regard.

[0106] Step S24, determine the fault data corresponding to the multi-source data according to the first fault detection result and the second fault detection result.

[0107] Specifically, the fault data includes a fault type and a flag information. The flag information refers to whether a certain detected fault type is determined according to the first rule layer or the second rule layer. The flag information corresponding to the fault type determined according to the first rule layer is a primary flag, and the flag information corresponding to the fault type determined according to the second rule layer is a secondary flag.

[0108] In some embodiments, the third rule layer can also be used to replace the first rule layer and the second rule layer, that is, a device fault judgment model including the third rule layer is constructed for each data queue. The third rule layer is a multi-threshold judgment rule in the form of a multi-level key-value constructed according to the device type, device model, fault type, and parameter combination. The multi-thresholds of the parameters in the parameter combination of the third rule layer include: static thresholds and dynamic thresholds.

[0109] Among them, the static threshold is the threshold of each parameter in the parameter combination of the first rule layer, and the dynamic threshold is the threshold of each parameter in the parameter combination of the second rule layer. At this time, step S13 "input the standardized data of the four data queues into the corresponding device fault judgment model for processing to obtain fault data" specifically includes: for any data queue, obtain the multi-source data corresponding to the device model from the standardized data with aligned timestamps according to the device model; input the multi-source data into the third rule layer for multi-threshold judgment of parameter thresholds, and output the third fault detection result, where the third fault detection result includes unmarked multi-source data and / or first-level marked fault data and / or second-level marked fault data.

[0110] In the above embodiment, by performing parameter judgment of static thresholds on the standardized data in different operation stages, real-time, accurate and efficient detection of various device fault types is achieved. At the same time, early warning of faults is also achieved through dynamic threshold judgment, realizing early warning before the occurrence of faults, and effectively improving the fault detection effect.

[0111] Figure 3 is a schematic structural diagram of a fault real-time detection device provided by the present disclosure. Refer to Figure 3 , the fault real-time detection device 300 includes: a data acquisition module 310, a data cache module 320, a processing module 330, and an early warning module 340.

[0112] The data acquisition module 310 is configured to acquire the operation status data of at least one steel plant device, perform standardized processing on the operation status data to generate standardized data with aligned timestamps, and divide the standardized data into standardized data of four operation stages: raw material preparation, smelting and processing, forming and processing, and shutdown according to the operation stage. The data cache module 320 is configured to build four data queues on the cache device for receiving the standardized data of the four operation stages, calculate the data flow fluctuation coefficients of the four data queues respectively according to the inflow rate of the standardized data of the four data queues and the remaining capacity of the cache device, and adjust the cache strategy of the four data queues according to the data flow fluctuation coefficients. The processing module 330 is configured to input the standardized data of the four data queues into the corresponding device fault judgment model for processing to obtain fault data; among them, different data queues correspond to different device fault judgment models, and the device fault judgment model includes a first rule layer and a second rule layer. The first rule layer is used to perform static threshold judgment on the standardized data, and the second rule layer is used to perform dynamic threshold judgment on the standardized data. The early warning module 340 is configured to send an early warning message indicating that a steel plant device has a fault according to the fault data.

[0113] In some embodiments, the real-time fault detection device further includes: a report module, configured to generate metric data based on the first fault detection result and the second fault detection result, and generate a device status report based on the metric data and the fault data. Wherein, the metric data is unlabeled multi-source data, and the device status report includes information such as device health status, fault prediction results, maintenance records, etc. The device status report can be stored in the form of charts or dashboards, etc., to facilitate users to clearly and intuitively view the operating status and fault trends of the device.

[0114] For the detailed description of the above real-time fault detection device, please refer to the description of the relevant detection methods in the above embodiments. Repeated parts will not be elaborated. The embodiments of the real-time fault detection method and the real-time fault detection device described above are merely illustrative. The "module" described as a separated component can be a combination of software and / or hardware that implements a predetermined function, and may or may not be physically separated. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0115] Figure 4 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. The electronic device 400 includes: a processor 410 and a memory 420. The processor 410 and the memory 420 are connected through a bus 430. The memory 420 stores machine-readable instructions executable by the processor 410. The processor 410 reads the machine-readable instructions from the memory 420 and executes the machine-readable instructions to implement the real-time fault detection method provided by the embodiment of the present disclosure.

[0116] It should be understood that the electronic device can be an electronic device with logical computing functions such as a personal computer (PC), a tablet computer, a smart phone, etc.

[0117] The embodiment of the present disclosure also provides a computer-readable storage medium, which stores a computer program for executing the real-time fault detection method provided by the embodiment of the present disclosure.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present disclosure, and should all be covered by the protection scope of the present invention.

Claims

1. A real-time fault detection method, characterized in that: The method comprises: Acquire the operating status data of at least one steel plant equipment, and perform standardization processing on the operating status data to generate standardized data with time stamp alignment, and divide the standardized data into standardized data of four operating stages according to the operating stage: raw material preparation, smelting processing, forming processing, and shutdown; Constructing four data queues on the cache device for receiving the standardized data of the four operation stages, respectively calculating the data flow fluctuation coefficients of the four data queues according to the standardized data cache inflow rates of the four data queues and the remaining capacity of the cache device, and adjusting the cache strategies of the four data queues according to the data flow fluctuation coefficients; The standardized data of the four data queues are respectively input into the corresponding equipment fault judgment model for processing to obtain fault data; wherein different data queues correspond to different equipment fault judgment models and the equipment fault judgment model includes a first rule layer and a second rule layer, the first rule layer is used to perform static threshold judgment on the standardized data, and the second rule layer is used to perform dynamic threshold judgment on the standardized data; According to the fault data, early warning information of the steel plant equipment failure is sent.

2. The real-time fault detection method according to claim 1, characterized in that: The standardized data of the four data queues are respectively input into the corresponding equipment fault judgment model for processing to obtain fault data, specifically including: For any data queue, obtain multi-source data corresponding to the device model from the standardized data with aligned timestamps according to the device model; Inputting the multi-source data into the first rule layer to perform static judgment on parameter thresholds, and outputting a first fault detection result, wherein the first fault detection result includes: unlabeled multi-source data and / or first-level labeled fault data; Inputting the multi-source data into the second rule layer to perform dynamic judgment on parameter thresholds, and outputting a second fault detection result, wherein the second fault detection result includes: unlabeled multi-source data and / or secondary labeled fault data; Fault data corresponding to the multi-source data is determined according to the first fault detection result and the second fault detection result, where the fault data includes a fault type and marking information.

3. The real-time fault detection method according to claim 2, characterized in that: The first rule layer includes static judgment rules in the form of multi-level key values ​​constructed according to device type, device model, fault type, and parameter combination, wherein the determination of the threshold of each parameter in the parameter combination of the first rule layer includes at least one of the following: Determine the first static threshold of the parameter based on the basic threshold set in the equipment manufacturing specifications and the age of the equipment; or, Determine a second static threshold value of the parameter based on the historical fault data of the device and the health of the device; or, A third static threshold is determined according to the first static threshold and the second static threshold.

4. The real-time fault detection method according to claim 3, characterized in that: The second rule layer includes constructing a dynamic judgment rule in a multi-level key value form according to the device type, device model, fault type, and parameter combination. The determination of the threshold of each parameter in the parameter combination of the second rule layer includes: Obtaining the threshold of each parameter in the parameter combination determined by the first rule layer; Calculate the coupling between parameters in the parameter combination based on the historical fault data of the equipment, and obtain the parameter association weight matrix; The dynamic threshold of each parameter is calculated according to the parameter association matrix and the threshold of each parameter.

5. The real-time fault detection method according to claim 1, characterized in that: The calculating the data flow fluctuation coefficients of the four data queues respectively according to the standardized data cache inflow rates of the four data queues and the remaining capacity of the cache device includes: A flow monitoring sensor at a data transmission interface of a cache device is used to monitor the cache inflow rate of standardized data of four data queues; Using a capacity monitor integrated in the cache device to monitor the remaining capacity of the cache device, the remaining capacity of the cache device includes the remaining capacity of each data queue; According to the cache inflow rate and remaining capacity of the standardized data of each data queue, the data flow fluctuation coefficient of each data queue is calculated.

6. The real-time fault detection method according to claim 5, characterized in that: The data flow fluctuation coefficient of each data queue is calculated based on the cache inflow rate and the remaining capacity of the standardized data of each data queue, including: The data flow fluctuation coefficient of each data queue is calculated according to the first formula, where the first formula is: Among them, the data inflow rate standard deviation is the standard deviation of the data inflow rate within the preset time period, the data inflow rate average is the average inflow rate of the data inflow rate within the preset time period, α is the capacity weight coefficient, "1-remaining capacity percentage" is the cache capacity adjustment factor, and the remaining capacity percentage is the ratio between the remaining capacity of each data queue and the allocated capacity.

7. The real-time fault detection method according to claim 5, characterized in that: The adjusting the cache strategies of the four data queues according to the data flow fluctuation coefficient includes: According to the data flow fluctuation coefficient and fluctuation threshold of each data queue, the data cache inflow of each data queue is divided into a stable state, a violent fluctuation state, and an extreme fluctuation state; When the data cache inflow of any data queue is in an extremely fluctuating state, the standardized data to be cached in the data queue is temporarily stored in the cloud cache device using a high-speed data transmission link; When the data cache inflow of any data queue is in a state of violent fluctuation and there is at least one data queue in which the data cache inflow is in a stable state among the four data queues, the allocation amount of the data queue in which the data cache inflow is in a state of violent fluctuation is increased while the allocation amount of the data queue in which the data cache inflow is in a stable state is reduced; When the data cache inflow of any data queue is in a state of violent fluctuation and there is no data queue in the four data queues with a stable data cache inflow, the standardized data to be cached in the data queue is temporarily stored in the cloud cache device by using a high-speed data transmission link; When the data cache inflow of any data queue is in a stable state, the data that has not been accessed for a long time is cleared and the storage fragments of the local cache device are sorted.

8. A real-time fault detection device, characterized in that: The device comprises: A data acquisition module is used to acquire the operating status data of at least one steel plant equipment, and to perform standardization processing on the operating status data to generate standardized data with time stamp alignment, and to divide the standardized data into standardized data of four operating stages according to the operating stage: raw material preparation, smelting processing, forming processing, and shutdown; A data cache module, used to construct four data queues on a cache device for receiving standardized data of the four operation stages, calculate data flow fluctuation coefficients of the four data queues according to standardized data cache inflow rates of the four data queues and remaining capacity of the cache device, and adjust cache strategies of the four data queues according to the data flow fluctuation coefficients; A processing module, used for inputting the standardized data of the four data queues into the corresponding equipment fault judgment model for processing to obtain fault data; wherein different data queues correspond to different equipment fault judgment models and the equipment fault judgment model includes a first rule layer and a second rule layer, the first rule layer is used for performing static threshold judgment on the standardized data, and the second rule layer is used for performing dynamic threshold judgment on the standardized data; The early warning module is used to send early warning information of steel plant equipment failure based on the fault data.

9. An electronic device, characterized in that: The electronic device includes: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to read the machine-readable instructions from the memory and execute the machine-readable instructions to implement the real-time fault detection method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is used to execute the real-time fault detection method described in any one of claims 1 to 7.

Images