Intelligent management method and system for cement factory equipment management
By collecting and processing equipment parameters in cement plants, regression models and multi-objective optimization models were established, solving the problems of manual dependence and information silos in cement plant equipment management. This enabled predictive maintenance and refined management of equipment, improving management efficiency and reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGCAI BANGYE (HANGZHOU) INTELLIGENT TECH CO LTD
- Filing Date
- 2026-01-19
- Publication Date
- 2026-05-12
AI Technical Summary
Cement plant equipment management relies on manual operation, lacks a standardized online system, and suffers from insufficient data processing quality, resulting in information silos that cannot support decision-making and execution. The management model is mainly based on post-event maintenance.
Collect process parameters, perform preprocessing, establish equipment regression models, extract data features, construct multi-objective optimization models, and output maintenance decision reports to achieve predictive maintenance and comprehensive decision-making for equipment.
Improve equipment management efficiency, enhance operational reliability, reduce unplanned downtime and maintenance costs, and promote the transformation of equipment management from passive to proactive and refined.
Smart Images
Figure CN122022291A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of equipment management technology, and in particular to an intelligent management method and system for equipment management in cement plants. Background Technology
[0002] Currently, equipment management in cement plants generally adopts a traditional, manual-dominated model. Core management processes, such as equipment file storage and updating, task assignment and process recording for inspections, and measurement, statistics, and analysis of operational data, all rely on manual operation. Records are mostly kept in paper ledgers or spreadsheets. The deployment and coordination of daily equipment management work are mainly carried out offline through pre-shift meetings, telephone communication, and WeChat messages. The discovery of potential equipment hazards highly depends on the on-site experience and sense of responsibility of the staff, and the overall management model is mainly based on reactive maintenance.
[0003] However, the aforementioned existing technologies may suffer from over-reliance on offline workflows, resulting in a lack of standardized online systems and difficulty in controlling related decision-making and execution. Furthermore, data processing may lack quality and rely on manual intervention, creating "information silos" that also fail to support decision-making and execution. Summary of the Invention
[0004] To address the problems existing in the prior art, embodiments of the present invention provide an intelligent management method and system for equipment management in cement plants.
[0005] This invention provides an intelligent management method for equipment management in a cement plant, the method comprising:
[0006] The process parameters are collected and preprocessed. The data tags of the process parameters are determined by comparing them with the rule tags of the preset rule base. The data tags include steady-state data and non-steady-state data. The non-steady-state data further includes abnormal operating condition data.
[0007] Obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, establish a device regression model. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and then outputs the health status of the real-time parameters.
[0008] Obtain preceding data slices within a specific time window before the period corresponding to the abnormal operating condition data, extract data features from the preceding data slices, store the data features in a pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison results, and then trigger the corresponding hidden danger alarm.
[0009] A multi-objective optimization model is constructed for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. The total expected cost of each maintenance route in the model is solved, and then a maintenance decision report is output.
[0010] In one embodiment, the method further includes:
[0011] A regression model is established using the steady-state data as the independent variable and the equipment parameters as the dependent variable. The calculation formula of the regression model includes:
[0012] Y = f(X) + ε
[0013] Where Y represents the equipment parameters, f(X) is the expected normal value of the equipment determined by the independent variable X, and ε represents the residual.
[0014] In one embodiment, the method further includes:
[0015] Calculate the residual sequence between the actual values and predicted values of equipment parameters in historical steady-state data, and calculate the standard deviation of the historical residual sequence;
[0016] The residual control range of the corresponding parameter is determined by comparing the standard deviation with the preset health center line and the difference between the two.
[0017] In one embodiment, the data features include:
[0018] Time-domain features, frequency-domain features, and time-frequency domain features;
[0019] The step of storing data features into a pattern feature library includes:
[0020] Based on the fault type corresponding to the abnormal operating condition data, a feature extraction tendency is determined. Based on the feature extraction tendency, key features in the time domain features, frequency domain features, and time and frequency domain features are determined, and the key features are stored in the pattern feature library.
[0021] In one embodiment, the method further includes:
[0022] Obtain the fault information of the device corresponding to the hidden danger alarm, and then obtain the task data, resource data, and economic cost at the current time;
[0023] The system pre-determines the possible repair routes for each fault mode of the equipment and calculates the total expected cost of each repair route under the corresponding constraints. The total expected cost includes risk cost, direct repair cost, and downtime cost.
[0024] A corresponding structured decision report is generated based on the overall expectation.
[0025] This invention provides an intelligent management system for equipment management in a cement plant, the system comprising:
[0026] The data acquisition module is used to acquire process parameters and perform preprocessing steps, compare the rule tags of the preset rule base, and determine the data tags of the process parameters. The data tags include steady-state data and non-steady-state data, and the non-steady-state data further includes abnormal operating condition data.
[0027] The regression module is used to obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, a device regression model is established. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and then outputs the health status of the real-time parameters.
[0028] The matching module is used to obtain the preceding data slice within a specific time window before the period corresponding to the abnormal working condition data, extract the data features of the preceding data slice, store the data features in the pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison results, and then trigger the corresponding hidden danger alarm.
[0029] The decision-making module is used to construct a multi-objective optimization model for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. The module solves the total expected cost of each maintenance route in the model and then outputs a maintenance decision report.
[0030] In one embodiment, the system further includes:
[0031] The tendency module is used to determine the feature extraction tendency based on the fault type corresponding to the abnormal operating condition data, determine the key features in the time domain, frequency domain and time and frequency domain based on the feature extraction tendency, and store the key features in the pattern feature library.
[0032] In one embodiment, the system further includes:
[0033] The maintenance module is used to obtain fault information of the equipment corresponding to the hidden danger alarm, and then obtain the task data, resource data, and economic cost at the current time;
[0034] The cost module is used to preset the optional maintenance routes for each failure mode of the equipment and calculate the total expected cost of each maintenance route under the corresponding constraints. The total expected cost includes risk cost, direct maintenance cost and downtime cost.
[0035] The reporting module is used to generate a corresponding structured decision report based on the overall expectation.
[0036] This invention provides an electronic device, including a processor and a memory;
[0037] The processor is connected to the memory;
[0038] The memory is used to store executable program code;
[0039] The processor runs a program corresponding to the executable program code stored in the memory to perform the methods described in one or more embodiments.
[0040] This invention provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described intelligent management method for cement plant equipment management.
[0041] In view of the above, in one or more embodiments of this specification, process parameters are collected and preprocessed. The rule labels of the preset rule base are compared to determine the data labels of the process parameters. The data labels include steady-state data and non-steady-state data, with non-steady-state data further including abnormal operating condition data. Equipment parameters corresponding to the timestamps of the steady-state data are obtained, and the residuals of the equipment health status are determined. An equipment regression model is established based on the steady-state data and residuals. The equipment regression model outputs the health deviation residuals corresponding to the real-time parameters of the equipment, and outputs the health status of the real-time parameters. A slice of preceding data within a specific time window before the period corresponding to the abnormal operating condition data is obtained. Data features of the preceding data slices are extracted and stored in a pattern feature library. The obtained equipment parameters are continuously matched with the pattern feature library. A similarity score is determined based on the comparison results, thereby triggering the corresponding hidden danger alarm. A multi-objective optimization model for the equipment corresponding to the hidden danger alarm is constructed. The inputs of the multi-objective optimization model include fault information, task data, resource data, and economic costs. The total expected cost of each maintenance route in the model is solved, and a maintenance decision report is output. This eliminates tedious processes such as manual statistics, analysis, and judgment, significantly improving equipment management efficiency. At the same time, data integration enables predictive maintenance of equipment and early warning through comprehensive decision-making, enhancing equipment operational reliability, reducing unplanned downtime and maintenance costs, and extending equipment lifespan. Intelligent methods provide precise support for equipment management and enterprise decision-making, promoting the transformation of traditional passive equipment management to proactive and refined management. Attached Figure Description
[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0043] Figure 1 This is a flowchart of an intelligent management method for cement plant equipment management provided in one embodiment of this specification.
[0044] Figure 2 This is a schematic diagram of the structure of an intelligent management system for equipment management in a cement plant, provided in one embodiment of this specification.
[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this specification. Detailed Implementation
[0046] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the various examples. For example, the described methods may be performed in a different order than described, and steps may be added, omitted, or combined. Furthermore, features described in some examples may be combined in other examples.
[0047] As used herein, the term "comprising" and its variations are open terms meaning "including but not limited to". The term "based on" means "at least partially based on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other definitions, whether explicit or implicit, may be included below. Unless explicitly indicated by the context, the definition of a term shall remain consistent throughout the specification.
[0048] like Figure 1 As shown, this embodiment of the invention provides an intelligent management method for cement plant equipment management, including:
[0049] Step S102: Collect process parameters and perform preprocessing steps. Compare the rule tags in the preset rule base to determine the data tags of the process parameters. The data tags include steady-state data and non-steady-state data. The non-steady-state data further includes abnormal operating condition data.
[0050] Specifically, in the continuous and highly coupled production process of a cement plant, operating conditions change frequently due to operations such as feeding, shutdown, output adjustment, and coal type switching, leading to significant and drastic fluctuations in equipment operating parameters. Therefore, when collecting raw time-series data within a cement plant, it is necessary to decouple the process data representing the "production status" from the detection data representing the "equipment status" to identify operating conditions and classify the data. Data sources can include cement plant process parameters such as feed rate, kiln speed, kiln main drive power / current, key fan speed / current, total system air pressure, and grate speed. These process parameters directly reflect the overall compliance and operating status of the production line. Preprocessing steps are then performed, including but not limited to: outlier removal, such as filtering values with obvious physical unreliability caused by momentary sensor malfunctions (e.g., temperature jumps to extremes); data alignment, ensuring precise synchronization of data from different systems on the time axis (typically aligned to the second level); and smoothing and noise reduction, performing simple moving averages or low-pass filtering on key parameters to eliminate high-frequency electronic noise.
[0051] Furthermore, a pre-defined rule base is obtained, whose rules include labels for different data. For example, if (feed rate standard deviation < threshold A) and (kiln speed standard deviation < threshold A) and (|feed rate trend slope| < threshold B), the corresponding data label can be "steady-state condition"; or if (feed rate mean > feeding threshold) and (feed rate trend slope > rising threshold), the corresponding data label is "start-up feeding transition condition". Key characteristics of process parameters (feed rate, kiln speed) are calculated over a fixed length T (e.g., 10 minutes), including the mean (average load level), standard deviation (degree of fluctuation), and linear trend slope (whether the parameter is rising, falling, or stable). The rules in the rule base can be clustered based on historical data to determine the data requirements corresponding to different data label conditions, or they can be preset by relevant personnel. Then, the rule base is used to logically judge the process parameters and determine the corresponding label type, ensuring the timeliness of the judgment. In addition, cluster analysis can be performed on the rules in the rule base and their corresponding process parameters to check whether the categories divided by the existing rules are consistent with the categories naturally clustered by the data, thereby identifying new transitional operating condition subclasses and determining adjustment suggestions for the parameters in the rule base (such as thresholds A and B).
[0052] Furthermore, based on the process parameters with established data labels, for steady-state data, all data segments marked as steady-state on the continuous time axis can be segmented to form steady-state data slices. Each steady-state slice records metadata such as its start and end times, average load level, and duration. For non-steady-state data, data further labeled as transitional operating conditions can be analyzed to examine the dynamic response characteristics of the equipment under load changes, as well as data on abnormal operating conditions, which is crucial data for constructing a fault mode library.
[0053] Step S104: Obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, establish a device regression model. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and outputs the health status of the real-time parameters.
[0054] Specifically, based on the steady-state data slices output in step S102, key data pairs are identified, categorized into independent and dependent variables. The independent variables (averaged process parameters strongly correlated with the equipment during the slice period) are the process parameters in the steady-state data slice: for the rotary kiln main motor: feed rate, kiln speed, kiln torque; for the high-temperature blower: system negative pressure (kiln tail pressure), gas temperature, blower speed. The dependent variables are the average or feature-extracted equipment state parameters at the same time stamp during the slice period. Examples include: effective vibration velocity, bearing temperature, effective motor current, and vibration energy in a specific frequency band. Then, the relationship between process parameters and equipment state is quantified using a regression model, and health standards are defined through residual analysis, including:
[0055] For different key status parameters of different equipment, a corresponding regression model Y=f(X)+ε is established. Here, f(X) represents the "expected normal value" determined by the independent variable X, and ε represents the residual (i.e., the deviation caused by the equipment's own health status). Then, the calculation formula for Y is determined through multiple linear regression (for linearly correlated parameters), i.e., Y=β0+β1X1+β2X2+...+β n X n Furthermore, for parameters with nonlinear correlations, multinomial regression or generalized additive models can be used to determine the relationship between equipment state parameters. The model training process can use long-term (e.g., one year) historical steady-state slice data, divided chronologically into a training set (e.g., the first 8 months) and a test set (the last 4 months). The model is fitted on the training set, aiming to minimize the error between predicted and actual values. The model performance is evaluated on the test set.
[0056] Model f(X) defines the "expected normal value" of the equipment, while the health standard is determined based on the difference between the actual and predicted Y values of the residual ε. The difference between the actual and predicted Y values in historical steady-state slices is calculated to obtain the historical residual sequence. A health centerline is preset (i.e., the actual state perfectly matches the process expectation). The standard deviation σ of the historical residual sequence is calculated. Combining the difference between the standard deviation and the health centerline, the degree of deviation of the equipment from the expected normal result is judged based on the difference, i.e., the residual control range. For example, if it is within ±2σ, it means the equipment is within the health standard; if it is outside ±2σ or within ±3σ, the equipment begins to deviate from the health standard; and if it is outside ±3σ, it means the equipment has completely deviated from the health benchmark. Furthermore, for the real-time process parameter Xreal-time, the corresponding model Ypredicted = f(Xreal-time) can be immediately invoked. Then, the actual Y is compared with the predicted Y to obtain the real-time residual. This allows us to determine the regression model corresponding to the parameters of each piece of equipment, and each parameter also has a corresponding residual control range for the degree of deviation. This helps us determine the health status of each parameter for each piece of equipment, and based on the health status, we can determine the preliminary warning results, indicating the degree to which the equipment status deviates from the normal baseline under its process conditions.
[0057] Step S106: Obtain the preceding data slice within a specific time window before the period corresponding to the abnormal operating condition data, extract the data features of the preceding data slice, store the data features in the pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison results, and then trigger the corresponding hidden danger alarm.
[0058] Specifically, based on the time-series data with abnormal operating condition labels output in step S102, a preceding steady-state data slice is determined within a specific time window before the fault occurs (e.g., 24 hours or 3 days before the fault). The original signal is then filtered and demodulated to highlight fault characteristics and suppress background noise. Data features of the continuous steady-state data slices within the specific event window are extracted, including but not limited to: time-domain features (calculating statistical features that characterize the waveform morphology within the data segment, such as kurtosis, peak factor, waveform factor, and impulse factor), frequency-domain features (obtaining the spectrum through fast Fourier transform and calculating the energy proportion, main frequency amplitude ratio, and sideband energy of specific frequency bands), and time-frequency-domain features (using continuous wavelet transform or Hilbert-Huang transform to generate the time-frequency spectrum of the signal and extracting two-dimensional image features of energy distribution in the time and frequency dimensions). After standardizing the extracted features, they are stored in the pattern feature library along with their corresponding fault type, device ID, and development stage (e.g., "3 days before the fault").
[0059] Furthermore, in actual processing, due to different fault mechanisms, the corresponding feature extraction tendency can be determined according to different fault types. For example, for local damage faults of bearings and gears, the core feature is the characteristic frequency and its harmonic amplitude in the demodulation spectrum; for faults such as imbalance, misalignment, and looseness, the amplitude and phase of the power frequency and its harmonics in the vibration spectrum are preferred; for dynamic process or impact faults (such as grate plate breakage in a grate cooler), wavelet transform can be used to extract energy at a specific scale as a feature. A fault-feature selection lookup table can be preset to determine the feature types with higher weighting in subsequent comparison steps.
[0060] For real-time acquired process parameters, continuous comparisons are performed with the pattern feature library. This can be achieved using a sliding window method to calculate a feature set (with the same time-domain and frequency-domain indices) that is identical to the pattern feature library for the real-time process parameters in the steady-state data window. The continuous comparison methods include: for waveform morphology in the time domain, dynamic time warping (DTW) algorithms can be used to calculate morphological similarity for comparison; for spectral features in the frequency domain, cosine similarity or Euclidean distance can be used to calculate the similarity between the real-time feature vector and each pattern vector in the library; for time-frequency domain features such as the time-spectrum graph, lightweight convolutional neural networks can be used for feature extraction and similarity comparison. Based on the above comparison results, the similarity scores between the real-time features and each feature in the pattern feature library are calculated in parallel. When the highest similarity score exceeds a preset threshold, a potential fault alarm is triggered. The alarm includes the matched fault mode type, the current similarity score, the historical fault development stage corresponding to the mode, and suggested inspection locations and measures.
[0061] Step S108: Construct a multi-objective optimization model for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. Solve the total expected cost of each maintenance route in the model, and then output a maintenance decision report.
[0062] Specifically, while the preceding steps establish early warning results, in actual operation, a technical warning does not necessarily equate to an immediate maintenance order. Given the multiple constraints of a cement plant, the decision to maintain or continue operation requires consideration of various factors. This necessitates seeking a dynamic optimal balance between equipment reliability, production continuity, and operational economics. Specifically, this is achieved by: obtaining the potential hazard alarm from step S106; based on this alarm, further fault information can be determined, including but not limited to: equipment ID, fault mode, severity level, predicted remaining useful life and its confidence interval, fault development rate estimation, etc. Then, operational data within the current system is acquired, including task data (detailed production schedules for the next few days to weeks, order lists, key delivery dates, etc.), resource data (spare parts inventory, manpower, and collaborative equipment status), and economic costs, including but not limited to equipment downtime costs, preventative maintenance costs, reactive maintenance costs, and spare parts unit prices.
[0063] Furthermore, multiple optional maintenance routes are preset for each failure mode. These routes range from immediate repair to operation to complete failure, encompassing multiple paths. Each maintenance route includes not only its action type but also resource requirements, execution effects, and execution constraints. The total expected cost of each maintenance route can include three parts: 1. Risk cost, calculated based on the probability distribution of remaining lifetime prediction (usually a normal distribution), multiplied by the probability of a sudden failure occurring before the decision maintenance time point, multiplied by the total loss caused by the sudden failure (including greater maintenance costs and production losses); 2. Direct maintenance costs, including manpower and spare parts costs; 3. Downtime cost, the amount converted from planned production losses caused by the selected maintenance time window. When calculating these costs, corresponding time constraints (maintenance actions must be completed before the failure occurs), resource constraints (spare parts, tools, and personnel are available within the selected time window), production constraints (business is in a coordinated downtime or low-load state), and logical constraints (some maintenance actions have sequential dependencies) are determined. For the calculation of the total expected cost, a genetic algorithm can be used to search for a high-quality non-dominated solution set in the vast solution space, i.e., a scheme that achieves different balances in cost, risk, and production impact. This process generates the optimal solution and multiple suboptimal solutions, or a corresponding structured decision report comparing the total cost, risk probability, estimated downtime, and critical resource requirements of each solution in tabular and graphical form. After maintenance, the actual time, cost, and resources used in the maintenance process, as well as the actual operating time of the equipment after maintenance, are recorded. This data is then used to further optimize the models corresponding to maintenance costs and fault prediction.
[0064] This invention provides an intelligent management method for cement plant equipment. The method involves collecting process parameters and performing preprocessing steps. It compares these parameters with rule tags in a pre-defined rule base to determine data tags for the process parameters. These data tags include steady-state and non-steady-state data, with non-steady-state data further including abnormal operating condition data. The method then acquires equipment parameters corresponding to the timestamps of the steady-state data to determine the residuals of the equipment's health status. Based on the steady-state data and residuals, an equipment regression model is established. This model outputs the health deviation residuals corresponding to the real-time equipment parameters, thus showing the health status of the real-time parameters. The method also acquires preceding data slices within a specific time window before the period corresponding to the abnormal operating condition data, extracts data features from these slices, stores these features in a pattern feature library, and continuously matches the acquired equipment parameters with the pattern feature library. Based on the comparison results, a similarity score is determined, triggering corresponding hazard alarms. Finally, a multi-objective optimization model is constructed for the equipment corresponding to the hazard alarm. The inputs to the multi-objective optimization model include fault information, task data, resource data, and economic costs. The model calculates the total expected cost of each maintenance route and outputs a maintenance decision report. This eliminates tedious processes such as manual statistics, analysis, and judgment, significantly improving equipment management efficiency. At the same time, data integration enables predictive maintenance of equipment and early warning through comprehensive decision-making, enhancing equipment operational reliability, reducing unplanned downtime and maintenance costs, and extending equipment lifespan. Intelligent methods provide precise support for equipment management and enterprise decision-making, promoting the transformation of traditional passive equipment management to proactive and refined management.
[0065] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of an intelligent management system for equipment management in a cement plant, provided in an embodiment of this application. Figure 2 As shown, the system includes:
[0066] The acquisition module S202 is used to acquire process parameters and perform preprocessing steps, compare the rule tags of the preset rule base, and determine the data tags of the process parameters. The data tags include steady-state data and non-steady-state data, and the non-steady-state data further includes abnormal operating condition data.
[0067] The regression module S204 is used to obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, a device regression model is established. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and outputs the health status of the real-time parameters.
[0068] The matching module S206 is used to obtain the preceding data slice within a specific time window before the period corresponding to the abnormal working condition data, extract the data features of the preceding data slice, store the data features in the pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison result, and then trigger the corresponding hidden danger alarm.
[0069] The decision module S208 is used to construct a multi-objective optimization model for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. The model solves the total expected cost of each maintenance route and then outputs a maintenance decision report.
[0070] In another embodiment, an intelligent management system for cement plant equipment management further includes:
[0071] The tendency module is used to determine the feature extraction tendency based on the fault type corresponding to the abnormal operating condition data, determine the key features in the time domain, frequency domain and time and frequency domain based on the feature extraction tendency, and store the key features in the pattern feature library.
[0072] In another embodiment, an intelligent management system for cement plant equipment management further includes:
[0073] The maintenance module is used to obtain fault information of the equipment corresponding to the hidden danger alarm, and then obtain the task data, resource data, and economic cost at the current time;
[0074] The cost module is used to preset the optional maintenance routes for each failure mode of the equipment and calculate the total expected cost of each maintenance route under the corresponding constraints. The total expected cost includes risk cost, direct maintenance cost and downtime cost.
[0075] The reporting module is used to generate a corresponding structured decision report based on the overall expectation.
[0076] Those skilled in the art will clearly understand that the technical solutions of the embodiments of this application can be implemented by means of software and / or hardware. In this specification, "unit" and "module" refer to software and / or hardware that can independently complete or cooperate with other components to complete a specific function, wherein the hardware may be, for example, a field-programmable gate array (FPGA), an integrated circuit (IC), etc.
[0077] Each processing unit and / or module in the embodiments of this application can be implemented by an analog circuit that implements the functions described in the embodiments of this application, or by software that executes the functions described in the embodiments of this application.
[0078] See Figure 3 It shows a schematic diagram of the structure of an electronic device according to an embodiment of this application, which can be used to implement... Figure 1 The method in the illustrated embodiment. (As shown) Figure 3 As shown, the electronic device 300 may include: at least one processor 301, at least one network interface 304, user interface 303, memory 305, and at least one communication bus 302.
[0079] The communication bus 302 is used to enable communication between these components.
[0080] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.
[0081] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).
[0082] The processor 301 may include one or more processing cores. The processor 301 connects to various parts within the electronic device 300 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling data stored in the memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.
[0083] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. Figure 3 As shown, the memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and program instructions.
[0084] exist Figure 3 In the illustrated electronic device 300, the user interface 303 is mainly used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call the image-based interactive application stored in the memory 305 and specifically perform the following operations: acquire process parameters and perform preprocessing steps, compare the rule labels of the preset rule base to determine the data labels of the process parameters, the data labels include steady-state data and non-steady-state data, and the non-steady-state data further includes abnormal operating condition data; acquire the equipment parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the equipment health status; establish an equipment regression model based on the steady-state data and the residual; and the equipment... The regression model outputs the health deviation residuals corresponding to the real-time parameters of the equipment, thus outputting the health status of the real-time parameters. It acquires preceding data slices within a specific time window before the period corresponding to the abnormal operating condition data, extracts the data features of these preceding data slices, stores the data features in a pattern feature library, continuously matches the acquired equipment parameters with the pattern feature library, determines the similarity score based on the comparison results, and then triggers the corresponding hidden danger alarm. A multi-objective optimization model is constructed for the equipment corresponding to the hidden danger alarm. The inputs of the multi-objective optimization model include fault information, task data, resource data, and economic costs. The model solves for the total expected cost of each maintenance route, and then outputs a maintenance decision report.
[0085] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.
[0086] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.
[0087] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.
[0088] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.
[0089] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0090] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0091] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.
[0092] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.
[0093] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.
Claims
1. An intelligent management method for equipment management in a cement plant, the method comprising: The process parameters are collected and preprocessed. The data tags of the process parameters are determined by comparing them with the rule tags of the preset rule base. The data tags include steady-state data and non-steady-state data. The non-steady-state data further includes abnormal operating condition data. Obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, establish a device regression model. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and then outputs the health status of the real-time parameters. Obtain preceding data slices within a specific time window before the period corresponding to the abnormal operating condition data, extract data features from the preceding data slices, store the data features in a pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison results, and trigger the corresponding hidden danger alarm. A multi-objective optimization model is constructed for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. The total expected cost of each maintenance route in the model is solved, and then a maintenance decision report is output.
2. The method according to claim 1, characterized in that, The method further includes: A regression model is established using the steady-state data as the independent variable and the equipment parameters as the dependent variable. The calculation formula of the regression model includes: Y = f(X) + ε Where Y represents the equipment parameters, f(X) is the expected normal value of the equipment determined by the independent variable X, and ε represents the residual.
3. The method according to claim 2, characterized in that, The method further includes: Calculate the residual sequence between the actual values and predicted values of equipment parameters in historical steady-state data, and calculate the standard deviation of the historical residual sequence; The residual control range of the corresponding parameter is determined by comparing the standard deviation with the preset health center line and the difference between the two.
4. The method according to claim 1, characterized in that, The data features include: Time-domain features, frequency-domain features, and time-frequency domain features; The step of storing data features into a pattern feature library includes: Based on the fault type corresponding to the abnormal operating condition data, a feature extraction tendency is determined. Based on the feature extraction tendency, key features in the time domain features, frequency domain features, and time and frequency domain features are determined, and the key features are stored in the pattern feature library.
5. The method according to claim 1, characterized in that, The method further includes: Obtain the fault information of the device corresponding to the hidden danger alarm, and then obtain the task data, resource data, and economic cost at the current time; The system pre-determines the possible repair routes for each fault mode of the equipment and calculates the total expected cost of each repair route under the corresponding constraints. The total expected cost includes risk cost, direct repair cost, and downtime cost. A corresponding structured decision report is generated based on the overall expectation.
6. An intelligent management system for equipment management in a cement plant, characterized in that, The system includes; The data acquisition module is used to acquire process parameters and perform preprocessing steps, compare the rule tags of the preset rule base, and determine the data tags of the process parameters. The data tags include steady-state data and non-steady-state data, and the non-steady-state data further includes abnormal operating condition data. The regression module is used to obtain the device parameters corresponding to the timestamp of the steady-state data, and then determine the residual of the device health status. Based on the steady-state data and the residual, a device regression model is established. The device regression model outputs the health deviation residual corresponding to the real-time parameters of the device, and then outputs the health status of the real-time parameters. The matching module is used to obtain the preceding data slice within a specific time window before the period corresponding to the abnormal working condition data, extract the data features of the preceding data slice, store the data features in the pattern feature library, continuously match the obtained equipment parameters with the pattern feature library, determine the similarity score based on the comparison results, and trigger the corresponding hidden danger alarm. The decision-making module is used to construct a multi-objective optimization model for the equipment corresponding to the hidden danger alarm. The input of the multi-objective optimization model includes fault information, task data, resource data and economic cost. The module solves the total expected cost of each maintenance route in the model and then outputs a maintenance decision report.
7. The system according to claim 6, characterized in that, The system also includes: The tendency module is used to determine the feature extraction tendency based on the fault type corresponding to the abnormal operating condition data, determine the key features in the time domain, frequency domain and time and frequency domain based on the feature extraction tendency, and store the key features in the pattern feature library.
8. The system according to claim 6, characterized in that, The system also includes: The maintenance module is used to obtain fault information of the equipment corresponding to the hidden danger alarm, and then obtain the task data, resource data, and economic cost at the current time; The cost module is used to preset the optional maintenance routes for each failure mode of the equipment and calculate the total expected cost of each maintenance route under the corresponding constraints. The total expected cost includes risk cost, direct maintenance cost and downtime cost. The reporting module is used to generate a corresponding structured decision report based on the overall expectation.
9. An electronic device, comprising a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, in order to perform the method as described in any one of claims 1-5.
10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of claims 1-5.