Smelting furnace production risk assessment method, device and equipment and storage medium

Abstract

The invention is suitable for the technical field of risk assessment, and provides a smelting furnace production risk assessment method, device and equipment and a storage medium, and the method comprises the steps: obtaining historical operation data, historical accident records and historical equipment fault frequency of a target smelting furnace; labeling the historical operation data based on the historical accident record to obtain labeling information, and constructing a training sample data set based on the historical operation data, the historical equipment fault frequency and the labeling information; based on the training sample data set, a risk assessment model is constructed and trained, and the risk assessment model comprises an operation data prediction model and a fault probability prediction model; the current operation data and the current equipment fault frequency of the target smelting furnace are obtained and input to the risk assessment model, and a risk assessment result is obtained based on two times of prediction of the operation data prediction model and the fault probability prediction model, the risk assessment result comprises the target fault prediction probability, and the accuracy of smelting furnace production risk assessment is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of risk assessment, and in particular, relates to a smelting furnace production risk assessment method, device, equipment and storage medium. Background Art

[0002] In industrial production, the safety and stability of the melting furnace, as a key production equipment, are directly related to production efficiency and product quality. Since the melting process involves many complex factors such as high temperature, high pressure, and chemical reactions, risk control in the production process becomes a challenge. In addition, the melting furnace is a common industrial equipment, widely used in metallurgy, chemical industry, construction and other fields. If the operation is improper or the equipment fails, safety accidents such as explosions and fires may occur during the operation of the melting furnace. Therefore, it is particularly important to conduct an effective risk assessment of the melting furnace production process.

[0003] Traditional smelting furnace risk assessment methods mainly rely on manual experience and rules, which are subjective and limited. At the same time, due to the complexity and variability of smelting furnace operation data, manual processing and analysis are inefficient, making it difficult to discover hidden safety hazards, resulting in low accuracy of risk assessment.

[0004] Therefore, how to improve the accuracy of smelting furnace production risk assessment has become an urgent problem to be solved. Summary of the invention

[0005] The embodiments of the present application provide a smelting furnace production risk assessment method, device, equipment and storage medium, aiming to improve the accuracy of smelting furnace production risk assessment.

[0006] In a first aspect, an embodiment of the present application provides a method for assessing the production risk of a smelting furnace, the method comprising: obtaining historical operating data, historical accident records, and historical equipment failure frequencies of a target smelting furnace, the historical operating data comprising one or more of temperature, pressure, and vibration; annotating the historical operating data based on the historical accident records to obtain annotation information, and constructing a training sample data set based on the historical operating data, the historical equipment failure frequencies, and the annotation information; based on the training sample data set, constructing and training a risk assessment model with the operating data and equipment failure frequencies of the target smelting furnace as input and the failure prediction probability as output, the risk assessment model comprising an operating data prediction model and a failure probability prediction model; obtaining the current operating data and the current equipment failure frequency of the target smelting furnace, and inputting them into the risk assessment model, and obtaining a risk assessment result of the target smelting furnace based on two predictions of the operating data prediction model and the failure probability prediction model, the risk assessment result comprising a target failure prediction probability.

[0007] In a possible implementation, based on the training sample data set, constructing and training a risk assessment model with the operating data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, including: calculating the historical operating data change value based on the historical operating data, and the historical operating data change value includes the historical temperature change rate and the historical pressure fluctuation amplitude; training a deep learning neural network model based on the historical operating data and the historical operating data change value to obtain the operating data prediction model with the operating data of the target smelting furnace as input and the operating data prediction value as output; training a gradient boosting decision tree model based on the historical operating data, the historical operating data change value, the historical equipment failure frequency and the annotation information to obtain the failure probability prediction model with the operating data change value of the target smelting furnace, the operating data prediction value and the equipment failure frequency as input and the failure prediction probability as output; constructing the risk assessment model based on the operating data prediction model and the failure probability prediction model.

[0008] In one possible implementation, the risk assessment model includes: a model input module, an operation data prediction model, a fault probability prediction model and a model output module; the operation data prediction model is a deep learning neural network model, including a first convolutional layer, a random inactivation layer, a second convolutional layer, a pooling layer and a fully connected layer connected in sequence; the input of the first convolutional layer is connected to the model input module; the fault probability prediction model is a gradient boosting decision tree model, whose input is connected to the model input module and the output of the fully connected layer of the operation data prediction model; the model output module is connected to the output of the gradient boosting decision tree model.

[0009] In a possible implementation, the convolution kernel of the first convolution layer is 3×3, the step size is 3, the padding is 2, and the activation function is a hyperbolic tangent function; the convolution kernel of the second convolution layer is 3×3, the step size is 7, the padding is 2, and the activation function is a hyperbolic tangent function.

[0010] In a possible implementation, the current operating data and the current equipment failure frequency of the target melting furnace are acquired and input into the risk assessment model, and the risk assessment result of the target melting furnace is obtained based on two predictions of the operating data prediction model and the failure probability prediction model, including: calculating a target operating data change value based on the current operating data of the target melting furnace; inputting the current operating data and the target operating data change value into the operating data prediction model to obtain a target operating data prediction value of the target melting furnace; and inputting the target operating data prediction value, the target operating data change value and the current equipment failure frequency into the failure probability prediction model to obtain a target failure prediction probability of the target melting furnace.

[0011] In a possible implementation, the risk assessment result also includes a target risk failure level, and the current operating data and the current equipment failure frequency of the target melting furnace are obtained and input into the risk assessment model, and the risk assessment result of the target melting furnace is obtained based on two predictions of the operating data prediction model and the failure probability prediction model. The method also includes: obtaining the current operating data and the current equipment failure frequency of the target melting furnace and inputting into the risk assessment model, and outputting the target failure prediction probability of the target melting furnace based on the two predictions of the operating data prediction model and the failure probability prediction model; determining the target risk failure level of the target melting furnace based on the target failure prediction probability and a preset failure level range, and the target risk failure level is any one of normal, concern and high risk.

[0012] In a possible implementation, after determining the target risk failure level of the target smelting furnace based on the target failure prediction probability and the preset failure level range, the method further includes: determining a target intervention strategy based on the target risk failure level and a preset mapping relationship, and outputting a prompt signal based on the target risk failure level, the preset mapping relationship representing the mapping relationship between the risk failure level and the intervention strategy.

[0013] In a second aspect, an embodiment of the present application provides a smelting furnace production risk assessment device, the device comprising: a data acquisition module, used to acquire historical operating data, historical accident records and historical equipment failure frequencies of a target smelting furnace, the historical operating data comprising one or more of temperature, pressure and vibration; a sample construction module, used to annotate the historical operating data based on the historical accident records to obtain annotation information, and to construct a training sample data set based on the historical operating data, the historical equipment failure frequencies and the annotation information; a model training module, used to construct and train a risk assessment model based on the training sample data set, with the operating data and equipment failure frequencies of the target smelting furnace as input and the failure prediction probability as output, the risk assessment model comprising an operating data prediction model and a failure probability prediction model; a risk assessment module, used to acquire the current operating data and the current equipment failure frequency of the target smelting furnace, and input them into the risk assessment model, and obtain a risk assessment result of the target smelting furnace based on two predictions of the operating data prediction model and the failure probability prediction model, the risk assessment result comprising a target failure prediction probability.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the method described in the first aspect or any one of the implementation methods thereof.

[0017] Compared with the prior art, the embodiments of the present application have the following beneficial effects: first, the historical operating data, historical accident records, and historical equipment failure frequencies of the target smelting furnace are obtained, and the historical operating data are annotated based on the historical accident records to obtain annotated information. Compared with data annotation based on experience, the accuracy of the annotated information is improved; in addition, the historical operating data, historical equipment failure frequencies, and annotated information are used as training sample data sets, and based on the training sample data sets, a risk assessment model is constructed and trained with the operating data and equipment failure frequencies of the target smelting furnace as input and the failure prediction probability as output. Compared with model training using only the historical operating data and annotated information of the smelting furnace, the historical equipment failure frequencies are added, making the training sample data set richer. , which improves the accuracy of the trained risk assessment model; furthermore, the risk assessment model includes an operation data prediction model and a fault probability prediction model. After the current operation data and the current equipment failure frequency of the target smelting furnace are input into the trained risk assessment model, the target smelting furnace is predicted twice by the operation data prediction model and the fault probability prediction model respectively, and the target failure prediction probability in the risk assessment result of the target smelting furnace is obtained. Compared with relying on artificial experience and rules to conduct smelting furnace production risk assessment, or directly obtaining the failure prediction probability after only one prediction of the current operation data of the smelting furnace, two predictions based on the current operation data of the target smelting furnace and the current equipment failure frequency can make the target failure prediction probability more accurate, thereby improving the accuracy of the smelting furnace production risk assessment.

[0018] It can be understood that the smelting furnace production risk assessment device, electronic device, computer-readable storage medium and computer program product provided in the embodiments of the present application have the same beneficial effects as the above-mentioned smelting furnace production risk assessment method, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic diagram of a process flow of a smelting furnace production risk assessment method provided in one embodiment of the present application;

[0021] Figure 2 A schematic diagram of the structure of a risk assessment model provided in one embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of another risk assessment model provided in one embodiment of the present application;

[0023] Figure 4 A structural block diagram of a smelting furnace production risk assessment device provided in one embodiment of the present application;

[0024] Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0025] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0026] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0027] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0028] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0029] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0030] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0031] In industrial production, the safety and stability of the melting furnace, as a key production equipment, are directly related to production efficiency and product quality. However, since the melting process involves many complex factors such as high temperature, high pressure, and chemical reactions, risk control in the production process becomes a challenge. In addition, the melting furnace is a common industrial equipment, widely used in metallurgy, chemical industry, construction and other fields. Due to improper operation or equipment failure, safety accidents such as explosions and fires may occur during the operation of the melting furnace. Therefore, how to effectively assess the risks of melting furnace operation and promptly discover and deal with potential safety hazards has become an important issue.

[0032] Traditional smelting furnace risk assessment methods mainly rely on manual experience and rules, which are subjective and limited. At the same time, due to the complexity and variability of smelting furnace operation data, manual processing and analysis are inefficient and difficult to discover hidden safety hazards. Currently, many companies face the following problems in smelting furnace production risk management:

[0033] 1. Lack of systematic risk assessment tools and reliance on traditional manual experience judgments, resulting in low accuracy of risk assessment;

[0034] 2. The phenomenon of information islands is serious, and risk data between different departments is difficult to share and integrate;

[0035] 3. The emergency response mechanism is not perfect and lacks the ability to respond quickly to emergencies.

[0036] In order to solve the above technical problems, the present application provides a smelting furnace production risk assessment method, which obtains historical operating data, historical accident records and historical equipment failure frequencies of a target smelting furnace, wherein the historical operating data includes one or more of temperature, pressure and vibration; the historical operating data is labeled based on the historical accident records to obtain labeled information, and a training sample data set is constructed based on the historical operating data, the historical equipment failure frequency and the labeled information; based on the training sample data set, a risk assessment model is constructed and trained with the operating data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, the risk assessment model including an operating data prediction model and a failure probability prediction model; the current operating data and the current equipment failure frequency of the target smelting furnace are obtained and input into the risk assessment model, and the risk assessment result of the target smelting furnace is obtained based on two predictions of the operating data prediction model and the failure probability prediction model, and the risk assessment result includes the target failure prediction probability, thereby improving the accuracy of the smelting furnace production risk assessment.

[0037] For ease of understanding, the technical solution of the present application will be described in detail below with reference to the accompanying drawings.

[0038] Figure 1 This is a flow chart of a smelting furnace production risk assessment method provided in one embodiment of the present application. For ease of description, only the part related to this embodiment is shown. The method provided in this embodiment includes the following steps:

[0039] S110, acquiring historical operation data, historical accident records, and historical equipment failure frequencies of a target smelting furnace, wherein the historical operation data includes one or more of temperature, pressure, and vibration.

[0040] Specifically, an industrial melting furnace is a general term for special thermal equipment used to change the form or physical properties of materials in industrial activities. The target melting furnace is any industrial melting furnace; the historical operating data of the target melting furnace is the operating data of the target melting furnace at each historical moment, including but not limited to temperature, pressure and vibration; the historical accident records include accident cases that occurred in the target melting furnace in the past and their cause analysis, and the accident cases or cause analysis record the operating data of the target melting furnace when the accident occurred; the historical equipment failure frequency refers to the number of failures of the target melting furnace within a historical unit time.

[0041] In the specific implementation, various sensors (such as temperature sensors, pressure sensors, vibration sensors, etc.) installed on the target melting furnace and its surrounding environment are used to collect operating data such as temperature, pressure, and vibration in real time. Past accident cases of the target melting furnace and their cause analysis are collected to obtain historical accident records. The historical equipment failure frequency of the target melting furnace is directly obtained from the inspection management platform.

[0042] As an example, the equipment failure frequency is usually calculated by counting the number of equipment failures within a certain period of time. Assuming that within a certain period of time, the equipment fails N times, and the total length of the period is T, the equipment failure frequency F can be expressed as:

[0043] F=N / T,

[0044] Where F is the equipment failure frequency, N is the number of failures, and T is the total duration of the time period.

[0045] S120, annotating historical operation data based on historical accident records to obtain annotation information, and constructing a training sample data set based on the historical operation data, historical equipment failure frequencies and the annotation information.

[0046] Specifically, the historical accident records include accident cases and cause analyses of the target smelting furnace in the past. The accident cases or cause analyses record the operating data of the target smelting furnace when the accident occurred. Based on the historical accident records, the historical operating data corresponding to the target smelting furnace when the accident occurred is marked as a fault, and the remaining historical operating data is marked as normal.

[0047] In the specific implementation, the historical operation data corresponding to the failure of the target smelting furnace recorded in the historical accident record is marked as 1, and the remaining historical operation data is marked as 0 to obtain the annotation information, and the historical operation data, historical equipment failure frequency and annotation information are constructed to obtain a training sample data set.

[0048] As an example, historical operation data, historical equipment failure frequencies, and annotation information are cleaned to remove incomplete, erroneous, or abnormal data before constructing a training sample data set.

[0049] S130, based on the training sample data set, construct and train a risk assessment model with the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, wherein the risk assessment model includes an operation data prediction model and a failure probability prediction model.

[0050] In the specific implementation, the model structure of the risk assessment model is first constructed, and the risk assessment model is constructed as a model with a secondary prediction function including a data prediction model and a fault probability prediction model. Then, the training sample data set is divided into a training set, a validation set and a test set based on a preset ratio. The training set is used to train the risk assessment model through cross-validation and other methods, the validation set is used to select and adjust the parameters and structure of the risk assessment model, and the test set is used to evaluate the performance and accuracy of the risk assessment model, so that the risk assessment model after training has appropriate model parameters to achieve optimal performance.

[0051] S140, obtaining the current operating data and the current equipment failure frequency of the target smelting furnace, and inputting them into the risk assessment model, and obtaining the risk assessment result of the target smelting furnace based on the two predictions of the operating data prediction model and the failure probability prediction model, wherein the risk assessment result includes the target failure prediction probability.

[0052] In the specific implementation, the current operating data such as temperature, pressure and vibration of the target melting furnace are collected in real time by temperature sensors, pressure sensors and vibration sensors installed on the target melting furnace and the surrounding environment, and the current equipment failure frequency of the target melting furnace is obtained from the inspection management platform. The acquired current operating data and current equipment failure frequency are input into the trained risk assessment model. First, based on the current operating data, a first prediction is made through the operating data prediction model to obtain a first prediction result. Then, based on the current operating data, the current equipment failure frequency and the first prediction result, a second prediction is made through the fault probability prediction model to obtain the target failure prediction probability in the risk assessment result of the target melting furnace.

[0053] In a possible implementation, the risk assessment result also includes a target risk failure level. After the target failure prediction probability of the target smelting furnace is obtained based on the risk assessment model, the target risk failure level of the target smelting furnace is determined based on the target failure prediction probability and a preset failure level range. The target risk failure level is any one of normal, concern, and high risk.

[0054] As an example, the preset fault level range is set to a normal risk fault level when the fault prediction probability is less than 0.5, a concern risk fault level when the fault prediction probability is 0.5-0.7, and a high risk fault level when the fault prediction probability is greater than 0.7.

[0055] Exemplarily, the target failure prediction probability of the target smelting furnace obtained based on the risk assessment model is 0.4, and the target risk failure level of the target smelting furnace is normal at this time.

[0056] Furthermore, after determining the target risk failure level of the target smelting furnace, the target intervention strategy is determined based on the target risk failure level and a preset mapping relationship, and based on the target risk failure level, a prompt signal is output, and the preset mapping relationship represents the mapping relationship between the risk failure level and the intervention strategy.

[0057] In a specific implementation, different signal prompt strategies are set in advance for different risk fault levels, as well as mapping relationships between different risk fault levels and intervention strategies.

[0058] As an example, a risk warning may be provided to the target smelting furnace by means of a sound prompt or an indicator light prompt to notify relevant personnel. For example, different prompt sounds may be set for different risk fault levels, or different indicator light colors may be set for different risk fault levels, or different indicator light flashing frequencies may be set for different risk fault levels.

[0059] As an example, the preset mapping relationship that characterizes the mapping relationship between the risk failure level and the intervention strategy can be set as follows: when the risk failure level is a concern, the corresponding intervention strategy is to adjust the operating parameters, etc.; when the risk failure level is high risk, the corresponding intervention strategy is set to emergency shutdown or start-up of backup equipment, etc.; when the risk failure level is normal, the corresponding intervention strategy is not set or is set to continuous observation, etc.

[0060] In addition, as production conditions change, the training sample data set needs to be continuously updated, and the risk assessment model needs to be retrained based on the updated training sample data set to maintain the accuracy of the risk assessment model.

[0061] Furthermore, the preset fault level range and the preset mapping relationship characterizing the mapping relationship between the risk fault level and the intervention strategy are continuously adjusted according to the actual situation of the target smelting, so as to effectively identify and warn of possible risks in the production process of the target smelting furnace, thereby improving production safety and efficiency.

[0062] In addition, the risk assessment results of the target smelting furnace will be shared with various departments to effectively avoid the phenomenon of information islands, improve the emergency response mechanism, and enhance the company's ability to respond quickly to emergencies.

[0063] The technical solution provided by the present application obtains historical operating data, historical accident records, and historical equipment failure frequencies of a target smelting furnace, and annotates the historical operating data based on the historical accident records to obtain annotated information. Compared with data annotation based on experience, the accuracy of the annotated information is improved; the historical operating data, historical equipment failure frequencies, and annotated information are used as training sample data sets, and based on the training sample data sets, a risk assessment model is constructed and trained with the operating data and equipment failure frequencies of the target smelting furnace as input and the failure prediction probability as output. Compared with model training using only the historical operating data and annotated information of the smelting furnace, the historical equipment failure frequencies are added, making the training sample data set richer and improving the training style. The accuracy of the risk assessment model; the risk assessment model includes an operation data prediction model and a fault probability prediction model. After the current operation data of the target smelting furnace and the current equipment failure frequency are input into the trained risk assessment model, the target failure prediction probability in the risk assessment result of the target smelting furnace is obtained by two predictions by the operation data prediction model and the failure probability prediction model respectively. Compared with relying on artificial experience and rules to conduct smelting furnace production risk assessment, or directly obtaining the failure prediction probability after only one prediction of the current operation data of the smelting furnace, two predictions based on the current operation data of the target smelting furnace and the current equipment failure frequency can make the obtained target failure prediction probability more accurate, thereby improving the accuracy of the smelting furnace production risk assessment.

[0064] On the basis of the above embodiments, this embodiment further illustrates and optimizes the technical solution. Specifically, in this embodiment, based on the training sample data set, a risk assessment model is constructed and trained with the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, including:

[0065] Based on the historical operation data, the historical operation data change value is calculated, and the historical operation data change value includes the historical temperature change rate and the historical pressure fluctuation range;

[0066] Based on the historical operation data and the change value of the historical operation data, a deep learning neural network model is trained to obtain an operation data prediction model with the operation data of the target smelting furnace as input and the operation data prediction value as output;

[0067] Based on the historical operation data, the historical operation data change value, the historical equipment failure frequency and the annotation information, a gradient boosting decision tree model is trained to obtain a failure probability prediction model with the operation data change value, the operation data prediction value and the equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output;

[0068] A risk assessment model is constructed based on the operation data prediction model and the failure probability prediction model.

[0069] Specifically, the temperature change rate is the rate of change of temperature with respect to a certain factor (such as time) when there is a corresponding relationship between temperature and this factor (such as time) in a specific environment, that is, the derivative of temperature with respect to time. It is used to describe how fast the temperature changes over time, usually in degrees Celsius per second (℃ / s) or degrees Celsius per minute (℃ / min); the pressure fluctuation amplitude is used to describe the degree of pressure fluctuation within a certain period of time.

[0070] As an example, the calculation formula for the temperature change rate can be expressed as:

[0071] dT / dt=Q / mc

[0072] Where dT / dt is the rate of temperature change in degrees Celsius per second (℃ / s); Q is the heat flow in the system in joules (J); m is the mass of the system in kilograms (kg); and c is the specific heat capacity of the substance in joules per kilogram Celsius (J / (kg·℃)).

[0073] As an example, the pressure fluctuation amplitude can be calculated by the formula: pressure fluctuation amplitude = (instantaneous pressure value - pressure average value) / pressure average value × 100%, or directly expressed by the difference between the maximum value and the minimum value of the pressure fluctuation.

[0074] In the specific implementation, the structure of the risk assessment model is first constructed, such as Figure 2 As shown, it includes a model input module, an operation data prediction model, a fault probability prediction model and a model output module connected in sequence, wherein the operation data prediction model adopts a deep learning neural network model, and the fault probability prediction model adopts a gradient boosting decision tree model; based on the temperature and pressure data in the historical operation data and the corresponding time, the historical temperature change rate and the historical pressure fluctuation amplitude are calculated, and the historical operation data, the historical temperature change rate and the historical pressure fluctuation amplitude are input into the operation data prediction model through the model input module, and the historical temperature change rate, the historical pressure fluctuation amplitude and the historical equipment failure frequency and other data are input into the failure probability prediction model; based on the historical operation data and the historical operation data change values ​​such as the historical temperature change rate and the historical pressure fluctuation amplitude, the deep learning neural network model is trained to obtain the operation data prediction model to realize the prediction of future operation data; then based on the historical operation data, the historical operation data change value, w* the historical equipment failure frequency and the annotation information, the gradient boosting decision tree model is trained to obtain the fault probability prediction model to realize the prediction of the fault probability, and the final prediction result is output from the model output module.

[0075] As an example, the value of w is 2.

[0076] As an example, the cross entropy function is used as the loss function during training. The formula of the cross entropy loss function is:

[0077]

[0078] Where y is the true label (0 or 1), is the predicted probability.

[0079] As an example, Figure 3 As shown in the figure, the risk assessment model includes a model input module, an operation data prediction model, a fault probability prediction model and a model output module; the operation data prediction model is a deep learning neural network model, including a first convolution layer, a random dropout layer, a second convolution layer, a pooling layer and a fully connected layer connected in sequence; the input of the first convolution layer is connected to the model input module; the fault probability prediction model is a gradient boosting decision tree model, whose input is connected to the model input module and the output of the fully connected layer of the operation data prediction model; the model output module is connected to the output of the gradient boosting decision tree model. Among them, the convolution kernel of the first convolution layer is 3×3, the step size is 3, the padding is 2, and the activation function is the hyperbolic tangent function; the convolution kernel of the second convolution layer is 3×3, the step size is 7, the padding is 2, and the activation function is the hyperbolic tangent function. The hyperbolic tangent (Tanh) function is selected because it is similar to the sigmoid function, but the output range is (-1, 1), which can provide a more balanced activation in some cases.

[0080] Exemplarily, the hyperbolic tangent function is mathematically defined as the ratio of the hyperbolic sine function (sinh(x)) to the hyperbolic cosine function (cosh(x)), that is:

[0081] tanh(x)=sinh(x) / cosh(x),

[0082] Furthermore, the hyperbolic sine and cosine functions can be defined using exponential functions:

[0083] tanh(x)=sinh(x) / cosh(x)=(e x -e -x ) / (e x +e -x ),

[0084] Where e is the base of the natural logarithm and x is the input value.

[0085] As an example, Gradient Boosting Decision Trees (GBDT) is a powerful machine learning algorithm that combines the simplicity of decision trees with the power of the gradient boosting framework for regression and classification tasks. GBDT gradually adds decision trees in an iterative manner, and each tree is learned based on the residual of the previous tree, with the goal of minimizing the overall prediction error. Specifically, the residual calculation formula is: residual = actual observation value - model prediction value; expressed in mathematical symbols is:

[0086] e=y-y',

[0087] Among them, e represents the residual, y represents the actual observation value, and y' represents the model prediction value.

[0088] Specifically, the basic principles of GBDT include:

[0089] Gradient boosting: GBDT uses the idea of ​​gradient descent to minimize the loss function. In each iteration, the algorithm tries to fit the residual between the current total predicted value and the true target value (i.e., the gradient of the loss function) to gradually improve the prediction of the entire model;

[0090] Decision tree: As a base learner, GBDT uses decision trees. Each tree learns the remaining errors in the previous iteration, so the order of the trees is very important, and the later trees will correct the errors of the previous trees.

[0091] Furthermore, the workflow of GBDT includes the following steps:

[0092] Step 1, initialization: First, use a constant value or a simple model (such as an average value) as the initial prediction value, for example, use an average risk failure prediction value as the initial prediction value.

[0093] Preferably, an average risk assessment value of 75 points is used as the initial prediction value.

[0094] Step 2, iteration: For each iteration, GBDT trains a decision tree. The goal of the tree is to minimize the overall prediction error of the current stage. Each leaf node of the tree corresponds to a prediction value.

[0095] Step 3, residual calculation: Calculate the residual of all samples in the current stage, that is, the difference between the true value and the predicted value of all current trees.

[0096] Step 4: Train the next tree: Use the residual as the new target variable to train the next decision tree.

[0097] Step 5: Merge predictions: Add the prediction value of the newly generated tree to the prediction values ​​of all previous trees to obtain the updated overall prediction value.

[0098] Step 6: Repeat steps 2 to 5 until the stopping condition is met (such as reaching a preset number of iterations or the loss function converges, etc.).

[0099] Exemplarily, the workflow of GBDT includes:

[0100] 1. Initialization: Set an initial prediction value, such as the mean of the target values ​​of all samples, denoted as F0(x)=c, and the residual is r0=y-F0(x).

[0101] 2. Iterative training: For m = 1, 2, ..., M (M is the number of decision trees), perform the following steps:

[0102] Calculate the current residual r mi =y i -f m -1(x i ), i = 1, 2, ..., N (N is the number of samples);

[0103] Using {(x i ,r mi )}i=1,2,...,N train a decision tree (regression tree) Tm and get T(x;Θm);

[0104] Update f m (x) = f m -1(x)+T(x;Θm).

[0105] 3. Get the final model: After completing the above iterations, we get the boosted tree f M (x) = ∑i = 1 MT (x;Θm).

[0106] The technical solution provided in this embodiment trains a deep learning neural network model based on historical operating data and historical operating data change values ​​to obtain an operating data prediction model, trains a gradient boosting decision tree model based on historical operating data, historical operating data change values, historical equipment failure frequencies and annotation information to obtain a failure probability prediction model, and then constructs a risk assessment model based on the operating data prediction model and the failure probability prediction model, thereby improving the accuracy of the risk assessment model.

[0107] On the basis of the above embodiment, this embodiment further illustrates and optimizes the technical solution. Specifically, in this embodiment, the current operation data and the current equipment failure frequency of the target smelting furnace are obtained and input into the risk assessment model. Based on the two predictions of the operation data prediction model and the failure probability prediction model, the risk assessment result of the target smelting furnace is obtained, including:

[0108] Calculating a target operating data change value based on the current operating data of the target smelting furnace;

[0109] Inputting the current operating data and the target operating data change value into the operating data prediction model to obtain the target operating data prediction value of the target smelting furnace;

[0110] The target operation data prediction value, the target operation data change value and the current equipment failure frequency are input into the failure probability prediction model to obtain the target failure prediction probability of the target smelting furnace.

[0111] Specifically, the current operating data of the target melting furnace includes the temperature, pressure, vibration and other data of the target melting furnace at the current moment; the target operating data change value includes the target temperature change rate and target pressure fluctuation amplitude calculated based on the temperature and pressure of the target melting furnace at the current moment; the target operating data prediction value is the temperature prediction value, pressure prediction value and vibration prediction value corresponding to the target melting furnace at a preset time in the future, wherein the preset time in the future can be tomorrow or three days later, and this application does not limit this; the target fault prediction probability is the probability value of a failure of the target melting furnace.

[0112] In the specific implementation, based on Figure 2 or Figure 3 The risk assessment model shown in FIG. 1 inputs the current operating data of the target smelting furnace, the target operating data change value and the current equipment failure frequency of the target smelting furnace through the model input module; the current operating data and the target operating data change value of the target smelting furnace are input into the operating data prediction model; after the prediction of the operating data prediction model, the target operating data prediction value of the target smelting furnace is output; the target operating data prediction value is input into the failure probability prediction model; at the same time, the target operating data change value and the current equipment failure frequency are also input into the failure probability prediction model; the failure probability prediction model calculates the target failure prediction probability, and the model output module outputs the target failure prediction probability.

[0113] The technical solution provided in this embodiment uses an operation data prediction model to perform a first prediction based on the current operation data and target operation data change value of the target smelting furnace to obtain the target operation data prediction value of the target smelting furnace; then, based on the target operation data prediction value, target operation data change value and current equipment failure frequency of the target smelting furnace, a fault probability prediction model is used to perform a second prediction on the basis of the first prediction to obtain the target failure prediction probability of the target smelting furnace; compared with directly obtaining the failure prediction probability by performing a single prediction based on the current operation data of the target smelting furnace, the prediction result is more accurate and the accuracy of risk assessment is improved.

[0114] In summary, the technical solution provided in this application improves the accuracy and efficiency of the safety risk assessment of the smelting furnace; can process a large amount of smelting furnace operation data and discover hidden safety hazards; can achieve real-time monitoring and prediction, and issue alarms in time to avoid accidents; provides scientific technical support for the safety management of smelting furnaces and reduces management costs; optimizes production plans and resource allocation through data analysis, improves production efficiency, strengthens cross-departmental communication and collaboration, and improves overall operational efficiency; establishes a more complete emergency management system, effectively reduces the probability of accidents, protects the lives of employees and the property of the enterprise, and helps enterprises meet regulatory requirements and enhance their corporate image and market competitiveness.

[0115] Figure 4 This is a structural block diagram of a smelting furnace production risk assessment device provided in one embodiment of the present application. For the sake of convenience, only the parts related to the embodiment of the present application are shown. Figure 4 The smelting furnace production risk assessment device 400 may include a data acquisition module 401 , a sample construction module 402 , a model training module 403 and a risk assessment module 404 .

[0116] The data acquisition module 401 is used to acquire the historical operation data, historical accident records and historical equipment failure frequency of the target smelting furnace, and the historical operation data includes one or more of temperature, pressure and vibration.

[0117] The sample construction module 402 is used to annotate the historical operation data based on the historical accident records to obtain the annotation information, and to construct a training sample data set based on the historical operation data, the historical equipment failure frequency and the annotation information.

[0118] The model training module 403 is used to construct and train a risk assessment model based on a training sample data set, which takes the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output. The risk assessment model includes an operation data prediction model and a failure probability prediction model.

[0119] The risk assessment module 404 is used to obtain the current operating data and current equipment failure frequency of the target smelting furnace, and input them into the risk assessment model. Based on the two predictions of the operating data prediction model and the failure probability prediction model, the risk assessment result of the target smelting furnace is obtained. The risk assessment result includes the target failure prediction probability.

[0120] A smelting furnace production risk assessment device provided in an embodiment of the present application has the same beneficial effects as the above-mentioned smelting furnace production risk assessment method.

[0121] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0122] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0123] Figure 5 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 5 As shown, the electronic device 5 of this embodiment includes: at least one processor 50 ( Figure 5 Only one is shown in the figure), a memory 51, and a computer program 52 stored in the memory 51 and executable on at least one processor 50, the processor 50 executes the computer program 52 to implement the above Figure 1 The steps in the method embodiment, or the implementation of the above Figure 4 Functions of each module / unit in the device embodiment.

[0124] The electronic device 5 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 5 may include but is not limited to a processor 50 and a memory 51. Those skilled in the art will appreciate that Figure 5 It is only an example of the electronic device 5 and does not constitute a limitation on the electronic device 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0125] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0126] In some embodiments, the memory 51 may be an internal storage unit of the electronic device 5, such as a hard disk or memory of the electronic device 5. In other embodiments, the memory 51 may also be an external storage device of the electronic device 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 5. Further, the memory 51 may also include both an internal storage unit of the electronic device 5 and an external storage device. The memory 51 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as program codes of a computer program. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0127] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0128] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include at least: any entity or device capable of carrying the computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0129] A computer-readable storage medium provided in an embodiment of the present application has the same beneficial effects as the above-mentioned smelting furnace production risk assessment method.

[0130] An embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0131] A computer program product provided in an embodiment of the present application has the same beneficial effects as the above-mentioned smelting furnace production risk assessment method.

[0132] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0133] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0134] In the embodiments provided in the present application, it should be understood that the disclosed devices / equipment and methods can be implemented in other ways. For example, the device / equipment embodiments described above are merely schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A smelting furnace production risk assessment method, characterized in that: The method comprises: Acquire historical operation data, historical accident records, and historical equipment failure frequencies of the target smelting furnace, wherein the historical operation data includes one or more of temperature, pressure, and vibration; Annotating the historical operation data based on the historical accident records to obtain annotation information, and constructing a training sample data set based on the historical operation data, the historical equipment failure frequency and the annotation information; Based on the training sample data set, construct and train a risk assessment model with the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, wherein the risk assessment model includes an operation data prediction model and a failure probability prediction model; The current operating data and the current equipment failure frequency of the target smelting furnace are obtained and input into the risk assessment model, and the risk assessment result of the target smelting furnace is obtained based on the two predictions of the operating data prediction model and the failure probability prediction model, wherein the risk assessment result includes the target failure prediction probability.

2. The method according to claim 1, characterized in that The method of constructing and training a risk assessment model based on the training sample data set, which takes the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, comprises: Based on the historical operation data, calculating a historical operation data change value, wherein the historical operation data change value includes a historical temperature change rate and a historical pressure fluctuation amplitude; Based on the historical operation data and the change value of the historical operation data, a deep learning neural network model is trained to obtain the operation data prediction model which takes the operation data of the target smelting furnace as input and outputs the operation data prediction value; Based on the historical operation data, the historical operation data change value, the historical equipment failure frequency and the annotation information, a gradient boosting decision tree model is trained to obtain the fault probability prediction model with the operation data change value of the target smelting furnace, the operation data prediction value and the equipment failure frequency as input and the fault prediction probability as output; The risk assessment model is constructed based on the operation data prediction model and the failure probability prediction model.

3. The method according to claim 2, characterized in that The risk assessment model includes: a model input module, an operation data prediction model, a failure probability prediction model and a model output module; The running data prediction model is a deep learning neural network model, which includes a first convolutional layer, a random dropout layer, a second convolutional layer, a pooling layer, and a fully connected layer connected in sequence; the input of the first convolutional layer is connected to the model input module; The fault probability prediction model is a gradient boosting decision tree model, whose input is connected to the model input module and the output of the fully connected layer of the running data prediction model; Model output module, connected to the output of the gradient boosted decision tree model.

4. The method according to claim 3, characterized in that The convolution kernel of the first convolution layer is 3×3, the step size is 3, the padding is 2, and the activation function is the hyperbolic tangent function; the convolution kernel of the second convolution layer is 3×3, the step size is 7, the padding is 2, and the activation function is the hyperbolic tangent function.

5. The method according to claim 1, characterized in that The current operation data and the current equipment failure frequency of the target smelting furnace are obtained and input into the risk assessment model, and the risk assessment result of the target smelting furnace is obtained based on the two predictions of the operation data prediction model and the failure probability prediction model, including: Calculating a target operating data change value based on the current operating data of the target smelting furnace; Inputting the current operating data and the target operating data change value into the operating data prediction model to obtain a target operating data prediction value of the target smelting furnace; The target operation data prediction value, the target operation data change value and the current equipment failure frequency are input into the failure probability prediction model to obtain the target failure prediction probability of the target smelting furnace.

6. The method according to claim 1, characterized in that The risk assessment result also includes a target risk failure level. The current operation data and the current equipment failure frequency of the target smelting furnace are obtained and input into the risk assessment model. Based on the two predictions of the operation data prediction model and the failure probability prediction model, the risk assessment result of the target smelting furnace is obtained, and the following is also included: Acquire current operation data and current equipment failure frequency of the target smelting furnace, and input them into the risk assessment model, and output a target failure prediction probability of the target smelting furnace based on two predictions of the operation data prediction model and the failure probability prediction model; Based on the target fault prediction probability and the preset fault level range, a target risk fault level of the target smelting furnace is determined, and the target risk fault level is any one of normal, concern, and high risk.

7. The method according to claim 6, characterized in that After determining the target risk failure level of the target smelting furnace based on the target failure prediction probability and the preset failure level range, the method further includes: A target intervention strategy is determined based on the target risk failure level and a preset mapping relationship, and a prompt signal is output based on the target risk failure level, wherein the preset mapping relationship represents a mapping relationship between the risk failure level and the intervention strategy.

8. A smelting furnace production risk assessment device, characterized in that: The device comprises: A data acquisition module, used to acquire historical operation data, historical accident records, and historical equipment failure frequencies of a target smelting furnace, wherein the historical operation data includes one or more of temperature, pressure, and vibration; A sample construction module, used to annotate the historical operation data based on the historical accident records to obtain annotation information, and to construct a training sample data set based on the historical operation data, the historical equipment failure frequency and the annotation information; A model training module, for constructing and training a risk assessment model based on the training sample data set, which takes the operation data and equipment failure frequency of the target smelting furnace as input and the failure prediction probability as output, wherein the risk assessment model includes an operation data prediction model and a failure probability prediction model; The risk assessment module is used to obtain the current operating data and current equipment failure frequency of the target smelting furnace and input them into the risk assessment model, and obtain the risk assessment result of the target smelting furnace based on the two predictions of the operating data prediction model and the failure probability prediction model, wherein the risk assessment result includes the target failure prediction probability.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images