Sintering furnace temperature control method and system based on BP neural network prediction model

Abstract

The invention provides a sintering furnace temperature control method and system based on a BP neural network prediction model, and the method comprises the steps: constructing an initial model of a BP neural network based on Adaboost, and training a prediction model through test data and the initial model; the test data comprises the sintering furnace temperature and power of the test sintering target; and on the basis of the prediction model, according to the sintering temperature curve of the target needing to be sintered, the heating power needed by the sintering furnace is predicted and adjusted in real time. According to the sintering furnace temperature control method and system based on the BP neural network prediction model, the Adaboost-based BP neural network is trained by using the test data to obtain the prediction model, and the heating power required by the sintering furnace is predicted according to the sintering temperature curve of the required sintering target and is adjusted in real time; the heating process of the sintering furnace is more timely and more linear, the anti-interference capability is higher, further, the product stability of sintered products is improved, and the defective rate of the products is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology based on neural networks, and in particular to a sintering furnace temperature control method and system based on a BP neural network prediction model. Background Art

[0002] Sintering is the process of heating metal powder or powder compacts to below their melting point, allowing the particles to bond and achieve the desired strength. A sintering furnace is the device that accomplishes this process. During the sintering of powder compacts, precise control of heating and cooling rates is required to achieve dewaxing, reduction, alloying, and structural transformation of the powder. Therefore, temperature control during the sintering process is crucial.

[0003] However, the heating process of traditional sintering furnaces has the characteristics of delay, nonlinearity and strong disturbance, which leads to unstable production and high defective rate of sintered products.

[0004] In view of this, there is an urgent need for a sintering furnace temperature control method and system based on a BP neural network prediction model to at least solve the above-mentioned shortcomings. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a sintering furnace temperature control method and system based on a BP neural network prediction model. The prediction model is obtained by training a BP neural network based on Adaboost using experimental data. The heating power required by the sintering furnace is predicted and adjusted in real time according to the sintering temperature curve of the required sintering target. The heating process of the sintering furnace is more timely and more linear, and the anti-interference ability is also stronger. Furthermore, the output stability of the sintered products is improved and the product defect rate is reduced.

[0006] The sintering furnace temperature control method based on the BP neural network prediction model provided in an embodiment of the present invention includes:

[0007] Step 1: Construct an initial model of the BP neural network based on Adaboost, and use the test data and the initial model to train the prediction model; the test data includes: the sintering furnace temperature and power of the test sintering target;

[0008] Step 2: Based on the prediction model and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

[0009] Preferably, the method for measuring the sintering furnace temperature includes:

[0010] K-type thermocouples were used to measure the temperature distribution of each point in the sintering furnace over time.

[0011] The sintering furnace temperature control method based on the BP neural network prediction model provided in an embodiment of the present invention further includes:

[0012] Based on the prediction model and the sintering temperature curve of the required sintering target, the heating power required for the sintering furnace is predicted and adjusted in real time to determine whether the sintering furnace temperature is evenly distributed;

[0013] If the sintering furnace temperature distribution in the sintering furnace is uneven, a pre-built sintering furnace physical model is obtained based on the energy conservation, mass conservation, momentum conservation equations and the sintering furnace structural dimension parameters;

[0014] Determine the simulation results based on the power of the experimental sintering target and the pre-built sintering furnace physical model;

[0015] Comparing and analyzing the simulation results with the sintering furnace temperature of the experimental sintering target and correcting the pre-built sintering furnace physical model to obtain the first sintering furnace physical model;

[0016] Based on the prediction model, the physical model of the first sintering furnace and the sintering temperature curve of the required sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

[0017] Preferably, comparing and analyzing the simulation results and the sintering furnace temperature of the experimental sintering target and correcting the pre-built sintering furnace physical model to obtain a first sintering furnace physical model includes:

[0018] The pre-built sintering furnace physical model is loaded into the virtual furnace body, and the differential temperature points in the virtual furnace body are determined based on the comparative analysis results between the simulation results and the sintering furnace temperature of the experimental sintering target;

[0019] Present known influencing factors of temperature differences at preset viewing locations;

[0020] Indexing the target knowledge structure according to known influencing factors; the target knowledge structure includes a first unknown influencing factor;

[0021] Extracting knowledge structure information according to the target knowledge structure, inputting the knowledge structure information into an error analysis model trained based on the structure storage knowledge of the target knowledge structure, and obtaining an error analysis result;

[0022] Describing a first unknown influencing factor in the virtual furnace body according to the semantics of the error analysis result, and indexing the engine driving data of the described first unknown influencing factor;

[0023] The pre-built sintering furnace physical model is modified based on the engine drive data.

[0024] Preferably, indexing the target knowledge structure according to known influencing factors includes:

[0025] Determining a second knowledge structure viewed by a viewer based on a first knowledge structure indexed in the big data by known influencing factors;

[0026] The virtual furnace body is positioned for viewing by a viewer according to the second unknown influencing factor in the second knowledge structure, and the first unknown influencing factor selected by the viewer is determined.

[0027] Preferably, based on the prediction model, the physical model of the first sintering furnace and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time, including:

[0028] Determine a target correspondence relationship based on the prediction model and the physical model of the first sintering furnace, the target correspondence relationship being: a correspondence relationship between the sintering rack temperature distribution and the predicted power;

[0029] Calculate the degree of deviation between each target temperature in the sintering rack temperature distribution and the required temperature at the current moment;

[0030] The predicted power corresponding to the sintering rack temperature distribution with the smallest deviation is used as the target power for real-time adjustment.

[0031] The sintering furnace temperature control method based on the BP neural network prediction model provided in an embodiment of the present invention further includes:

[0032] Calculating a target ratio according to the minimum deviation degree and the corresponding required temperature corresponding to each sintering moment; if the length of the sintering period during which the target ratio is greater than a preset ratio threshold is greater than a preset period length threshold, determining the interval distribution of the target sintering moment in the sintering time interval;

[0033] Determine the deviation correction strategy based on the interval distribution characteristics.

[0034] Preferably, the deviation correction strategy is determined based on the interval distribution characteristics, including:

[0035] If the interval distribution feature meets the first standard distribution feature, read the heating device working state library according to the interval distribution feature to determine the heating device working state feature;

[0036] According to the working state characteristics of the heating device and the abnormal working state characteristic library, the abnormal heating device is identified and abnormal maintenance is performed.

[0037] Preferably, determining the deviation correction strategy based on the interval distribution characteristics further includes:

[0038] If the interval distribution characteristics meet the second standard distribution characteristics, the layout design of the multiple sintering racks in the sintering furnace is determined;

[0039] The physical model of the first sintering furnace is updated according to the sintering rack arrangement position design to obtain the physical model of the second sintering furnace;

[0040] Simulating a first temperature field line at a designed position under different simulated powers according to a physical model of a second sintering furnace;

[0041] According to the sintering temperature curve of the desired sintering target, a second temperature field line passing through the designed position at different sintering times is determined;

[0042] Determine a third temperature field line in the first temperature field line that is most similar to the second temperature field line and associate them accordingly;

[0043] determining an evaluation value of the sintering rack arrangement position design according to a temperature field line difference between each associated second temperature field line and the third temperature field line;

[0044] The sintering rack arrangement position designed corresponding to the sintering rack arrangement position with the highest evaluation value is used as the reset sintering rack arrangement position.

[0045] The sintering furnace temperature control system based on the BP neural network prediction model provided in an embodiment of the present invention includes:

[0046] A training module is used to build an initial model of the BP neural network based on Adaboost and train the prediction model using the test data and the initial model; the test data includes: the sintering furnace temperature and power of the test sintering target;

[0047] The control module is used to predict the heating power required by the sintering furnace based on the prediction model and the sintering temperature curve of the required sintering target, and adjust it in real time.

[0048] The beneficial effects of the present invention are:

[0049] The present invention uses experimental data to train a BP neural network based on Adaboost to obtain a prediction model. The heating power required by the sintering furnace is predicted according to the sintering temperature curve of the required sintering target and is adjusted in real time. The heating process of the sintering furnace is more timely and more linear, and the anti-interference ability is also stronger. Furthermore, the output stability of the sintered products is improved and the product defect rate is reduced.

[0050] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0051] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0053] Figure 1 Schematic diagram of a sintering furnace temperature control method based on a BP neural network prediction model in an embodiment of the present invention;

[0054] Figure 2 Schematic diagram of a sintering furnace temperature control system based on a BP neural network prediction model in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0056] The embodiment of the present invention provides a sintering furnace temperature control method based on a BP neural network prediction model, such as Figure 1 Shown, including:

[0057] Step 1: Construct an initial model of the BP neural network based on Adaboost, and use the test data and the initial model to train the prediction model; the test data includes: the sintering furnace temperature and power of the test sintering target;

[0058] Among them, the measurement methods of sintering furnace temperature include:

[0059] K-type thermocouples were used to measure the temperature distribution of each point in the sintering furnace over time.

[0060] Step 2: Based on the prediction model and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

[0061] In this embodiment, the test data are historical measurement data of powder material (eg, polytetrafluoroethylene) sintering, including the temperature of each point in the space of the sintering furnace measured by a K-type thermocouple at different powers during the test.

[0062] In this embodiment, when the initial model is trained using experimental data, the sintering furnace temperature of the experimental sintering target is used as the input of the initial model, and the power of the experimental sintering target is used as the output of the initial model. When the initial model is trained to convergence, it is used as a prediction model.

[0063] In this embodiment, the desired sintering target is the sintering target that actually needs to be sintered. The sintering temperature curve of the desired sintering target is an ideal temperature change curve of the entire sintering stage, which is determined according to the sintering requirements of the desired sintering target.

[0064] In this embodiment, based on the prediction model and the sintering temperature curve of the required sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time, which means: according to the current sintering stage, the ideal sintering temperature in the sintering temperature curve is read, the ideal sintering temperature is input into the prediction model, and the power output by the prediction model is used as the heating power for real-time adjustment.

[0065] The working principle and beneficial effects of the above technical solution are:

[0066] The present invention uses experimental data to train a BP neural network based on Adaboost to obtain a prediction model. The heating power required by the sintering furnace is predicted according to the sintering temperature curve of the required sintering target and is adjusted in real time. The heating process of the sintering furnace is more timely and more linear, and the anti-interference ability is also stronger. Furthermore, the output stability of the sintered products is improved and the product defect rate is reduced.

[0067] The embodiment of the present invention provides a sintering furnace temperature control method based on a BP neural network prediction model, further comprising:

[0068] Based on the prediction model and the sintering temperature curve of the required sintering target, the heating power required for the sintering furnace is predicted and adjusted in real time to determine whether the sintering furnace temperature is evenly distributed;

[0069] Among them, the determination of whether the sintering furnace temperature is uniform is made by comparing the temperatures of multiple spatial points under the same power. If the difference between the temperatures corresponding to two spatial points is greater than the set value, for example: 15°C, it is determined that the sintering furnace temperature distribution is uneven;

[0070] If the sintering furnace temperature distribution in the sintering furnace is uneven, a pre-built sintering furnace physical model is obtained based on the energy conservation, mass conservation, momentum conservation equations and the sintering furnace structural dimension parameters;

[0071] Among them, the pre-built sintering furnace physical model is: a numerical simulation model of the sintering furnace, which simulates the temperature field in the furnace and predicts the temperature response under different powers;

[0072] Determine the simulation results based on the power of the experimental sintering target and the pre-built sintering furnace physical model;

[0073] The experimental sintering target is the sintering target in the experimental data; the simulation result is the temperature distribution of the power corresponding to the experimental sintering target output by the pre-built sintering furnace physical model;

[0074] Comparing and analyzing the simulation results with the sintering furnace temperature of the experimental sintering target and correcting the pre-built sintering furnace physical model to obtain the first sintering furnace physical model;

[0075] The first sintering furnace physical model is a physical model that can truly reflect the temperature distribution within the sintering furnace by adjusting the pre-built sintering furnace physical model based on the simulation results and the analysis results of the sintering furnace temperature of the experimental sintering target. During the adjustment, the engineering personnel's historical adjustment records are used to make corresponding adjustments. The adjustment records include adjustment strategies based on the comparison of different simulation results and actual results.

[0076] Based on the prediction model, the physical model of the first sintering furnace and the sintering temperature curve of the required sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

[0077] Among them, based on the prediction model, the physical model of the first sintering furnace and the sintering temperature curve of the required sintering target, predicting the heating power required for the sintering furnace and adjusting it in real time means: reading the ideal sintering temperature in the sintering temperature curve according to the current sintering stage, and at the same time, taking into account the influencing factor of uneven temperature distribution in the furnace, and finally using the prediction model to predict the power that can make the required sintering target reach the ideal sintering temperature of the current sintering stage.

[0078] The working principle and beneficial effects of the above technical solution are:

[0079] Due to the different types and distribution of heating devices, the temperature distribution in the sintering furnace may be uneven during temperature control, which will lead to poor sintering quality. Therefore, before predicting the required heating power of the sintering furnace based on the prediction model and the sintering temperature curve of the required sintering target and adjusting it in real time, it is necessary to determine whether the sintering furnace temperature is evenly distributed.

[0080] Specifically, based on the energy conservation, mass conservation, momentum conservation equations and the sintering furnace structural dimension parameters, a pre-built sintering furnace physical model was constructed in the ANSYS numerical simulation software for simulation research. Based on the experimental data and the simulation results, a comparative analysis was conducted and the pre-built sintering furnace physical model was modified to obtain a physical model that can truly reflect the temperature distribution in the sintering furnace (the first sintering furnace physical model). Finally, relevant information about temperature unevenness was extracted from the first sintering furnace physical model. When predicting the heating power, the temperature unevenness factor was taken into account. For example, if the temperature unevenness is not considered, the ideal temperature of the sintering stage is 360°C. The heating power corresponding to 360°C is determined according to the prediction model for real-time power adjustment. However, considering the uneven temperature distribution, the actual temperature at the sintering rack A is 355°C at the heating power corresponding to 360°C. In fact, the heating power needs to be increased so that the temperature at the actual position of the desired sintering target reaches the ideal temperature of the corresponding stage.

[0081] The present invention takes into account the situation of uneven temperature distribution in the sintering furnace, uses experimental data and a numerical model of the sintering furnace to perform simulation, compares the simulation results with the actual results, and obtains a physical model (the first sintering furnace physical model) that can truly reflect the temperature distribution in the sintering furnace. The prediction model, the first sintering furnace physical model and the sintering temperature curve of the required sintering target are then combined to jointly predict the heating power required for the sintering furnace and adjust it in real time, further improving the rationality and accuracy of the determination of the heating power.

[0082] In one embodiment, the simulation results are compared and analyzed with the sintering furnace temperature of the experimental sintering target, and the pre-built sintering furnace physical model is modified to obtain a first sintering furnace physical model, including:

[0083] The pre-built sintering furnace physical model is loaded into the virtual furnace body, and the differential temperature points in the virtual furnace body are determined based on the comparative analysis results between the simulation results and the sintering furnace temperature of the experimental sintering target;

[0084] The virtual furnace body is a digital twin model of the sintering furnace corresponding to the pre-built physical model of the sintering furnace. The differential temperature points are the twin points where the simulated temperature in the virtual furnace body is inconsistent with the actual test sintering furnace temperature.

[0085] Present known influencing factors of temperature differences at preset viewing locations;

[0086] The preset viewing position is, for example, the portion of the display screen near the differential temperature point that does not cover the virtual furnace body; the known influencing factors are the influencing factors of the known differential temperature points determined based on the construction principle of the pre-built sintering furnace physical model, which are input by the simulation error attribution personnel;

[0087] Indexing the target knowledge structure according to known influencing factors; the target knowledge structure includes a first unknown influencing factor;

[0088] The target knowledge structure is a knowledge graph structure, and the known influencing factors can all find corresponding items in the corresponding entities of the target knowledge structure; the first unknown influencing factor is the influencing factor corresponding to the entity that has a structural association with the target knowledge structure;

[0089] Extracting knowledge structure information according to the target knowledge structure, inputting the knowledge structure information into an error analysis model trained based on the structure storage knowledge of the target knowledge structure, and obtaining an error analysis result;

[0090] The knowledge structure information is the relevant information of the sintering furnace in reality corresponding to the target knowledge structure, such as the distribution of the sintering racks, the structural dimension parameters of the sintering furnace, etc. The structural storage knowledge of the target knowledge structure is the entity relationship corresponding to its graph entities; the error analysis result is the reason for the error in the simulation analyzed by the error analysis model based on the real information;

[0091] Describing a first unknown influencing factor in the virtual furnace body according to the semantics of the error analysis result, and indexing the engine driving data of the described first unknown influencing factor;

[0092] The first unknown influencing factor is described in the virtual furnace body to generate a corresponding virtual part in the virtual furnace body according to the described first unknown influencing factor; the engine driving data is reversely deduced according to the twin rules preset in the virtual furnace body and the virtual part generated by the description;

[0093] Modify the pre-built sintering furnace physical model based on engine drive data;

[0094] When the pre-built sintering furnace physical model is modified according to the engine driving data, the pre-built sintering furnace physical model is modified according to the correlation between the engine driving data and the pre-built sintering furnace physical model;

[0095] Among them, indexing the target knowledge structure according to known influencing factors includes:

[0096] Determining a second knowledge structure viewed by a viewer based on a first knowledge structure indexed in the big data by known influencing factors;

[0097] Among them, the first knowledge structure is a knowledge graph structure containing all graph entities corresponding to known influencing factors based on big data index;

[0098] The virtual furnace body is positioned for viewing by a viewer according to the second unknown influencing factor in the second knowledge structure, and the first unknown influencing factor selected by the viewer is determined.

[0099] Among them, positioning the virtual furnace body according to the second unknown influencing factor refers to automatically adjusting the viewing angle of the virtual furnace body and displaying the position of the virtual furnace body generated by the second unknown influencing factor to the viewer, so that the viewer can verify whether the influence of the second unknown influencing factor does exist. If it is determined that the influence of the second unknown influencing factor exists, the corresponding second knowledge structure will be used as the target knowledge structure, and the corresponding second unknown influencing factor will be used as the first unknown influencing factor in the target knowledge structure.

[0100] The working principle and beneficial effects of the above technical solution are:

[0101] The present invention performs digital twinning on the pre-built sintering furnace physical model, determines the differential temperature points in the virtual furnace body, and determines the known influencing factors. At the same time, it indexes the target knowledge structure containing the known influencing factors, extracts the actual knowledge structure information based on the target knowledge structure, and inputs the knowledge structure information into the error analysis model trained based on the structural storage knowledge of the target knowledge structure to obtain the error analysis results.

[0102] According to the semantics of the error analysis results, the first unknown influencing factor is automatically described in the virtual furnace body. Based on the preset twin rules and the corresponding engine driving data of the virtual part, the pre-built sintering furnace physical model is corrected according to the correspondence between the engine driving data and the numerical model. This assists viewers in correcting the pre-built sintering furnace physical model, making it more user-friendly.

[0103] In one embodiment, based on the prediction model, the physical model of the first sintering furnace, and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time, including:

[0104] Determine a target correspondence relationship based on the prediction model and the physical model of the first sintering furnace, the target correspondence relationship being: a correspondence relationship between the sintering rack temperature distribution and the predicted power;

[0105] The corresponding relationship between the sintering rack temperature distribution and the predicted power is: the corresponding relationship between different predicted powers and the temperature distribution at each sintering rack under the predicted power;

[0106] Calculate the degree of deviation between each target temperature in the sintering rack temperature distribution and the required temperature at the current moment;

[0107] The deviation degree is: the mean square error between each target temperature and the required temperature;

[0108] The predicted power corresponding to the sintering rack temperature distribution with the smallest deviation is used as the target power for real-time adjustment.

[0109] The working principle and beneficial effects of the above technical solution are:

[0110] The present invention determines a target correspondence, quantifies the degree of deviation between the sintering rack temperature distribution and the required temperature in the target correspondence, and uses the predicted power corresponding to the sintering rack temperature distribution with the smallest degree of deviation as the target power for real-time adjustment, thereby further improving the suitability of the real-time adjusted heating power setting.

[0111] In one embodiment, based on the prediction model, the physical model of the first sintering furnace, and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time, further comprising:

[0112] Calculating a target ratio according to the minimum deviation degree and the corresponding required temperature corresponding to each sintering moment; if the length of the sintering period during which the target ratio is greater than a preset ratio threshold is greater than a preset period length threshold, determining the interval distribution of the target sintering moment in the sintering time interval;

[0113] The target ratio is the result obtained by dividing the square root of the deviation by the required temperature; the preset ratio threshold and the preset time period threshold are both manually preset; the target sintering time is the sintering time when the target ratio is greater than the preset ratio threshold, and the sintering time period is the total time length of the target sintering time; the interval distribution is the interval correspondence between the target sintering time and the time interval corresponding to the total sintering process;

[0114] Determine the deviation correction strategy based on the interval distribution characteristics.

[0115] Among them, the interval distribution feature is: a characterization of the interval correspondence between the target sintering moment and the time interval corresponding to the total sintering process, for example: the image features extracted after marking the target sintering moment on the time axis corresponding to the time interval; the deviation correction strategy is: a plan to correct the temperature deviation.

[0116] The working principle and beneficial effects of the above technical solution are:

[0117] When the degree of temperature deviation is the same, the greater the temperature base in the sintering furnace (the required temperature at the time of sintering), the smaller the impact of the temperature deviation on the final sintering result. However, during the entire sintering process, the degree of temperature deviation and the required temperature at the time of sintering are different, and it is impossible to accurately quantify the impact of the temperature deviation on the sintering process.

[0118] Therefore, the present invention introduces a target ratio, which represents the ratio of the minimum temperature deviation degree corresponding to each sintering moment to the required temperature. The larger the target ratio is, the greater the impact of the temperature deviation degree on the sintering process.

[0119] Since temperature control based on the predicted power corresponding to the sintering rack temperature distribution with the minimum deviation at each sintering moment is already the optimal temperature control state of the current sintering furnace, if the statistical target ratio at this time is greater than the preset ratio threshold and the corresponding sintering period length still exceeds the period length threshold, it indicates that temperature deviation correction is required.

[0120] The present invention introduces a target ratio to quantify the impact of temperature deviation from the optimal temperature control state on the sintering process. Whether temperature deviation correction is needed is determined based on the duration of this impact throughout the entire sintering process. The deviation correction strategy is determined based on the time information of the deviation, further improving the subsequent temperature control accuracy.

[0121] In one embodiment, determining a deviation correction strategy based on interval distribution characteristics includes:

[0122] If the interval distribution feature meets the first standard distribution feature, read the heating device working state library according to the interval distribution feature to determine the heating device working state feature;

[0123] The first standard distribution feature is, for example, that the target sintering time points on the time axis appear in clusters or clusters. The heating device working state library stores the working states of the heating devices that intervene in heating at different time points. The working state of the heating device that intervenes in heating at the target sintering time is read according to the first standard distribution feature, and its working parameters are the heating device working state features.

[0124] According to the working state characteristics of the heating device and the abnormal working state characteristic library, the abnormal heating device is identified and abnormal maintenance is performed.

[0125] Among them, the abnormal working state feature library includes the working parameters and abnormal maintenance strategies of different heating devices during abnormal operation. The working state characteristics of the heating device are matched with the abnormal working state characteristics of the corresponding heating device in the library. If the match is consistent, maintenance is performed according to the corresponding abnormal maintenance strategy.

[0126] The working principle and beneficial effects of the above technical solution are:

[0127] The time information of the deviation reveals the possible cause of the deviation. For example, if the deviation occurs only within a period of time and a certain heating device is working at the same time during this period, then there is a high probability that the heating device has an abnormality, causing the heating deviation.

[0128] Therefore, when the target sintering times appear in clusters at the annotated points on the timeline, the operating parameters of the heating devices involved (heating device operating state characteristics) are determined based on the corresponding target sintering times and the heating device operating state library. By introducing an abnormal operating state feature library, the heating device operating state characteristics are matched with abnormal operating state features in the library to identify abnormal heating devices and perform abnormal maintenance, thus improving the targeted identification and handling of heating device anomalies.

[0129] In one embodiment, determining a deviation correction strategy based on interval distribution characteristics further includes:

[0130] If the interval distribution characteristics meet the second standard distribution characteristics, the layout design of the multiple sintering racks in the sintering furnace is determined;

[0131] The second standard distribution feature is that the target sintering time points on the time axis appear discretely on the entire time axis; the sintering rack arrangement position design is the design scheme of the arrangement positions of different sintering racks in the sintering furnace;

[0132] The physical model of the first sintering furnace is updated according to the sintering rack arrangement position design to obtain the physical model of the second sintering furnace;

[0133] When the physical model of the first sintering furnace is updated according to the sintering rack arrangement position design to obtain the physical model of the second sintering furnace, the thermodynamic relationship and fluid dynamics relationship in the numerical model are re-determined based on the sintering rack position corresponding to the sintering rack arrangement position design, and the model is updated, and then verified using experimental data;

[0134] Simulating a first temperature field line at a designed position under different simulated powers according to a physical model of a second sintering furnace;

[0135] The design position is the sintering rack position designed in the sintering rack arrangement position design; the first temperature field line is the temperature field line passed by the sintering position corresponding to the sintering rack arrangement position design under different powers;

[0136] According to the sintering temperature curve of the desired sintering target, a second temperature field line passing through the designed position at different sintering times is determined;

[0137] The second temperature field line is: the ideal temperature field line passed by the designed position at different sintering times;

[0138] Determine a third temperature field line in the first temperature field line that is most similar to the second temperature field line and associate them accordingly;

[0139] determining an evaluation value of the sintering rack arrangement position design according to a temperature field line difference between each associated second temperature field line and the third temperature field line;

[0140] The temperature field line difference is the degree of fit of the associated second and third temperature field lines; the evaluation value is the sum of the degrees of fit corresponding to each sintering moment;

[0141] The sintering rack arrangement position designed corresponding to the sintering rack arrangement position with the highest evaluation value is used as the reset sintering rack arrangement position.

[0142] The working principle and beneficial effects of the above technical solution are:

[0143] The time information of the deviation reveals the possible reasons for the deviation. For example, if the deviation occurs discretely during the entire sintering process, it should be attributed to internal factors, such as the unreasonable arrangement of the sintering rack, which leads to uneven heating in the sintering furnace and random and discrete heating deviations.

[0144] Therefore, multiple sintering rack layout designs other than the current one were obtained. Since changes to the sintering rack layout would cause changes in the fluid movement within the sintering furnace, the thermodynamic and fluid dynamics relationships in the numerical model were redefined based on the sintering rack positions corresponding to the sintering rack layout designs, and the model was updated and verified using experimental data.

[0145] The new physical model of the second sintering furnace simulates the first temperature field line at different simulated powers. The first temperature field line represents the temperature field line distribution for all heating scenarios of the second sintering furnace physical model. Furthermore, the ideal temperature field line (second temperature field line) passing through the design position at different sintering times is determined. The third temperature field line that is most similar to the second temperature field line in the first temperature field line is determined. This third temperature field line represents the temperature distribution that best meets the temperature control requirements for the design position. The sum of the fits of the second and third temperature field lines associated with each sintering time is used as the evaluation value. The sintering rack layout design corresponding to the highest evaluation value is determined, thereby improving the rationality of the sintering rack layout setting.

[0146] The embodiment of the present invention provides a sintering furnace temperature control system based on a BP neural network prediction model, such as Figure 2 Shown, including:

[0147] Training module 1 is used to build an initial model of the BP neural network based on Adaboost and train the prediction model using the test data and the initial model; the test data includes: the sintering furnace temperature and power of the test sintering target;

[0148] The control module 2 is used to predict the heating power required by the sintering furnace based on the prediction model and the sintering temperature curve of the required sintering target, and adjust it in real time.

[0149] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A sintering furnace temperature control method based on a BP neural network prediction model, characterized in that: include: Step 1: Construct an initial model of the BP neural network based on Adaboost, and use the test data and the initial model to train the prediction model; the test data includes: the sintering furnace temperature and power of the test sintering target; Step 2: Based on the prediction model and the sintering temperature curve of the desired sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

2. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 1, characterized in that: The measurement methods of sintering furnace temperature include: K-type thermocouples were used to measure the temperature distribution of each point in the sintering furnace over time.

3. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 1, characterized in that: Also includes: Based on the prediction model and the sintering temperature curve of the required sintering target, the heating power required for the sintering furnace is predicted and adjusted in real time to determine whether the sintering furnace temperature is evenly distributed; If the sintering furnace temperature distribution in the sintering furnace is uneven, a pre-built sintering furnace physical model is obtained based on the energy conservation, mass conservation, momentum conservation equations and the sintering furnace structural dimension parameters; Determine the simulation results based on the power of the experimental sintering target and the pre-built sintering furnace physical model; Comparing and analyzing the simulation results with the sintering furnace temperature of the experimental sintering target and correcting the pre-built sintering furnace physical model to obtain the first sintering furnace physical model; Based on the prediction model, the physical model of the first sintering furnace and the sintering temperature curve of the required sintering target, the heating power required by the sintering furnace is predicted and adjusted in real time.

4. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 3, characterized in that: Compare and analyze the simulation results with the sintering furnace temperature of the experimental sintering target and modify the pre-built sintering furnace physical model to obtain the first sintering furnace physical model, including: The pre-built sintering furnace physical model is loaded into the virtual furnace body, and the differential temperature points in the virtual furnace body are determined based on the comparative analysis results between the simulation results and the sintering furnace temperature of the experimental sintering target; Present known influencing factors of temperature differences at preset viewing locations; Indexing the target knowledge structure according to known influencing factors; the target knowledge structure includes a first unknown influencing factor; Extracting knowledge structure information according to the target knowledge structure, inputting the knowledge structure information into an error analysis model trained based on the structure storage knowledge of the target knowledge structure, and obtaining an error analysis result; Describing a first unknown influencing factor in the virtual furnace body according to the semantics of the error analysis result, and indexing the engine driving data of the described first unknown influencing factor; The pre-built sintering furnace physical model is modified based on the engine drive data.

5. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 4, characterized in that: Index the target knowledge structure based on known influencing factors, including: Determining a second knowledge structure viewed by a viewer based on a first knowledge structure indexed in the big data by known influencing factors; The virtual furnace body is positioned for viewing by a viewer according to the second unknown influencing factor in the second knowledge structure, and the first unknown influencing factor selected by the viewer is determined.

6. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 3, characterized in that: Based on the prediction model, the physical model of the first sintering furnace, and the sintering temperature curve of the desired sintering target, the required heating power of the sintering furnace is predicted and adjusted in real time, including: Determine a target correspondence relationship based on the prediction model and the physical model of the first sintering furnace, the target correspondence relationship being: a correspondence relationship between the sintering rack temperature distribution and the predicted power; Calculate the degree of deviation between each target temperature in the sintering rack temperature distribution and the required temperature at the current moment; The predicted power corresponding to the sintering rack temperature distribution with the smallest deviation is used as the target power for real-time adjustment.

7. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 6, characterized in that: Also includes: Calculating a target ratio according to the minimum deviation degree and the corresponding required temperature corresponding to each sintering moment; if the length of the sintering period during which the target ratio is greater than a preset ratio threshold is greater than a preset period length threshold, determining the interval distribution of the target sintering moment in the sintering time interval; Determine the deviation correction strategy based on the interval distribution characteristics.

8. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 7, characterized in that: Determine the deviation correction strategy based on the interval distribution characteristics, including: If the interval distribution feature meets the first standard distribution feature, read the heating device working state library according to the interval distribution feature to determine the heating device working state feature; According to the working state characteristics of the heating device and the abnormal working state characteristic library, the abnormal heating device is identified and abnormal maintenance is performed.

9. The sintering furnace temperature control method based on the BP neural network prediction model according to claim 7, characterized in that: Determine the deviation correction strategy based on the interval distribution characteristics, including: If the interval distribution characteristics meet the second standard distribution characteristics, the layout design of the multiple sintering racks in the sintering furnace is determined; The physical model of the first sintering furnace is updated according to the sintering rack arrangement position design to obtain the physical model of the second sintering furnace; Simulating a first temperature field line at a designed position under different simulated powers according to a physical model of a second sintering furnace; According to the sintering temperature curve of the desired sintering target, a second temperature field line passing through the designed position at different sintering times is determined; Determine a third temperature field line in the first temperature field line that is most similar to the second temperature field line and associate them accordingly; determining an evaluation value of the sintering rack arrangement position design according to a temperature field line difference between each associated second temperature field line and the third temperature field line; The sintering rack arrangement position designed corresponding to the sintering rack arrangement position with the highest evaluation value is used as the reset sintering rack arrangement position.

10. The sintering furnace temperature control system based on the BP neural network prediction model is characterized in that: include: A training module is used to build an initial model of the BP neural network based on Adaboost and train the prediction model using the test data and the initial model; the test data includes: the sintering furnace temperature and power of the test sintering target; The control module is used to predict the heating power required by the sintering furnace based on the prediction model and the sintering temperature curve of the required sintering target, and adjust it in real time.

Images