Heating furnace temperature monitoring system based on neural network

Abstract

The invention belongs to the technical field of heating furnace supervision, and particularly relates to a heating furnace temperature monitoring system based on a neural network, which comprises a comprehensive monitoring transmission module, a furnace temperature abnormity diagnosis module, a neural network prediction module, an intelligent alarm regulation and control module and a display alarm terminal, the furnace temperature abnormity diagnosis module is used for carrying out furnace temperature abnormity diagnosis analysis on the heating furnace, the operation of the heating furnace is suspended when a furnace temperature diagnosis unqualified signal is generated, corresponding improvement measures are taken for the heating furnace, and the furnace temperature in a period of time in the future is predicted and corresponding parameter adjustment is carried out when a furnace temperature diagnosis qualified signal is generated. The furnace temperature monitoring difficulty of the heating furnace is reduced by reducing manual intervention, the risk position in the heating furnace is accurately captured through analysis when furnace temperature diagnosis disqualification signals are not generated in the monitoring time period, operation of the heating furnace is stopped when the risk position is captured, the corresponding position is checked, the subsequent operation effect of the heating furnace is further guaranteed, and the safety of the heating furnace is improved. And the intelligent level and the automation level are high.

Description

Technical Field

[0001] The present invention relates to the technical field of heating furnace supervision, and specifically to a heating furnace temperature monitoring system based on a neural network. Background Art

[0002] A heating furnace is an important device in industrial production. Its main function is to transfer heat to the materials or workpieces in the furnace chamber to make them reach the predetermined heating temperature. It is widely used in multiple industrial fields such as metallurgy, machinery, chemical engineering, and building materials. During the operation of the heating furnace, the precise control of the furnace temperature has a direct impact on product quality and production efficiency; Currently, the furnace temperature monitoring mainly uses physical sensors, such as thermocouples and thermal resistors, to monitor the internal temperature of the heating furnace. However, in actual operation, it is difficult to achieve automatic diagnosis of furnace temperature anomalies and prediction and regulation of furnace temperature, and it is impossible to reasonably analyze and timely warn of the potential hazards in the regulation execution and supervision of the heating furnace, which is not conducive to ensuring the operation effect of the heating furnace and reducing the operation supervision difficulty, and the level of intelligence and automation is low; In view of the above technical defects, a solution is proposed. Summary of the Invention

[0003] The purpose of the present invention is to provide a heating furnace temperature monitoring system based on a neural network, which solves the problems that in the prior art, it is difficult to achieve automatic diagnosis of furnace temperature anomalies and prediction and regulation of furnace temperature, and it is impossible to reasonably analyze and timely warn of the potential hazards in the regulation execution and supervision of the heating furnace, which is not conducive to ensuring the operation effect of the heating furnace and reducing the operation supervision difficulty, and the level of intelligence and automation is low.

[0004] To achieve the above purpose, the present invention provides the following technical solutions: A heating furnace temperature monitoring system based on a neural network includes a comprehensive monitoring and transmission module, a furnace temperature anomaly diagnosis module, a neural network prediction module, an intelligent alarm and regulation module, and a display and alarm terminal; the comprehensive monitoring and transmission module collects temperature data at several positions inside the heating furnace, as well as auxiliary data including heating power, air flow velocity inside the furnace, and ambient temperature, thereby obtaining a heating furnace monitoring data set, and sending the heating furnace monitoring data set to the furnace temperature anomaly diagnosis module and the neural network prediction module; The furnace temperature anomaly diagnosis module conducts furnace temperature anomaly diagnosis and analysis on the heating furnace, thereby generating a furnace temperature diagnosis qualified signal or a furnace temperature diagnosis unqualified signal, and sending the furnace temperature diagnosis unqualified signal to the intelligent alarm and regulation module when it is generated; When generating a qualified signal for furnace temperature diagnosis, the neural network prediction module predicts the furnace temperature within a future period according to the monitoring data set of the heating furnace, and sends the furnace temperature prediction result to the intelligent alarm and regulation module; when receiving an unqualified signal for furnace temperature diagnosis, the intelligent alarm and regulation module generates corresponding warning information and sends it to the display and alarm terminal, and when receiving the furnace temperature prediction result, it intelligently regulates the operating state of the heating furnace, dynamically adjusts the heating power and air flow speed parameters of the heating furnace, realizes precise control of the furnace temperature, and sends the regulation information to the display and alarm terminal in real time.

[0005] Further, the specific analysis process of furnace temperature anomaly diagnosis and analysis includes: Obtain the temperatures at several positions in the heating furnace, establish a set of actual furnace temperatures for all positions, calculate the variance of all subsets in the set of actual furnace temperatures, and obtain the furnace temperature distribution anomaly coefficient accordingly. Compare the furnace temperature distribution anomaly coefficient with the preset furnace temperature distribution anomaly coefficient threshold. If the furnace temperature distribution anomaly coefficient exceeds the preset furnace temperature distribution anomaly coefficient threshold, generate an unqualified signal for furnace temperature diagnosis.

[0006] Further, if the furnace temperature distribution anomaly coefficient does not exceed the preset furnace temperature distribution anomaly coefficient threshold, compare the temperature at the corresponding position with the preset temperature range that it matches. If the temperature at the corresponding position is not within the preset temperature range, mark the corresponding position as an abnormal furnace temperature position; Obtain the ratio of the number of abnormal furnace temperature positions in the heating furnace and mark it as the furnace temperature anomaly detection value. Compare the furnace temperature anomaly detection value with the preset furnace temperature anomaly detection threshold. If the furnace temperature anomaly detection value exceeds the preset furnace temperature anomaly detection threshold, generate an unqualified signal for furnace temperature diagnosis; If the furnace temperature anomaly detection value does not exceed the preset furnace temperature anomaly detection threshold, calculate the average value of the temperatures at all positions to obtain the furnace temperature analysis value, and calculate the absolute value of the difference between the furnace temperature analysis value and the median of the preset temperature range to obtain the furnace temperature deviation performance value; Compare the furnace temperature deviation performance value with the preset furnace temperature deviation performance threshold. If the furnace temperature deviation performance value exceeds the preset furnace temperature deviation performance threshold, generate an unqualified signal for furnace temperature diagnosis; if the furnace temperature deviation performance value does not exceed the preset furnace temperature deviation performance threshold, generate a qualified signal for furnace temperature diagnosis.

[0007] Further, the furnace temperature anomaly diagnosis module is communicatively connected to the risk position capture and output module. The risk position capture and output module is used to set a monitoring period with a duration of L1. When the duration reaches L1, if no unqualified signal for furnace temperature diagnosis is generated during the monitoring period, mark the number of times the corresponding position in the heating furnace is marked as an abnormal furnace temperature position during the monitoring period as the furnace temperature position abnormal frequency; Numerically compare the abnormal frequency of the furnace temperature with the preset abnormal frequency threshold of the furnace temperature. If the abnormal frequency of the furnace temperature exceeds the preset abnormal frequency threshold of the furnace temperature, mark the corresponding position as a risk position; if there is a risk position within the monitoring period, send the risk position to the intelligent alarm and control module. When the intelligent alarm and control module receives the risk position, it generates a corresponding alarm message and sends it to the display alarm terminal.

[0008] Furthermore, the intelligent alarm and control module is communicatively connected to the control execution hidden danger analysis module. The intelligent alarm and control module sends the control information for the heating furnace to the control execution hidden danger analysis module. The control execution hidden danger analysis module analyzes the degree of control execution hidden danger for the heating furnace per unit time, generates a high control hidden danger signal or a low control hidden danger signal through the analysis, and sends the high control hidden danger signal to the intelligent alarm and control module when it is generated. When the intelligent alarm and control module receives the high control hidden danger signal, it generates a corresponding alarm message and sends it to the display alarm terminal.

[0009] Furthermore, the specific analysis process of the control execution hidden danger analysis module is as follows: Real-time collect the heating power and air flow velocity of the heating furnace, mark the deviation value of the heating power compared to the preset heating power standard value currently matched as the power detection value, and mark the deviation value of the air flow velocity compared to the preset air flow velocity standard value currently matched as the air velocity detection value; Numerically compare the power detection value and the air velocity detection value with the preset power detection threshold and the preset air velocity detection threshold respectively. If the power detection value or the air velocity detection value exceeds the corresponding preset threshold, it is judged that the current is in an execution obstacle state; Obtain the total duration of the heating furnace in the execution obstacle state per unit time and mark it as the execution obstacle value. Numerically compare the execution obstacle value with the preset execution obstacle threshold. If the execution obstacle value exceeds the preset execution obstacle threshold, generate a high control hidden danger signal; If the execution obstacle value does not exceed the preset execution obstacle threshold, calculate the average value of all power detection values per unit time to obtain the power execution judgment value, and calculate the average value of all air velocity detection values per unit time to obtain the air velocity execution judgment value; Calculate the heating furnace risk value by numerically calculating the execution obstacle value, the power execution judgment value, and the air velocity execution judgment value. Numerically compare the heating furnace risk value with the preset heating furnace risk threshold. If the heating furnace risk value exceeds the preset heating furnace risk threshold, generate a high control hidden danger signal; if the heating furnace risk value does not exceed the preset heating furnace risk threshold, generate a low control hidden danger signal.

[0010] Further, the intelligent alarm control module is communicatively connected to the supervision hidden danger auxiliary analysis module. The supervision hidden danger auxiliary analysis module is used to set the detection period, analyze the degree of supervision hidden danger for the heating furnace during the detection period, generate a supervision high hidden danger signal or a supervision low hidden danger signal through the analysis, and send the supervision high hidden danger signal to the intelligent alarm control module when it is generated. When the intelligent alarm control module receives the supervision high hidden danger signal, it generates corresponding alarm information and sends it to the display alarm terminal.

[0011] Further, the specific analysis process of the supervision hidden danger auxiliary analysis module includes: Collect the generation time of the corresponding alarm information and the time when the display alarm terminal displays and alarms, and mark them as the first time and the second time respectively. Calculate the time difference between the first time and the second time to obtain the silent duration; obtain all the silent durations during the detection period and calculate their average value to obtain the silent performance value, and mark the number of occurrences of the silent duration exceeding the preset silent duration threshold during the detection period as the silent anomaly value; And monitor the display alarm area corresponding to the display alarm terminal, and determine that the display alarm area is in a supervision risk state when there is no management personnel in the display alarm area; obtain the total duration of the display alarm area being in a supervision risk state during the detection period and mark it as the supervision risk value, and mark the number of occurrences of the single continuous duration of the display alarm area being in a supervision risk state exceeding the preset single continuous duration threshold during the detection period as the supervision risk frequency value; Calculate the supervision hidden danger decision value by performing numerical calculations on the silent performance value, the silent anomaly value, the supervision risk value, and the supervision risk frequency value. Compare the supervision hidden danger decision value with the preset supervision hidden danger decision threshold. If the supervision hidden danger decision value exceeds the preset supervision hidden danger decision threshold, generate a supervision high hidden danger signal; if the supervision hidden danger decision value does not exceed the preset supervision hidden danger decision threshold, generate a supervision low hidden danger signal.

[0012] Compared with the prior art, the beneficial effects of the present invention are: 1. In the present invention, the furnace temperature of the heating furnace is diagnosed and analyzed by the furnace temperature abnormal diagnosis module. When the furnace temperature diagnosis fails signal is generated, the operation of the heating furnace is suspended and corresponding improvement measures are taken for the heating furnace. When the furnace temperature diagnosis passes signal is generated, the furnace temperature is predicted and corresponding parameter adjustments are made, reducing manual intervention to lower the furnace temperature supervision difficulty of the heating furnace, and accurately capturing the risk positions in the heating furnace through analysis when the furnace temperature diagnosis fails signal is not generated during the monitoring period, so as to check the corresponding positions in time, further ensuring the subsequent operation effect of the heating furnace, with high intelligence and automation level; 2. In the present invention, the regulation and execution hidden danger analysis module is used to analyze the degree of regulation and execution hidden danger of the heating furnace per unit time. When a high-regulation hidden danger signal is generated, a cause investigation and analysis are carried out and reasonable countermeasures are taken to ensure the accuracy of subsequent regulation and execution, which is beneficial to improving the operation effect and operation safety of the heating furnace. Moreover, the supervision hidden danger auxiliary analysis module is used to analyze the degree of supervision hidden danger of the heating furnace during the detection period. When a high-supervision hidden danger signal is generated, the operation supervision of the heating furnace is strengthened to further ensure the operation effect of the heating furnace and the quality of the products produced. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings; Figure 1 It is a system block diagram of the first embodiment in the present invention; Figure 2 It is a system block diagram of the second and third embodiments in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0015] Embodiment 1: As Figure 1 shown, a heating furnace temperature monitoring system based on a neural network proposed by the present invention includes a comprehensive monitoring and transmission module, a furnace temperature abnormal diagnosis module, a risk position capture and output module, a neural network prediction module, an intelligent alarm and regulation module, and a display and alarm terminal; Among them, the comprehensive monitoring and transmission module collects temperature data at several positions inside the heating furnace, as well as auxiliary data including heating power, air flow velocity inside the furnace, and ambient temperature, etc., thereby obtaining a heating furnace monitoring data set, and sending the heating furnace monitoring data set to the furnace temperature abnormal diagnosis module and the neural network prediction module; The furnace temperature abnormal diagnosis module conducts furnace temperature abnormal diagnosis and analysis on the heating furnace, thereby generating a furnace temperature diagnosis qualified signal or a furnace temperature diagnosis unqualified signal. When the furnace temperature diagnosis unqualified signal is generated, it is sent to the intelligent alarm and regulation module, and the intelligent alarm and regulation module generates a corresponding warning message and sends it to the display and alarm terminal when receiving the furnace temperature diagnosis unqualified signal; The display alarm terminal displays the corresponding early warning information and issues an early warning to remind the management personnel to suspend the operation of the heating furnace as needed and conduct a cause investigation and analysis, and promptly take corresponding improvement measures for the heating furnace to ensure the uniformity and stability of the furnace temperature distribution inside the heating furnace and the operation effect of the heating furnace. The specific analysis process of the furnace temperature abnormal diagnosis and analysis is as follows: Obtain the temperatures at several positions in the heating furnace, establish a set of actual furnace temperatures for all positions, and calculate the variance of all subsets in the set of actual furnace temperatures. Accordingly, obtain the furnace temperature distribution abnormal coefficient, and compare the furnace temperature distribution abnormal coefficient with the preset furnace temperature distribution abnormal coefficient threshold. If the furnace temperature distribution abnormal coefficient exceeds the preset furnace temperature distribution abnormal coefficient threshold, it indicates that the furnace temperature distribution in the heating furnace is uneven, and a furnace temperature diagnosis unqualified signal is generated; If the furnace temperature distribution abnormal coefficient does not exceed the preset furnace temperature distribution abnormal coefficient threshold, it indicates that the furnace temperature distribution in the heating furnace is relatively uniform. Then, compare the temperature at the corresponding position with the preset temperature range that is currently matched. If the temperature at the corresponding position is not within the preset temperature range, it indicates that the temperature at the corresponding position is relatively abnormal, and the corresponding position is marked as a non-normal furnace temperature position; Obtain the ratio of the number of non-normal furnace temperature positions in the heating furnace and mark it as the furnace temperature abnormal point detection value. Compare the furnace temperature abnormal point detection value with the preset furnace temperature abnormal point detection threshold. If the furnace temperature abnormal point detection value exceeds the preset furnace temperature abnormal point detection threshold, it indicates that the furnace temperature performance in the heating furnace is relatively abnormal, and a furnace temperature diagnosis unqualified signal is generated; If the furnace temperature abnormal point detection value does not exceed the preset furnace temperature abnormal point detection threshold, calculate the average value of the temperatures at all positions to obtain the furnace temperature analysis value, and calculate the difference between the furnace temperature analysis value and the median of the preset temperature range and take the absolute value to obtain the furnace temperature deviation performance value; Compare the furnace temperature deviation performance value with the preset furnace temperature deviation performance threshold. If the furnace temperature deviation performance value exceeds the preset furnace temperature deviation performance threshold, it indicates that the furnace temperature performance in the heating furnace is relatively abnormal, and a furnace temperature diagnosis unqualified signal is generated; if the furnace temperature deviation performance value does not exceed the preset furnace temperature deviation performance threshold, it indicates that generally the furnace temperature performance in the heating furnace is relatively normal, and a furnace temperature diagnosis qualified signal is generated.

[0016] When generating a qualified furnace temperature diagnosis signal, the neural network prediction module predicts the furnace temperature within a certain period in the future based on the monitored data set of the heating furnace, and sends the furnace temperature prediction result to the intelligent alarm and regulation module; when receiving the furnace temperature prediction result, the intelligent alarm and regulation module intelligently regulates the operating state of the heating furnace, dynamically adjusts the heating power and air flow velocity parameters of the heating furnace, realizes precise control of the furnace temperature, improves production efficiency and product quality, and sends the regulation information to the display and alarm terminal in real time, which is conducive to detailed understanding of the automatic control information of the heating furnace. Through intelligent monitoring and analysis, it reduces manual intervention, reduces the difficulty of furnace temperature supervision of the heating furnace, and has a high level of intelligence and automation.

[0017] Furthermore, the risk position capture output module sets a monitoring period with a duration of L1. Preferably, L1 is ten minutes; when the duration reaches L1, all diagnostic result information of the furnace temperature abnormal diagnosis module during the monitoring period is obtained. If no unqualified furnace temperature diagnosis signal is generated during the monitoring period, the number of times the corresponding position in the heating furnace is marked as an abnormal furnace temperature position during the monitoring period is marked as the furnace temperature position abnormal frequency; The furnace temperature position abnormal frequency is numerically compared with the preset furnace temperature position abnormal frequency threshold. If the furnace temperature position abnormal frequency exceeds the preset furnace temperature position abnormal frequency threshold, the corresponding position is marked as a risk position; if there is a risk position during the monitoring period, the risk position is sent to the intelligent alarm and regulation module. When the intelligent alarm and regulation module receives the risk position, it generates corresponding alarm information and sends it to the display and alarm terminal. The display and alarm terminal displays the corresponding alarm information and issues a warning to remind the management personnel to suspend the operation of the heating furnace as needed and check the corresponding position, further ensuring the subsequent operation effect of the heating furnace.

[0018] Embodiment 2: As Figure 2 shown, the difference between this embodiment and Embodiment 1 is that the intelligent alarm and regulation module is communicatively connected to the regulation execution hidden danger analysis module. The intelligent alarm and regulation module sends the regulation information for the heating furnace to the regulation execution hidden danger analysis module. The regulation execution hidden danger analysis module analyzes the degree of regulation execution hidden danger for the heating furnace per unit time and generates a regulation high hidden danger signal or a regulation low hidden danger signal through the analysis; And when generating a regulation high hidden danger signal, it is sent to the intelligent alarm and regulation module. When the intelligent alarm and regulation module receives the regulation high hidden danger signal, it generates corresponding alarm information and sends it to the display and alarm terminal. The display and alarm terminal displays the corresponding alarm information and issues a warning to remind the management personnel to conduct cause investigation and analysis and make reasonable countermeasures to ensure the subsequent regulation execution accuracy, which is conducive to improving the operation effect and operation safety of the heating furnace; the specific analysis process of the regulation execution hidden danger analysis module is as follows: The heating power and air flow velocity of the heating furnace are collected in real time. The deviation value of the heating power from the preset heating power standard value currently matched is marked as the power detection value, and the deviation value of the air flow velocity from the preset air flow velocity standard value currently matched is marked as the air velocity detection value; The power detection value and the air velocity detection value are respectively compared numerically with the preset power detection threshold and the preset air velocity detection threshold. If the power detection value or the air velocity detection value exceeds the corresponding preset threshold, indicating that the real-time execution status is poor, it is determined that the current is in an execution obstacle state; The total duration of the heating furnace in the execution obstacle state within a unit time is obtained and marked as the execution obstacle value. The execution obstacle value is compared numerically with the preset execution obstacle threshold. If the execution obstacle value exceeds the preset execution obstacle threshold, a regulation high-risk signal is generated; If the execution obstacle value does not exceed the preset execution obstacle threshold, the average value of all power detection values within a unit time is calculated to obtain the power execution judgment value, and the average value of all air velocity detection values within a unit time is calculated to obtain the air velocity execution judgment value; The heating furnace execution risk value is obtained by numerically calculating the execution obstacle value, the power execution judgment value, and the air velocity execution judgment value. That is, corresponding preset weight coefficients are assigned to the execution obstacle value, the power execution judgment value, and the air velocity execution judgment value in advance, and the execution obstacle value, the power execution judgment value, and the air velocity execution judgment value are respectively multiplied by the corresponding preset weight coefficients, and the results of the three groups of multiplications are summed up to obtain the heating furnace execution risk value; among them, the larger the value of the heating furnace execution risk value, the higher the comprehensive regulation execution hidden danger of the heating furnace within a unit time; The heating furnace execution risk value is compared numerically with the preset heating furnace execution risk threshold. If the heating furnace execution risk value exceeds the preset heating furnace execution risk threshold, indicating that the comprehensive regulation execution hidden danger of the heating furnace within a unit time is relatively high and it is not conducive to ensuring the furnace temperature stability and operation effect of the heating furnace, a regulation high-risk signal is generated; if the heating furnace execution risk value does not exceed the preset heating furnace execution risk threshold, indicating that the comprehensive regulation execution hidden danger of the heating furnace within a unit time is relatively low, a regulation low-risk signal is generated.

[0019] Embodiment 3: As Figure 2 shown, the difference between this embodiment and Embodiment 1 and Embodiment 2 is that the intelligent alarm regulation module is communicatively connected to the supervision hidden danger auxiliary analysis module. The supervision hidden danger auxiliary analysis module is used to set the detection period, analyze the supervision hidden danger degree of the heating furnace during the detection period, and generate a supervision high-risk signal or a supervision low-risk signal through the analysis; And when a high-risk supervision signal is generated, it is sent to the intelligent alarm control module. When the intelligent alarm control module receives the high-risk supervision signal, it generates corresponding alarm information and sends it to the display alarm terminal. The display alarm terminal displays the alarm information and issues a warning to strengthen the subsequent operation supervision of the heating furnace in a timely manner, further ensuring the operation effect of the heating furnace and the quality of the products produced. The specific analysis process of the supervision risk auxiliary analysis module is as follows: Collect the generation time of the corresponding alarm information and the time when the display alarm terminal displays and alarms, and mark them as the first time and the second time respectively. Calculate the time difference between the first time and the second time to obtain the silent duration. Among them, the larger the value of the silent duration, the slower the response to display and alarm for the corresponding alarm information, and the less conducive to making reasonable response measures in a timely manner; Obtain all the silent durations during the detection period and calculate their average value to obtain the silent performance value. Compare the silent duration with the preset silent duration threshold, and mark the number of times that the silent duration exceeds the preset silent duration threshold during the detection period as the silent anomaly value; And monitor the display alarm area corresponding to the display alarm terminal. When there is no management personnel in the display alarm area, it is judged that the display alarm area is in a supervision risk state. Obtain the total duration that the display alarm area is in a supervision risk state during the detection period and mark it as the supervision risk value, and mark the number of times that the single continuous duration that the display alarm area is in a supervision risk state during the detection period exceeds the preset single continuous duration threshold as the supervision risk frequency value; Calculate the supervision risk decision value by numerically calculating the silent performance value, the silent anomaly value, the supervision risk value, and the supervision risk frequency value. That is, assign corresponding preset weight coefficients to the silent performance value, the silent anomaly value, the supervision risk value, and the supervision risk frequency value in advance, multiply the silent performance value, the silent anomaly value, the supervision risk value, and the supervision risk frequency value by the corresponding preset weight coefficients respectively, and sum the four groups of product results to obtain the supervision risk decision value. Among them, the larger the value of the supervision risk decision value, the higher the comprehensive supervision risk degree of the heating furnace during the detection period; Compare the supervision risk decision value with the preset supervision risk decision threshold. If the supervision risk decision value exceeds the preset supervision risk decision threshold, it indicates that the comprehensive supervision risk degree of the heating furnace during the detection period is relatively high, and a high-risk supervision signal is generated; if the supervision risk decision value does not exceed the preset supervision risk decision threshold, it indicates that the comprehensive supervision risk degree of the heating furnace during the detection period is relatively low, and a low-risk supervision signal is generated.

[0020] Working principle of the present invention: During use, the operation of the heating furnace is monitored by the comprehensive monitoring transmission module to collect the monitoring data set of the heating furnace. The furnace temperature abnormal diagnosis module conducts furnace temperature abnormal diagnosis and analysis on the heating furnace. When a furnace temperature diagnosis unqualified signal is generated, the operation of the heating furnace is paused and corresponding improvement measures are taken for the heating furnace to ensure the uniformity and stability of the furnace temperature distribution inside the heating furnace. When a furnace temperature diagnosis qualified signal is generated, the furnace temperature in the next period of time is predicted by the neural network prediction module. The intelligent alarm and regulation module makes corresponding parameter adjustments based on the furnace temperature prediction result to control the furnace temperature, improve production efficiency and product quality, reduce manual intervention to lower the difficulty of furnace temperature supervision of the heating furnace, and capture the risk positions inside the heating furnace through analysis when no furnace temperature diagnosis unqualified signal is generated during the monitoring period. When the risk positions are captured, the operation of the heating furnace is paused and the corresponding positions are inspected to further ensure the subsequent operation effect of the heating furnace, with high levels of intelligence and automation.

[0021] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, many modifications and changes can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A heating furnace temperature monitoring system based on neural network, characterized in that: It includes a comprehensive monitoring transmission module, a furnace temperature abnormality diagnosis module, a neural network prediction module, an intelligent alarm control module and a display alarm terminal; the comprehensive monitoring transmission module collects the heating furnace monitoring data set, and sends the heating furnace monitoring data set to the furnace temperature abnormality diagnosis module and the neural network prediction module; the furnace temperature abnormality diagnosis module performs furnace temperature abnormality diagnosis analysis on the heating furnace, and when a furnace temperature diagnosis unqualified signal is generated, it is sent to the intelligent alarm control module; When a furnace temperature diagnosis qualification signal is generated, the furnace temperature in the future period is predicted based on the heating furnace monitoring data set through the neural network prediction module, and the furnace temperature prediction result is sent to the intelligent alarm and control module; the intelligent alarm and control module generates corresponding early warning information when receiving a furnace temperature diagnosis failure signal and sends it to the display alarm terminal, and intelligently controls the operating status of the heating furnace when receiving the furnace temperature prediction result, and sends the control information to the display alarm terminal in real time.

2. The heating furnace temperature monitoring system based on neural network according to claim 1 is characterized in that: The specific analysis process of abnormal furnace temperature diagnosis and analysis includes: The temperatures at several positions in the heating furnace are obtained, and the actual temperature set in the furnace is established based on the temperatures at all positions. The variance of all subsets in the actual temperature set in the furnace is calculated, and the furnace temperature distribution anomaly coefficient is obtained based on this. If the furnace temperature distribution anomaly coefficient exceeds the preset furnace temperature distribution anomaly coefficient threshold, a furnace temperature diagnosis failure signal is generated.

3. The heating furnace temperature monitoring system based on neural network according to claim 2 is characterized in that: If the furnace temperature distribution anomaly coefficient does not exceed the preset furnace temperature distribution anomaly coefficient threshold, the number of abnormal furnace temperature positions in the heating furnace is obtained and marked as the furnace temperature outlier detection value. If the furnace temperature outlier detection value exceeds the preset furnace temperature outlier detection threshold, a furnace temperature diagnosis unqualified signal is generated; If the furnace temperature outlier detection value does not exceed the preset furnace temperature outlier detection threshold, the furnace temperature deviation performance value is numerically compared with the preset furnace temperature deviation performance threshold. If the furnace temperature deviation performance value exceeds the preset furnace temperature deviation performance threshold, a furnace temperature diagnosis failure signal is generated; otherwise, a furnace temperature diagnosis pass signal is generated.

4. The heating furnace temperature monitoring system based on neural network according to claim 1 is characterized in that: The furnace temperature abnormality diagnosis module is connected to the dangerous position capture output module in communication. The dangerous position capture output module is used to set a monitoring period with a duration of L1. When the duration reaches L1, if no furnace temperature diagnosis failure signal is generated during the monitoring period, the number of times the corresponding position in the heating furnace is marked as an abnormal furnace temperature position during the monitoring period is marked as the furnace temperature position abnormal frequency; If the furnace temperature position frequency exceeds the preset furnace temperature position frequency threshold, the corresponding position will be marked as a risk position; if there is a risk position during the monitoring period, the risk position will be sent to the intelligent alarm control module. When the intelligent alarm control module receives the risk position, it will generate corresponding alarm information and send it to the alarm display terminal.

5. The heating furnace temperature monitoring system based on neural network according to claim 1 is characterized in that: The intelligent alarm control module is communicated with the control execution hidden danger analysis module. The intelligent alarm control module sends the control information of the heating furnace to the control execution hidden danger analysis module. The control execution hidden danger analysis module analyzes the degree of control execution hidden danger of the heating furnace per unit time, and generates a high control hidden danger signal or a low control hidden danger signal through analysis. When the high control hidden danger signal is generated, it is sent to the intelligent alarm control module. When the intelligent alarm control module receives the high control hidden danger signal, it generates corresponding alarm information and sends it to the alarm display terminal.

6. The heating furnace temperature monitoring system based on neural network according to claim 5 is characterized in that: The specific analysis process of the control execution hidden danger analysis module is as follows: The heating power and airflow speed of the heating furnace are collected in real time. If the power detection value or the airflow speed detection value exceeds the corresponding preset threshold, it is judged that the current state is in an execution obstruction state; the total time of the heating furnace in the execution obstruction state per unit time is obtained and marked as the execution obstruction value. If the execution obstruction value exceeds the preset execution obstruction threshold, a high risk signal of regulation is generated; If the execution obstacle value does not exceed the preset execution obstacle threshold, the heating furnace risk value is obtained by numerically calculating the execution obstacle value, the power judgment value and the gas speed judgment value. If the heating furnace risk value exceeds the preset heating furnace risk threshold, a high control risk signal is generated; otherwise, a low control risk signal is generated.

7. The heating furnace temperature monitoring system based on neural network according to claim 6 is characterized in that: The intelligent alarm control module is communicated with the supervision hidden danger auxiliary analysis module. The supervision hidden danger auxiliary analysis module is used to set the detection period, analyze the degree of supervision hidden danger of the heating furnace during the detection period, and send it to the intelligent alarm control module when a high supervision hidden danger signal is generated. When the intelligent alarm control module receives the high supervision hidden danger signal, it generates corresponding alarm information and sends it to the alarm display terminal.

8. The heating furnace temperature monitoring system based on neural network according to claim 7 is characterized in that: The specific analysis process of the regulatory hidden danger auxiliary analysis module includes: The regulatory hidden danger decision value is obtained by numerically calculating the silent performance value, silent abnormal value, regulatory risk time value and regulatory risk frequency value. If the regulatory hidden danger decision value exceeds the preset regulatory hidden danger decision threshold, a high regulatory hidden danger signal is generated; otherwise, a low regulatory hidden danger signal is generated.

Images