A VOCs energy efficiency automatic management system for a coating operation exhaust treatment device

Abstract

This invention provides an automatic VOCs energy efficiency management system for a coating operation exhaust gas treatment device, comprising: an energy consumption optimization analysis platform for analyzing the current coating work plan to obtain an accurate VOCs concentration prediction curve and performing energy consumption optimization analysis based on the VOCs concentration prediction curve; and an equipment management platform for adjusting the equipment according to the analysis results of the energy consumption optimization analysis platform. Within the energy consumption optimization analysis platform, a current coating work plan acquisition module acquires and records the current coating work plan under the current production conditions of the enterprise; a prediction module uses the recorded current coating work plan to derive a VOCs online monitoring concentration prediction curve through a VOCs online monitoring concentration prediction model; and a verification module uses the VOCs online monitoring concentration prediction curve through a verification model to derive an accurate VOCs concentration prediction curve. This invention's system achieves automated control based on the prediction of the concentration curve according to the coating work plan and the prediction results.

Description

Technical Field

[0001] This invention relates to an automatic VOCs energy efficiency management system for a coating operation exhaust gas treatment device, belonging to the field of computer automation management technology. Background Technology

[0002] Shipbuilding companies require extensive painting operations, which generate volatile organic compounds (VOCs). VOCs are harmful substances and must be treated to meet emission standards before being released into the atmosphere. Currently, shipbuilding companies typically use a "zeolite rotor + RTO" method for VOC treatment. The VOCs undergo thermal desorption via the zeolite rotor, and the desorbed VOCs are then fed into the RTO (Regenerative Thermal Oxidizer) for combustion and purification. However, painting operations are intermittent, not continuous, resulting in significant fluctuations in the concentration and composition of the generated VOCs. In actual production, the energy efficiency management of waste gas treatment devices is carried out manually by staff based on the concentration data monitored by the online concentration detection instruments installed in the regenerative thermal oxidizer. Monitoring concentration does not equate to accuracy, and manual management relies on human subjectivity and past experience. These two factors make it difficult to achieve high accuracy in manual control. Therefore, in order to avoid the inability of natural gas in the furnace to maintain the high temperature conditions required for VOCs waste gas combustion due to inaccurate human judgment, shipbuilding companies often use excessive natural gas to assist combustion in order to maintain the high temperature of the combustion chamber. However, this energy efficiency management method consumes a lot of energy and creates a significant energy burden on the company's operation.

[0003] Based on the above situation, in order to achieve energy conservation and emission reduction in the treatment of exhaust gas in shipbuilding enterprises, there is an urgent need for a precise and efficient automatic VOCs energy efficiency management system. Summary of the Invention

[0004] The purpose of this invention is to provide an automatic VOCs energy efficiency management system for a coating operation exhaust gas treatment device.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] This invention provides an automatic VOCs energy efficiency management system for a coating operation exhaust gas treatment device. It is characterized by comprising: an energy consumption optimization analysis platform, used to analyze the current coating work plan to obtain an accurate VOCs concentration prediction curve and perform energy consumption optimization analysis based on the VOCs concentration prediction curve; and an equipment management platform, used to adjust the equipment accordingly based on the analysis results of the energy consumption optimization analysis platform. The energy consumption optimization analysis platform includes: a current coating work plan acquisition module, used to acquire and record the current coating work plan under the current production conditions of the enterprise; a model storage module, storing a VOCs online monitoring concentration prediction model and a verification model; an optimization rule storage module, used to store energy consumption optimization rules; a prediction module, used to derive a VOCs online monitoring concentration prediction curve from the recorded current coating work plan through the VOCs online monitoring concentration prediction model; a verification module, used to derive an accurate VOCs concentration prediction curve from the VOCs online monitoring concentration prediction curve through a verification model; and an optimization calculation module, used to calculate the corresponding analysis results based on the VOCs online monitoring concentration prediction curve and the energy consumption optimization rules.

[0007] Furthermore, the VOCs energy efficiency automatic management system for the coating operation exhaust gas treatment device provided by this invention is characterized by: further including a modeling platform, which includes: a modeling coating work plan management module, which collects and records historical coating work plans that the enterprise has implemented; a modeling online monitoring concentration management module, which collects and records historical online monitoring concentration data that was simultaneously monitored during the production process of implementing historical coating work plans; a standard reference concentration storage module, which stores standard reference concentration data, which is the concentration obtained by directly sampling and analyzing VOCs gas during the production process of implementing historical coating work plans; a first modeling module, which constructs a VOCs online monitoring concentration prediction model through the correlation between historical coating work plans and historical online monitoring concentration data; and a second modeling module, which constructs a verification model through the correlation between historical online monitoring concentration data and standard reference concentration data.

[0008] Furthermore, the VOCs energy efficiency automatic management system for the coating operation exhaust gas treatment device provided by the present invention may also have the following features: the energy consumption optimization analysis platform further includes: a chart interface display module, used to display various data tables and various parameter curves collected and recorded or calculated during system operation; the data tables include, but are not limited to, recording the current coating work plan in tabular form; the parameter curves include, but are not limited to, VOCs online monitoring concentration prediction curves, VOCs accurate concentration prediction curves, and natural gas consumption curve comparison charts before and after optimization.

[0009] Furthermore, the VOCs energy efficiency automatic management system of the coating operation exhaust gas treatment device provided by the present invention may also have the following features: wherein, the energy consumption optimization rules include the host air volume calculation rules, the desorption fan air volume calculation rules, the concentration ratio calculation rules, and the natural gas consumption calculation rules; the equipment management platform includes: a host air volume control module for controlling the host air volume, a desorption fan air volume control module for controlling the desorption fan air volume, a natural gas consumption control module for controlling the natural gas intake, and a concentration ratio control module for controlling the concentration ratio.

[0010] The beneficial effects of this invention are:

[0011] The VOCs energy efficiency automatic management system for coating operation exhaust gas treatment device provided by this invention is equipped with an energy consumption optimization analysis platform and an equipment management platform, realizing automatic management of exhaust gas treatment energy efficiency. From the perspective of enterprise management, this system is automated management by computer software, which has the advantages of high efficiency and timeliness, and reduced manpower and material resources compared with the manual management of existing technologies. From the perspective of enterprise energy saving, this system optimizes VOCs energy efficiency, which can greatly reduce the energy consumption of exhaust gas treatment, alleviate the energy consumption burden of enterprises, and has the advantages of energy saving and emission reduction. Attached Figure Description

[0012] Figure 1 This is a system architecture diagram of the VOCs energy efficiency automatic management system of the painting operation exhaust gas treatment device in this embodiment of the invention;

[0013] Figure 2 These are the online monitoring concentration curve actually measured at the main pipe inlet before optimization in a certain working condition in this embodiment of the invention, the predicted online monitoring concentration curve of VOCs, and the verified accurate concentration prediction curve of VOCs.

[0014] Figure 3 This is a comparison chart of natural gas consumption curves before and after optimization under a certain operating condition in an embodiment of the present invention. Detailed Implementation

[0015] To make the technical means, creative features, objectives and effects of this invention easier to understand, the following embodiments, in conjunction with the accompanying drawings, will specifically illustrate the technical solution of this invention.

[0016] This embodiment provides an automatic VOCs energy efficiency management system for a painting operation exhaust gas treatment device. It is used in shipbuilding enterprises to automatically manage the energy efficiency of exhaust gas treatment devices during the exhaust gas treatment production process. The exhaust gas treatment device refers to a complete exhaust gas treatment device using the "zeolite rotor + RTO" method.

[0017] like Figure 1As shown, the VOCs energy efficiency automatic management system of the coating operation exhaust gas treatment device in this embodiment includes: a modeling platform 1, an energy consumption optimization analysis platform 2, and an equipment management platform 3. The modeling platform 1 is used for online monitoring and concentration prediction models and verification models for VOCs. The energy consumption optimization analysis platform 2 is used to analyze the current coating work plan to obtain an accurate VOCs concentration prediction curve and perform energy consumption optimization analysis based on the VOCs concentration prediction curve. The equipment management platform 3 is used to adjust the equipment accordingly based on the analysis results of the energy consumption optimization analysis platform.

[0018] Modeling platform 1 includes: modeling and painting work plan management module 11, modeling online monitoring concentration management module 12, standard reference concentration storage module 13, first modeling module 14, and second modeling module 15.

[0019] The modeling and painting work plan management module 11 collects and records historical painting work plans that the company has already implemented. The painting work plan S consists of N work conditions S. i Composition, i = 1, ..., N. Each working condition S i The operating data includes the start time of the operation. Homework end time Number of spray guns (G) i Paint volume P i Solvent usage Q i Spraying area A i The collected historical painting work plans, including the number of segments and working condition data, are recorded in tabular form. These historical painting work plans are used for modeling. The resulting set of working condition data for each segment is represented as follows:

[0020]

[0021] The online monitoring concentration management module 12 collects and records historical online monitoring concentration data that is synchronously monitored during the production process of implementing historical coating work plans. Because the segmented work conditions in the coating work plan S have a high degree of dispersion, the operation times of the segmented work conditions are too scattered on the work condition data record table. Correspondingly, the historical online monitoring concentration data collected for the historical coating work plan S shows multiple bands in the curve graph. The collected historical online monitoring concentration data is recorded using a curve graph (this curve graph is a curve graph of concentration changing over time), and each band in the curve graph is associated with the segment number of the corresponding work condition. For example, the first band in the historical online monitoring concentration curve is associated with the first work condition of the historical coating work plan.

[0022] The standard reference concentration storage module 13 stores standard reference concentration data. Currently, the monitoring results of VOC concentration detection instruments on regenerative thermal oxidizers (the online monitoring concentration data collected by the modeling online monitoring concentration management module) are often greater than the true values ​​(values ​​obtained through concentration analysis of direct sampling of VOCs gas), and the monitored values ​​are not accurate enough. The standard reference concentration data stored in the standard reference concentration storage module are concentration values ​​obtained by the company through synchronous direct sampling and analysis of VOCs gas by a third-party testing agency during the production process of implementing historical coating work plans.

[0023] The first modeling module 14 constructs a VOCs online monitoring concentration prediction model based on the correlation between historical coating work plans and historical online monitoring concentration data. In this embodiment, the first modeling module 14 uses a log-Gaussian function to fit the VOCs online monitoring concentration curve to facilitate subsequent data analysis; and uses an XGBoost model to construct the mapping relationship. The specific workflow of the first modeling module 14 is as follows:

[0024] First, curve fitting was performed on the online monitoring concentrations for each operating condition:

[0025] ① Any i-th working condition S i The online monitoring concentration of VOCs was fitted using the logarithmic Gaussian function of formula (1):

[0026]

[0027] In formula (1), t represents the working condition S. i At a certain moment in time; The value represents the concentration monitored online at time t; e represents the natural logarithm; t i This indicates that the operating condition S i The time difference between the start time of the current painting operation and the start time of the planned painting operation (the start time of the painting operation plan S is the start time of the first working condition, and the end time of the painting operation plan S is the end time of the last working condition); 1n(*) represents the natural logarithm function; K i μ i σ i 2 These are the characteristic parameters of the logarithmic Gaussian function, calculated from the Gaussian curve.

[0028] ② Define the entire coating operation plan S. Online monitoring of concentration can be achieved through N operating conditions S. i The concentrations are summed, and it is represented as:

[0029]

[0030] In formula (2), This indicates the online monitoring concentration of the coating operation plan S.

[0031] Then, an online monitoring concentration prediction model is constructed:

[0032] S for each working condition is obtained from formula (1). i This will fit a Gaussian curve, thus obtaining a set of characteristic parameters K. i μ i σ i 2 Then, the set of characteristic parameters {K} is obtained for the N working conditions of the entire painting operation plan S. i μ i , σ i 2}, i = 1, ..., N;

[0033] The set of operating conditions {S} is constructed using the XGBoost model. i}, i = 1, ..., N, the set of each working condition and characteristic parameter {K i μ i , σ i 2 The mapping relationship g of the feature parameters in groups i = 1, ..., N is expressed as follows:

[0034]

[0035] Based on formula (2), the following online monitoring concentration prediction model for coating operation plan is constructed:

[0036]

[0037] In formula (4), This represents the predicted concentration from online monitoring of the current coating operation plan; the current coating operation plan is divided into M working conditions S. i , i = 1, ..., M; This indicates the working condition S in the current painting operation plan. i The online monitoring concentration prediction results are based on the operating conditions S in the current coating operation plan. i Gaussian curves are plotted based on a set of characteristic parameters of a logarithmic Gaussian function obtained from formula (3).

[0038] The second modeling module 15 constructs a verification model based on the correlation between historical online monitoring concentration data and standard reference concentration data. Since the online monitoring concentration is not the true value, this verification model can map the online monitoring concentration to the true value. In this embodiment, the second modeling module 15 uses the GPR model to construct a verification and correction model based on the mapping relationship between the online monitoring concentration and the standard reference concentration of the modeling coating operation plan, as shown below:

[0039]

[0040] Among them, C O Indicates the concentration monitored online; C L Indicates the standard reference concentration.

[0041] The energy consumption optimization analysis platform 2 includes: a current coating work plan acquisition module 21, a model storage module 22, an optimization rule storage module 23, a prediction module 24, a verification module 25, an optimization calculation module 26, and a chart interface display module 27.

[0042] The current painting work plan acquisition module 21 is used to acquire and record the current painting work plan under the current production conditions of the enterprise. The number of segments and working condition data of the acquired current painting work plan are recorded in tabular form.

[0043] The model storage module 22 stores the VOCs online monitoring concentration prediction model obtained by the first modeling module and the verification model obtained by the second modeling module.

[0044] Based on the recorded current coating work plan, the prediction module 24 derives the VOCs online monitoring concentration prediction curve through the VOCs online monitoring concentration prediction model.

[0045] The verification module 25 uses the online monitoring concentration prediction curve of VOCs to derive the accurate concentration prediction curve of VOCs through the verification model.

[0046] The optimization rule storage module 23 is used to store energy consumption optimization rules. The energy consumption optimization rules include the host air volume calculation rules, the desorption fan air volume calculation rules, the concentration ratio calculation rules, and the natural gas consumption calculation rules.

[0047] ①The rules for calculating the air volume of the main unit are as follows:

[0048]

[0049] Where L1 represents the design value of the main fan's air volume; This indicates the air volume of the main fan during actual operation.

[0050] The painting and curing stages are divided according to the accurate concentration prediction curve: the time point at which the extreme value of the last peak of the accurate concentration prediction curve occurs is the dividing line. The time period before this point is the painting stage, and the time period after this point is the curing stage.

[0051] ②The calculation rules for the air volume of the desorption fan are as follows:

[0052]

[0053] Where, ρU These are the equipment design parameters for the main fan, ρ L These are the equipment design parameters of the desorption fan, all of which are known parameters.

[0054] ③ The rules for calculating the concentration ratio are as follows:

[0055] Concentration ratio The upper and lower limits of the concentration ratio are calculated based on the ratio of the main fan air volume to the desorption fan air volume.

[0056]

[0057] The actual concentration ratio R during the operation of the RTO equipment True :

[0058]

[0059] Here, F refers to the amount of natural gas used.

[0060] ④ The rules for calculating natural gas consumption are as follows:

[0061] First, calculate the concentration C on the accurate concentration prediction curve. L Critical point C that does not require natural gas for combustion T :

[0062]

[0063] in, The RTO equipment maintains the concentration of VOCs for spontaneous combustion.

[0064] When C L >C T When C is zero, the natural gas consumption is 0; when C is zero, the natural gas consumption is 0. L ≤C T At that time, the natural gas consumption is the same as that under the unoptimized operating conditions.

[0065] Then, based on the critical point C T Calculate the time T when the gas supply stops. Start and the start time of gas supply T End .

[0066] In the accurate concentration prediction curve graph with time as the horizontal axis and organic waste gas concentration as the vertical axis, the critical point C is plotted. T Critical point C T It is a straight line parallel to the horizontal axis, and in this embodiment, the critical point is C. T The VOCs concentration is 200 mg / m³ 3The straight line intersects the accurate concentration prediction curve. The intersection point is determined as follows: if the slope of the accurate concentration prediction curve at the intersection point is positive, then the time point on the x-axis corresponding to the intersection point is the time point T at which gas supply ceases. End For example, in this embodiment Figure 2 T shown End , Figure 3 This indicates that gas supply will stop at that moment; if the slope of the accurate concentration prediction curve at the intersection point is negative, then the time point on the x-axis corresponding to the intersection point is the gas supply start time T. Start For example, in this embodiment Figure 2 T shown Start , Figure 3 This indicates that gas supply will begin at that moment.

[0067] The optimization calculation module 26 is used to calculate the corresponding analysis results by running the above energy consumption optimization rules based on the VOCs online monitoring concentration prediction curve.

[0068] The chart interface display module 27 is used to display various data tables and parameter curves collected, recorded, or calculated during system operation. Data tables include, but are not limited to, records of the current coating work plan in tabular form. Parameter curves include, but are not limited to, online VOCs concentration prediction curves, accurate VOCs concentration prediction curves, and comparison curves of natural gas consumption before and after optimization.

[0069] The equipment management platform 3 includes: a main unit airflow control module 31, a desorption fan airflow control module 32, a natural gas consumption control module 33, and a concentration ratio control module 34. The main unit airflow control module 31 controls the main unit airflow based on the calculation results of the main unit airflow calculation rules. The desorption fan airflow control module 32 controls the desorption fan airflow based on the calculation results of the desorption fan airflow calculation rules. The natural gas consumption control module 33 controls the natural gas intake based on the calculation results of the natural gas consumption calculation rules. The concentration ratio control module 34 controls the concentration ratio based on the calculation results of the concentration ratio calculation rules. When using this invention system to implement energy efficiency management of the waste gas treatment device, the corresponding control valves and operation buttons in the waste gas treatment device need to be replaced with electrically controlled valves and buttons. The control operation of each control module in the equipment management platform 3 outputs corresponding control electrical signals to the corresponding electrically controlled valves and buttons.

[0070] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An automatic VOCs energy efficiency management system for a coating operation exhaust gas treatment device, characterized in that, include: Modeling platform for VOCs online monitoring concentration prediction model and validation model; The energy consumption optimization analysis platform is used to analyze the current coating work plan to obtain an accurate VOCs concentration prediction curve and perform energy consumption optimization analysis based on the VOCs concentration prediction curve. The equipment management platform is used to adjust the equipment accordingly based on the analysis results from the energy consumption optimization analysis platform. The modeling platform includes: The modeling and painting work plan management module collects and records historical painting work plans that the company has already implemented. Each painting work plan S consists of N work conditions. i Composition, i = 1, ..., N, each working condition S i The operating data includes the start time of the operation. Homework end time Number of spray guns (G) i Paint volume P i Solvent usage Q i Spraying area A i The number of segments and working condition data of the collected historical painting work plans are recorded in tabular form. The recorded historical painting work plans are used for modeling, and the resulting set of working condition data for each segment is represented as follows: The modeling online monitoring concentration management module collects and records historical online monitoring concentration data that is synchronously monitored during the implementation of historical coating work plans. The standard reference concentration storage module stores standard reference concentration data, which is the concentration obtained by directly sampling and analyzing VOCs gas during the production process of implementing the company's historical coating work plan. The first modeling module constructs a VOCs online monitoring concentration prediction model by analyzing the correlation between historical coating work plans and historical online monitoring concentration data. It uses a log-Gaussian function to fit the VOCs online monitoring concentration curve and employs an XGBoost model to build the mapping relationship. The specific workflow of the first modeling module is as follows: First, curve fitting was performed on the online monitoring concentrations for each operating condition: ① Any i-th working condition S i The online monitoring concentration of VOCs was fitted using the logarithmic Gaussian function of formula (1): In formula (1), t represents the working condition S. i At a certain moment in time; The value represents the concentration monitored online at time t; e represents the natural logarithm; t i This indicates that the operating condition S i The time difference between the start time of the current painting operation and the start time of the planned painting operation (the start time of the painting operation plan S is the start time of the first working condition, and the end time of the painting operation plan S is the end time of the last working condition); ln(*) represents the natural logarithm function; K i μ i σ i 2 These are the characteristic parameters of the logarithmic Gaussian function, calculated from the Gaussian curve; ② Define the entire coating operation plan S. Online monitoring of concentration can be achieved through N operating conditions S. i The concentrations are summed, and it is represented as: In formula (2), This indicates the online monitoring concentration of the coating operation plan S; Then, an online monitoring concentration prediction model is constructed: S for each working condition is obtained from formula (1). i This will fit a Gaussian curve, thus obtaining a set of characteristic parameters K. i μ i σ i 2 Then, the set of characteristic parameters {K} is obtained for the N working conditions of the entire painting operation plan S. i μ i , σ i 2 }, i = 1, ..., N; The set of operating conditions {S} is constructed using the XGBoost model. i }, i = 1, ..., N, the set of each working condition and characteristic parameter {K i μ i , σ i 2 The mapping relationship g of the feature parameters in i = 1, ..., N is expressed as follows: Based on formula (2), the following online monitoring concentration prediction model for coating operation plan is constructed: In formula (4), This represents the predicted concentration from online monitoring of the current coating operation plan; the current coating operation plan is divided into M working conditions S. i , i = 1, ..., M; This indicates the working condition S in the current painting operation plan. i The online monitoring concentration prediction results are based on the operating conditions S in the current coating operation plan. i Gaussian curves are plotted based on a set of characteristic parameters of a logarithmic Gaussian function obtained from formula (3); The second modeling module constructs a verification model by establishing the correlation between historical online monitoring concentration data and standard reference concentration data. This module uses the GPR model h to construct the mapping relationship between the online monitoring concentration and the standard reference concentration of the coating operation plan, establishing a verification and correction model, as shown below: Among them, C O Indicates the concentration monitored online; C L Indicates the standard reference concentration; The energy consumption optimization and analysis platform includes: The current coating work plan acquisition module is used to acquire and record the current coating work plan under the current production conditions of the enterprise; The model storage module stores the online monitoring concentration prediction model and the validation model for VOCs; An optimized rule storage module is used to store energy consumption optimization rules; The prediction module is used to derive the VOCs online monitoring concentration prediction curve from the recorded current coating work plan through the VOCs online monitoring concentration prediction model; The verification module is used to derive an accurate VOCs concentration prediction curve from the online monitoring concentration prediction curve of VOCs through the verification model; The optimization calculation module is used to calculate the corresponding analysis results based on the VOCs online monitoring concentration prediction curve and the energy consumption optimization rules.

2. The VOCs energy efficiency automatic management system for the coating operation exhaust gas treatment device as described in claim 1, characterized in that: in, The energy consumption optimization analysis platform also includes: a chart interface display module, used to display various data tables and various parameter curves collected and recorded or calculated during system operation; The data table includes, but is not limited to, recording the current painting work plan in tabular form; The parameter curves include, but are not limited to, VOCs online monitoring concentration prediction curves, VOCs accurate concentration prediction curves, and natural gas consumption curves before and after optimization.

3. The VOCs energy efficiency automatic management system of the coating operation exhaust gas treatment device as described in claim 1, Its features are: The energy consumption optimization rules include rules for calculating the air volume of the main unit, the air volume of the desorption fan, the concentration ratio, and the natural gas consumption. The equipment management platform includes: a main unit airflow control module for regulating the airflow of the main unit, a desorption fan airflow control module for regulating the airflow of the desorption fan, a natural gas consumption control module for regulating the natural gas intake, and a concentration ratio control module for regulating the concentration ratio.

Images