A ship painting plant field environment monitoring system

Abstract

The application discloses a kind of ship painting workshop field environment monitoring systems, including field detection unit, control center and data acquisition network, field detection unit is installed in the integrated monitoring point of ship painting workshop field, for measuring the environmental parameter of integrated monitoring point;Control center is connected with field detection unit by data acquisition network, receives the data collected by field detection unit;The control center at least includes real-time database and and software management system, and software management system includes eight functional modules of system management module, real-time display module, prediction analysis module, alarm management module, interface module, report management, basic data management and green evaluation system etc..The application precisely monitors and traces the environmental quality of painting workshop, realizes targeted management;It can also carry out multi-dimensional analysis on monitoring data, provide decision support for environmental pollution prevention and control, and enhance the rapid response and disposal capacity of environmental emergencies.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology in ship painting workshops, and specifically to a field environmental monitoring system for ship painting workshops. Background Technology

[0002] The shipbuilding industry is a crucial component of the development of high-end equipment manufacturing. Ship painting not only consumes energy but also generates significant amounts of pollutants, primarily including air pollution, water pollution, and waste pollution. Air pollutants, such as dust and VOCs, are the most significant. Furthermore, pollution emissions from shipbuilding companies severely impact the health of their workers. For instance, welding, cutting, grinding, and sandblasting operations in shipbuilding generate substantial dust pollution, making workers long-term exposure to these tasks susceptible to pneumoconiosis, which can cause irreversible health damage.

[0003] Research on green production and emission control technologies in ship painting workshops can help shipbuilding companies improve HSE management, reduce pollution emissions, and promote green transformation and upgrading. Summary of the Invention

[0004] This invention addresses the issue of low stability and reliability in shipbuilding processes caused by discrepancies between the ship painting workshop's environment, equipment resources, and intermediate product status and the design and planning. It establishes a workshop environmental perception system in the ship painting workshop, creating a real-time monitoring and early warning system to meet safety management needs, understand the changes and distribution patterns of pollutants, improve the company's response speed and handling capabilities to accidents, facilitate relevant departments' monitoring of pollutant emissions, contribute to improving the occupational health and safety system, protect the physical and mental health of operators, and achieve green manufacturing.

[0005] Based on the above ideas, the purpose of this invention is to provide a field environmental monitoring system for ship painting workshops, which solves the problems of lack of real-time sensing system and low degree of data information integration in the prior art.

[0006] The technical solution adopted by the present invention to achieve the above objectives is as follows:

[0007] A marine painting workshop on-site environmental monitoring system, including

[0008] The on-site testing unit is installed at the integrated monitoring point in the ship painting workshop to measure the environmental parameters of the integrated monitoring point;

[0009] The control center is connected to the field testing units via a data acquisition network to receive data collected by the field testing units;

[0010] The control center includes at least a real-time database and a software management system. The software management system stores data collected by the field detection units in the real-time database, reads data from the real-time database, and manages and analyzes the data. The software management system includes eight functional modules: system management module, real-time display module, predictive analysis module, alarm management module, interface module, report management, basic data management, and green evaluation system. The system management module manages the login interface and interacts with the real-time database. The real-time display module displays the environmental quality of the ship painting workshop in real time as needed. The predictive analysis module has a built-in environmental quality prediction model for the painting workshop to predict the current concentration or future trend of volatile organic compound (VOC) concentration. The alarm management module displays alarms when environmental parameters exceed standards. The interface module links with the workshop's MES system or visualization module. The report management module manages, queries, and exports historical data. The basic data management module handles basic data input, modification, and parameter setting. The green production evaluation module retrieves data from the real-time database based on the painting workshop's working status to score the workshop's environment.

[0011] The method for establishing the environmental quality prediction model for the painting workshop is as follows:

[0012] Step 1: Parameter selection. Volatile organic compound concentration is selected as the environmental quality characterization parameter of the coating workshop, and coating process parameters and environmental parameters are selected as independent variables.

[0013] Step 2: Selection of monitoring points. Based on the layout of the painting workshop, N explosion-proof integrated monitoring points are installed in the painting room to conduct long-term environmental tracking and monitoring of the painting room, and to record the painting process parameters and environmental parameters. Each set of painting process parameters and the environmental parameters of one monitoring point constitute a sample.

[0014] Step 3: Organize the data. Based on the distribution of each explosion-proof integrated monitoring point, check whether there are any abnormalities in the data. If there are abnormal data, remove or supplement them.

[0015] Step 4: Filter the processed data for independent variables, remove independent variables with low correlation, and finally obtain K independent variables to build a prediction model;

[0016] Step 5: Based on the selected independent variables, establish a combined multiple linear regression-neural network model, as follows:

[0017] Step 5.1: Establish the following multiple linear regression model for the initial prediction:

[0018] Formula 1

[0019] in: It is a constant; Indicates the concentration of volatile organic compounds; Let m represent the m-th independent variable, where m takes the value of a positive integer from 1 to K; Let represent the partial regression coefficient of the m-th independent variable. For constant terms;

[0020] Step 5.2: Construct a composite sample. Use a multiple linear regression model to fit the data of each sample to obtain the fitted value. Use the fitted value and the corresponding sample as input, and use the predicted volatile organic compound concentration as output to reconstruct the composite sample.

[0021] Step 5.3: Establish a neural network model, which includes an input layer, intermediate layers, and an output layer. Neurons in each layer are fully connected only to their adjacent neurons. Determine the number of neurons M in the input layer, the number of intermediate layers P, the number of neurons in each layer, the number of neurons Q in the output layer, the activation function, the optimizer function, and the learning rate. Continuously learn and optimize the neural network model through composite samples to obtain a combined multiple linear regression-neural network model.

[0022] As a preferred technical solution, the field detection unit includes an explosion-proof box, a volatile organic compound gas sensor, a particulate matter sensor, a temperature and humidity sensor installed inside the explosion-proof box, an anemometer installed outside the explosion-proof box, and a data acquisition card for collecting data.

[0023] As a preferred technical solution, the explosion-proof box is also equipped with a vacuum pump, a power supply, and an air inlet pipe extending to the outside. The air inlet pipe is connected in sequence to a particulate matter sensor, a filter module, a vacuum pump, and a volatile organic compound (VOC) gas sensor. The power supply is used to power each sensor. The gas to be detected is first drawn into the particulate matter sensor by the vacuum pump for particulate matter concentration detection, then filtered by the filter module before entering the VOC gas sensor for detection, and finally discharged into the atmosphere.

[0024] As a preferred technical solution, the temperature and humidity sensor is installed at the bottom of the explosion-proof box and extends to the outside. The top of the explosion-proof box is also equipped with an alarm light that is connected to the control center signal. The outer side of the explosion-proof box is equipped with a human-machine interaction display screen that is connected to the control center.

[0025] As a preferred technical solution, the data acquisition network includes transmission cables and wireless communication.

[0026] As a preferred technical solution, the on-site detection unit is installed above the densely populated area of ​​the ship painting workshop by means of a support rod, at a height of 2.5~4m above the ground and 0.3~1m away from the wall.

[0027] As a preferred technical solution, 2-8 integrated monitoring points are installed on-site in the ship painting workshop.

[0028] As a preferred technical solution, the volatile organic compound gas sensor is a PID photoionization gas sensor, the particulate matter sensor is a laser particulate matter sensor, and the humidity sensor is a digital temperature and humidity sensor.

[0029] As a preferred technical solution, in step 3, the data is normalized after being sorted out.

[0030] As a preferred technical solution, in the process of layer-by-layer processing of the intermediate layers in the neural network model in step 5.3, the state of each neuron in each layer only affects the state of the neuron in the next layer, and the influence formula is as follows:

[0031] Formula 2

[0032] Each neuron in each layer receives information transmitted by all neurons in the previous layer. i Input information transmitted by each neuron x 1 corresponds to one input weight Its magnitude represents the strength of the transmitted signal; the weighted sum of the input signals of each neuron in the previous layer is used as the total input value of a neuron in this layer, and the total input value is compared with the threshold value in that neuron. θ For comparison, only when the sum exceeds θ Only then will the total input value be passed to the first layer as transmission information.

[0033] As a preferred technical solution, in step 5.3, the number of neurons M in the input layer is 3 to 6, the number of intermediate layers P is 1 to 2, and the number of neurons Q in the output layer is 1 to 3.

[0034] As a preferred technical solution, in step 5.3, the activation function is one of Sigmoid, Tanh, ReLU, LeakReLU, etc., the loss function is one of mean squared error loss function, cross-entropy error loss function, and exponential loss function, etc., the optimizer function is one of GradientDescent, Adadelta, Momentum, and Adam, etc., and the learning rate is between 0.001 and 0.1.

[0035] The beneficial effects of this invention are:

[0036] (1) Establishing a workshop environmental sensing system in the ship painting workshop to sense relevant parameters of the workshop environment can help to understand the changes and distribution patterns of pollutants in a timely manner, and at the same time facilitate relevant departments to supervise the emission of pollutants. This measure not only helps to achieve an occupational health and safety management system and promote the development and progress of related industries, but also helps to improve the level of environmental monitoring, ensure the safe operation of enterprises, and enhance the core competitiveness of the manufacturing industry.

[0037] (2) By conducting research on green production and emission control technologies in ship painting workshops, it is helpful for shipbuilding companies to improve HSE management, reduce pollution emissions, and promote green transformation and upgrading of shipbuilding companies.

[0038] (3) This invention, through its predictive analysis module, can predict the current concentration of volatile organic compounds (VOCs) based on coating process parameters and environmental parameters that can be monitored in real time on-site. It eliminates the need to wait for the VOCs gas sensor to output results before providing feedback, thus improving the timeliness of VOCs monitoring. Furthermore, it allows for cross-validation with subsequent measured values ​​from the VOCs gas sensor, enhancing reliability. Additionally, the predictive analysis module of this invention can predict changes in VOCs concentration based on variations in coating process parameters, thereby proactively adjusting the ventilation volume in the coating workshop. This transforms the passive adjustment in existing technologies into proactive adjustment, ensuring worker safety while achieving energy conservation and consumption reduction. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the on-site environmental monitoring system in the ship painting workshop according to an embodiment of the present invention.

[0040] Figure 2 This is a schematic diagram of the overall field detection unit in an embodiment of the present invention.

[0041] Figure 3 This is a schematic diagram showing the opening of the explosion-proof enclosure of the field testing unit in an embodiment of the present invention.

[0042] Figure 4 This is a flowchart of the field testing process of the field testing unit in an embodiment of the present invention.

[0043] Figure 5 This is a functional block diagram of the on-site environmental monitoring system for ship painting workshops in an embodiment of the present invention.

[0044] Figure 6 This is a topology diagram of the RS485 bus-based data acquisition network in an embodiment of the present invention.

[0045] Figure 7 This is a schematic diagram of the combined multiple linear regression and neural network model.

[0046] Figure 8 The process for constructing a combined multiple linear regression-neural network model.

[0047] Figure 9 The effect of painting time on VOC concentration.

[0048] Figure 10 The effect of spray painting distance on VOC concentration.

[0049] Figure 11The effect of airflow velocity on VOCs concentration.

[0050] Figure 12 The distribution of VOCs concentration measured by a toluene sensor at a horizontal distance of 1m from the nozzle is shown.

[0051] Figure 13 This is a distribution map of VOCs concentration measured by a xylene sensor at a horizontal distance of 1m from the nozzle.

[0052] 100-Control Center, 200-Field Detection Unit, 210-Explosion-proof Box, 211-Box Body, 212-Box Cover, 213-Explosion-proof Switch, 220-Volatile Organic Compound Gas Sensor, 221-Particulate Matter Sensor, 222-Temperature and Humidity Sensor, 223-Anemometer, 224-Data Acquisition Card, 225-Air Inlet, 226-Vacuum Pump, 227-Power Supply, 228-Primary Filter, 229-Secondary Filter, 230-Tertiary Filter, 231-HMI Interface, 232-Alarm Light, 300-Data Acquisition Network. Detailed Implementation

[0053] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0054] like Figures 1 to 6 As shown, the present invention provides an on-site environmental monitoring system for ship painting workshops, including...

[0055] The on-site detection unit 200 is installed at the integrated monitoring point in the ship painting workshop to measure the environmental parameters of the integrated monitoring point;

[0056] The control center 100 is connected to the field detection unit 200 through the data acquisition network 300 and receives the data collected by the field detection unit 200.

[0057] like Figure 5As shown, the control center 100 includes at least a real-time database and a software management system. The software management system reads data from the real-time database through a relational database and manages and analyzes the data. The functional modules of the software management system include eight modules: system management module, real-time display module, predictive analysis module, alarm management module, interface module, report management, green production evaluation, and basic data management. The system management module is used to manage the login interface and interact with the real-time database. The real-time display module is used to display the on-site environmental quality of the ship painting workshop in real time as needed. The predictive analysis module has a built-in environmental quality prediction model for the painting workshop to predict the current concentration or future trend of volatile organic compound concentration. The alarm management module is used to display alarms when environmental parameters exceed the standards. The interface module is used to link with the workshop MES system or visualization module. The report management module is used for historical data management, querying, and exporting. The basic data management module is used for basic data input, modification, and parameter setting. The green production evaluation module retrieves data from the real-time database to score the workshop environment based on the working status of the painting workshop.

[0058] Of course, to achieve the above functions, the control center 100 also needs hardware support, generally including at least an industrial control computer and storage devices, or a server. In this embodiment of the invention, the hardware system is a host computer, and the software management system is developed using the Microsoft Framework 4.0 platform, with the development environment being Microsoft Visual Studio 2012, the development language being C#, and the database being SQL Server. The system supports SOA system architecture and supports multiple interface methods such as database, WEB SERVICE, XML, JSON, and OPC. It has strong integration and expansion capabilities. The specific development algorithms or code can be based on existing technologies and are not the inventive points of this invention, nor do they affect the technical problems solved by this invention.

[0059] The software management system communicates with the field detection units 200 via a data acquisition network. Each field detection unit 200 collects data from various sensors via a data acquisition card, then transmits and stores the data in a real-time database through the data acquisition network. The real-time display module extracts data from the real-time database to achieve real-time monitoring of the painting workshop environment. The system management module provides periodic dumping of real-time data or storage of historical data in a relational database. The predictive analysis module can extract data from the real-time database and the relational database to achieve predictive analysis functions; other modules directly exchange data with the relational database to achieve related functions.

[0060] The predictive analysis module provides historical data query and trend analysis, and provides an entry point for using the workshop environmental quality prediction model to provide early warning of pollutant concentrations in the ship painting workshop area for the current or future period.

[0061] The interface module can exchange information with the painting workshop manufacturing execution system (MES) and the painting workshop visualization module through the JSON interface, and transmit the real-time values ​​of monitored environmental variables to the MES and visualization module. The MES transmits the current equipment operation status to the environmental monitoring system and intelligently controls the ventilation and dust removal devices inside the workshop.

[0062] The basic data management module provides functions for inputting and modifying basic data, editing calculation formulas, and setting parameters required for green production evaluation in the painting workshop. Through the built-in "Environmental Indicator Evaluation System for Ship Painting Workshops," it comprehensively considers the environmental friendliness, economic rationality, and technological advancement of the ship painting operation process, assessing and evaluating resource consumption and environmental emissions from various dimensions such as input and output in the ship painting workshop. It should be noted that the "Environmental Indicator Evaluation System for Ship Painting Workshops" is a document compiled concurrently with this invention application and is not a key document for solving the technical problems of this invention. It can also be managed by referring to existing technologies such as the "Clean Production Evaluation Indicator System for the Painting Industry."

[0063] In some embodiments, such as Figures 2 to 4 As shown, the field detection unit 200 includes an explosion-proof box 210, a volatile organic compound gas sensor 220, a particulate matter sensor 221, a temperature and humidity sensor 222 installed inside the explosion-proof box 210, an anemometer 223 installed outside the explosion-proof box 210, and a data acquisition card 224 for collecting data. The explosion-proof box 210 itself includes a box body 211 and a box cover 212. The data acquisition card 224 is connected to each sensor and anemometer via signal lines for automatic data collection from each sensor, and then the data is sent to the control center through the data acquisition network.

[0064] The explosion-proof enclosure 210 also houses a vacuum pump 226, a power supply 227, and an air inlet pipe extending to the outside. The air inlet pipe is connected sequentially to a particulate matter sensor 221, a filter module, the vacuum pump 226, and a volatile organic compound (VOC) gas sensor 220 via pipes. The power supply 227 powers each sensor, specifically the VOC gas sensor 220, the particulate matter sensor 221, the temperature and humidity sensor 222, the anemometer 223, and the data acquisition card 224. The gas to be detected is first drawn into the particulate matter sensor 221 by the vacuum pump 226 for particulate matter concentration detection, then filtered by the filter module before entering the VOC gas sensor 220 for detection, and finally discharged into the atmosphere. Figure 3As shown, the temperature and humidity sensor 222 is installed at the bottom inside the explosion-proof box 210 and extends to the outside to directly measure the temperature and humidity outside the explosion-proof box 210; the top of the explosion-proof box 210 is also equipped with an alarm light 232 that is connected to the control center 100. According to the detection and analysis of the control center 100, an alarm is issued after the alarm level is reached; the outer side of the explosion-proof box 210 is equipped with a human-machine interaction display screen, i.e., HMI interface 231, which is connected to the control center 100.

[0065] As a specific embodiment, the volatile organic compound gas sensor 220 is a PID photoionization gas sensor, the particulate matter sensor 221 is a laser particulate matter sensor 221, and the humidity sensor is a digital temperature and humidity sensor 222; the explosion-proof box 210 has an explosion-proof rating of ExdeIICT4 Gb.

[0066] In some embodiments, such as Figure 3 As shown, the air intake pipe extends outside the explosion-proof box 210 to form an air inlet 225. A primary filter 228 is installed on the air intake pipe. The filter module includes a secondary filter 229 and a tertiary filter 230; the primary filter 228 is a sintered filter, the secondary filter 229 is a mist filter, and the tertiary filter is a membrane filter. Figure 4 As shown, the primary filter 228 is followed by a pipeline ( Figure 3 (Not shown in the diagram) is connected sequentially to the PM2.5 analysis module, PM100 analysis module, fog filter (secondary filter 229), vacuum pump 226 (sampling vacuum pump), and sample flow meter ( Figure 3 The integrated monitoring point includes (not shown), membrane filter (tertiary filter 230), and PID analysis module (PID photoionization gas sensor). The on-site detection process is as follows: Under the suction of the sampling vacuum pump, the gas to be measured enters the particulate matter sensor 221 after primary filtration. The PM2.5 analysis module and PM100 analysis module respectively detect the corresponding particulate matter concentration to obtain the particulate matter concentration value. The analyzed sample gas enters the volatile organic compound gas sensor 220 after secondary and tertiary filtration. After the volatile organic compound concentration is measured, the detected gas is discharged. The temperature and humidity sensor 222 is located outside the explosion-proof box 210 211 and directly measures the temperature and humidity of the workshop environment. The anemometer 223 is located outside the explosion-proof box 210 211 and directly measures the wind speed of the workshop environment. Each sensor transmits the collected measurement factor data to the data processing unit for further analysis, processing, and calibration, and displays it in real time on the HMI interactive interface.

[0067] In some embodiments, such as Figure 6 As shown, the data acquisition network 300 uses RS485 to collect data from each field detection unit 200 and transmits it to the control center 100.

[0068] In some embodiments, the field detection unit 200 is installed above the densely populated area of ​​the ship painting workshop via support rods, at a height of 2.5-4m above the ground and 0.3-1m from the wall. To fully cover the internal environmental monitoring of the ship painting workshop, N (N ranges from 2 to 8) integrated monitoring points are arranged inside the workshop. For example, as a specific case, the integrated monitoring points are installed above the densely populated area of ​​the ship painting workshop via support rods, at a height of 3m above the ground and 0.6m from the wall. To fully cover the internal environmental monitoring of the ship painting workshop, 6 integrated monitoring points are arranged inside the workshop.

[0069] like Figure 8 As shown, this invention also provides a method for establishing a prediction model for the environmental quality of a painting workshop, the specific steps of which are as follows:

[0070] S1. Parameter selection: Volatile organic compound concentration (VOCs) is selected as the environmental quality characterization parameter of the coating workshop, and coating process parameters and environmental parameters are selected as independent variables.

[0071] S2. Selection of monitoring points: Based on the layout of the painting workshop, six explosion-proof integrated monitoring points are installed in the painting room to conduct long-term environmental tracking and monitoring of the painting room, and to record the painting process parameters and environmental parameters. Each set of painting process parameters and the environmental parameters of one monitoring point constitute a sample. The painting process parameters include at least the type of paint, spraying time and spraying distance, and the environmental parameters include particulate matter concentration, ambient temperature, ambient humidity, air velocity and air pressure.

[0072] This invention selects two integrated monitoring points in the coating process room as test objects, and the sensors used by the on-site detection unit 200 are shown in Table 1.

[0073] Table 1. Sensors Selected for Testing

[0074]

[0075] S3. Organize the data. Based on the distribution of each explosion-proof integrated monitoring point, check whether there are any abnormalities in the data. Remove or supplement abnormal data and normalize the data.

[0076] S4. The processed data is then filtered to remove variables with low correlation, resulting in K (K=3) independent variables for building the predictive model; details are as follows:

[0077] Step 1: Observe the distribution trend of volatile organic compound (VOCs) concentration with respect to spraying time, spraying distance, and air velocity. Figures 9-11 As shown;

[0078] Step 2: The model variables were selected by stepwise regression. The concentration of volatile organic compounds was selected as the dependent variable, and the painting time, painting distance and air velocity were selected as independent variables. The results are shown in the table below.

[0079] Table 2 Variable Table for Analysis Model

[0080]

[0081] Step 5: Based on the selected independent variables, establish a combined multiple linear regression-neural network model, as follows:

[0082] Step 5.1: Establish the following multiple linear regression model for the initial prediction:

[0083] Formula 1

[0084] in: It is a constant, when , ,… When both are 0, this constant represents The value of is called the intercept; Indicates the concentration of volatile organic compounds; Let m represent the m-th independent variable, where m takes the value of a positive integer from 1 to K; Let represent the partial regression coefficient of the m-th independent variable. , ,… () is the partial regression coefficient matrix; The constant term is the random error generated after removing the influence of all independent variables on the model, also known as the residual.

[0085] The correlation coefficient and t-test were performed on the fitted multiple linear regression model, as shown in Tables 3 and 4:

[0086] Table 3 Correlation coefficient test

[0087]

[0088] Table 4 t-test

[0089]

[0090] Step 5: Based on the above tests, the following model can be obtained:

[0091]

[0092] Step 5.2: Construct a composite sample. Use a multiple linear regression model to fit the data of each sample to obtain the fitted value. Use the fitted value and the corresponding sample as input, and use the predicted volatile organic compound concentration as output to reconstruct the composite sample. In this embodiment, the input samples are the fitted value of the multiple linear regression model, the painting time, the painting distance, and the air velocity.

[0093] Step 5.3: Establish a neural network model, such as... Figure 7 As shown, the neural network model includes an input layer, intermediate layers, and an output layer. The neurons in each layer are fully connected only to their adjacent neurons. The number of neurons M in the input layer, the number of intermediate layers P, the number of neurons in each layer, the number of neurons Q in the output layer, the activation function, the optimizer function, and the learning rate are determined. The neural network model is continuously learned and optimized through composite samples to obtain a combined multiple linear regression-neural network model.

[0094] In this embodiment, a BP neural network is used as the neural network model. The fitted value of the multiple linear regression model, the painting time, the painting distance, and the air velocity are selected as the input layer neurons of the combined model. The output layer neurons are the volatile organic compound concentration. There are 9 hidden layer neurons and 1 intermediate layer. The activation function of the combined model is Sigmoid, the loss function is the mean squared error loss function, the optimizer function is the Adam Optimizer function, and the learning rate is 0.01.

[0095] A comparative analysis of the multiple linear regression-neural network combined model is shown in Table 5:

[0096] Table 5 Results Analysis

[0097]

[0098] Multiple linear regression models offer good interpretability and low computational complexity, but their fitting performance is relatively weak. Nonlinear neural network models, on the other hand, exhibit strong fitting performance but are prone to overfitting and have high computational complexity. This invention combines linear and nonlinear models, using the fitting results of a multiple linear regression model along with the independent variables as input neurons in a neural network model. By following the neural network model construction steps, a combined multiple linear regression-neural network model is obtained, achieving the best fitting result.

[0099] Once the combined model of multiple linear regression and neural network is obtained, the environmental quality of the painting workshop can be predicted. The specific method is as follows:

[0100] Six monitoring points are set up in the painting room, and each monitoring point is equipped with a field detection unit 200 to monitor environmental parameters. Of course, for the multiple linear regression-neural network combined model established in the above embodiment, the only environmental parameter is air velocity. Therefore, the field detection unit 200 needs to be equipped with at least wind direction and wind speed sensors. Other sensors can be installed as needed. At least two painting process parameters, spraying time and spraying distance, are recorded. The air velocity, spraying time and spraying distance are combined to form a data collection sample. Each monitoring point is a data sample, and data collection is completed at regular intervals.

[0101] The data collected from the 6 monitoring points were examined, and abnormal data were removed or supplemented to obtain 6 samples.

[0102] The processed data from the six samples were input into a multivariate linear regression-neural network combined model. The output result is the concentration of volatile organic compounds (VOCs) that characterizes the environmental quality of the painting workshop. The average value is taken as the current average level of VOC concentration in the painting workshop.

[0103] The above embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Although the invention has been described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications, or equivalent substitutions of the technical solutions of the invention do not depart from the spirit and scope of the invention and should be covered within the scope of the claims of the invention.

Claims

1. A field environmental monitoring system for a ship painting workshop, characterized in that, include The on-site testing unit includes integrated monitoring points installed in the ship painting workshop to measure environmental parameters at the integrated monitoring points; The control center is connected to the field testing units via a data acquisition network to receive data collected by the field testing units; The control center includes at least a real-time database and a software management system. The software management system stores data collected by the field detection units in the real-time database, reads data from the real-time database, and manages and analyzes the data. The software management system comprises eight functional modules: a system management module, a real-time display module, a predictive analysis module, an alarm management module, an interface module, a report management module, a basic data management module, and a green evaluation system module. The system management module manages the login interface and interacts with the real-time database. The real-time display module displays the environmental quality of the ship painting workshop in real time as needed. The predictive analysis module has a built-in environmental quality prediction model for the painting workshop, used to predict the current concentration or future trend of volatile organic compounds (VOCs). The alarm management module displays alarms when environmental parameters exceed standards. The interface module links with the workshop's MES system or visualization module. The report management module manages, queries, and exports historical data. The basic data management module handles basic data input, modification, and parameter setting. The green evaluation system module retrieves data from the real-time database based on the painting workshop's operating status to score the workshop's environmental conditions. The method for establishing the environmental quality prediction model for the painting workshop is as follows: Step 1: Parameter selection. Volatile organic compound concentration is selected as the environmental quality characterization parameter of the coating workshop, and coating process parameters and environmental parameters are selected as independent variables. Step 2: Selection of monitoring points. Based on the layout of the painting workshop, N explosion-proof integrated monitoring points are installed in the painting room to conduct long-term environmental tracking and monitoring of the painting room, and to record the painting process parameters and environmental parameters. Each set of painting process parameters and the environmental parameters of one monitoring point constitute a sample. Step 3: Organize the data. Based on the distribution of each explosion-proof integrated monitoring point, check whether there are any abnormalities in the data. If there are abnormal data, remove or supplement them. Step 4: Filter the processed data for independent variables, remove independent variables with low correlation, and finally obtain K independent variables to build a prediction model; Step 5: Based on the selected independent variables, establish a combined multiple linear regression-neural network model, as follows: Step 5.1: Establish the following multiple linear regression model for the initial prediction: Formula 1 in: It is a constant; Indicates the concentration of volatile organic compounds; Let m represent the m-th independent variable, where m takes the value of a positive integer from 1 to K; Let represent the partial regression coefficient of the m-th independent variable. For constant terms; Step 5.2: Construct a composite sample. Use a multiple linear regression model to fit the data of each sample to obtain the fitted value. Use the fitted value and the corresponding sample as input, and use the predicted volatile organic compound concentration as output to reconstruct the composite sample. Step 5.3: Establish a neural network model, which includes an input layer, intermediate layers, and an output layer. Neurons in each layer are fully connected only to their adjacent neurons. Determine the number of neurons M in the input layer, the number of intermediate layers P, the number of neurons in each layer, the number of neurons Q in the output layer, the activation function, the optimizer function, and the learning rate. Continuously learn and optimize the neural network model through composite samples to obtain a combined multiple linear regression-neural network model. In step 5.3, during the layer-by-layer processing of the intermediate layers in the neural network model, the state of each neuron only affects the state of the neurons in the next layer, as shown in the following formula: Formula 2 Each neuron in each layer receives information transmitted by all neurons in the previous layer. i Input information transmitted by each neuron x i Corresponding to one input weight Its magnitude represents the strength of the transmitted signal; the weighted sum of the input signals of each neuron in the previous layer is used as the total input value of a neuron in this layer, and the total input value is compared with the threshold value in that neuron. θ For comparison, only when the sum exceeds θ Only then will the total input value be passed to the first layer as transmission information.

2. The on-site environmental monitoring system for ship painting workshops according to claim 1, characterized in that: The on-site detection unit includes an explosion-proof box, a volatile organic compound gas sensor, a particulate matter sensor, a temperature and humidity sensor installed inside the explosion-proof box, an anemometer installed outside the explosion-proof box, and a data acquisition card for collecting data.

3. The on-site environmental monitoring system for ship painting workshops according to claim 2, characterized in that: The explosion-proof enclosure is also equipped with a vacuum pump, a power supply, and an air inlet pipe extending to the outside. The air inlet pipe is connected in sequence to a particulate matter sensor, a filter module, a vacuum pump, and a volatile organic compound (VOC) gas sensor. The power supply is used to power each sensor. The gas to be detected is first drawn into the particulate matter sensor by the vacuum pump for particulate matter concentration detection, then filtered by the filter module before entering the VOC gas sensor for detection, and finally discharged into the atmosphere.

4. The on-site environmental monitoring system for ship painting workshops according to claim 3, characterized in that: The temperature and humidity sensor is installed at the bottom inside the explosion-proof box and extends to the outside. The top of the explosion-proof box is also equipped with an alarm light that is connected to the control center signal. The outer side of the explosion-proof box is equipped with a human-machine interaction display screen that is connected to the control center.

5. The on-site environmental monitoring system for ship painting workshops according to claim 3, characterized in that: The data acquisition network includes transmission cables and wireless communication.

6. The on-site environmental monitoring system for ship painting workshops according to claim 1, characterized in that: The on-site testing unit is installed above the densely populated area of ​​the ship painting workshop via a support rod, at a height of 2.5~4m above the ground and 0.3~1m from the wall.

7. The on-site environmental monitoring system for ship painting workshops according to claim 1, characterized in that: 2-8 integrated monitoring points are installed on-site in the ship painting workshop.

8. The on-site environmental monitoring system for ship painting workshops according to claim 3, characterized in that: The volatile organic compound gas sensor is a PID photoionization gas sensor, the particulate matter sensor is a laser particulate matter sensor, and the humidity sensor is a digital temperature and humidity sensor.

Images