A method for producing a composite water treatment agent

Abstract

The present invention relates to the technical field of information data processing. The present invention provides a production method of a composite water treatment agent. The present invention acquires and analyzes production control data of the composite water treatment agent and production history data of the composite water treatment agent, and uses a random forest model to construct a production prediction model of the composite water treatment agent. After receiving a production task of the composite water treatment agent, the present invention predicts various parameters in the production process and compares them with actual production data in real time. If the data deviation exceeds a set range, an abnormality detection is performed by constructing an equipment operation simulation model to generate early warning information. The present invention effectively improves the accuracy of production data, optimizes the production process, and realizes real-time monitoring and early warning in the production process of the composite water treatment agent.

Description

Technical Field

[0001] The invention relates to the technical field of information data processing, and in particular to a production method of a composite water treatment agent. Background Art

[0002] Composite water treatment agent is a multifunctional water treatment agent. Composite water treatment agent is composed of multiple chemical components. The composition of composite water treatment agent includes organic acid, surfactant and bactericide. The chemical components of composite water treatment agent can work synergistically to remove and treat pollutants in water. Composite water treatment agent has multiple treatment functions, such as redox reaction, flocculation precipitation and sterilization and disinfection. It can treat various metal scales and corrosion at the same time to enhance the treatment effect. Composite water treatment agent is widely used in various water treatment fields, including industrial water, agricultural water, municipal water supply, etc. In terms of industrial water, composite water treatment agent is used to treat industrial wastewater, circulating water, etc., to remove organic matter, metal ions, suspended particles and other pollutants in wastewater. In terms of agricultural water, it can be used to improve irrigation water quality, prevent soil salinization and promote plant growth. In terms of municipal water supply, composite water treatment agent can be used to treat tap water, remove pollutants such as discoloration and odor in water, and ensure the safety of citizens' drinking water.

[0003] The specific steps of data processing in the production process of composite water treatment agents include: obtaining data in the production process in real time through sensors, instruments and other equipment, including raw material input, reaction temperature, pressure, pH value, stirring speed, etc. In the data analysis stage, various statistical methods and mathematical models are used to conduct in-depth mining and analysis of the pre-processed data to monitor the stability of the production process, optimize production parameters, predict product quality, etc., and establish an efficient database system to store, back up and manage massive data.

[0004] The following technical pain points exist in data processing during the production of composite water treatment agents. Multiple types of data need to be collected during the production of composite water treatment agents. Data collection is the starting point of data processing. When data collection is inaccurate, such as sensor failure or equipment failure, it will affect the accuracy of subsequent data processing during the production of composite water treatment agents. For example, sensor or equipment failure leads to measurement value deviation, resulting in errors in the production control command of composite water treatment agents. If errors in the collected data cannot be discovered in time, the production quality of composite water treatment agents will be affected. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a method for producing a composite water treatment agent, which solves the problem that if errors in collected data cannot be discovered in time, the production quality of the composite water treatment agent will be affected.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0007] The present invention provides a method for producing a composite water treatment agent, comprising:

[0008] Step S101, obtaining composite water treatment agent production control data and composite water treatment agent production history data, using the composite water treatment agent production control data and composite water treatment agent production history data to train a random forest model, and obtaining a composite water treatment agent production prediction model;

[0009] Step S102, receiving a composite water treatment agent production task, substituting the composite water treatment agent production task into a composite water treatment agent production prediction model, and obtaining composite water treatment agent production prediction data, the composite water treatment agent production prediction data including predicted raw material ratio, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data, and predicted composite water treatment agent production equipment end operation data;

[0010] Step S103, receiving the composite water treatment agent production task and the composite water treatment agent production task error range value, sending the composite water treatment agent production task to the composite water treatment agent production equipment end, collecting the operation data generated by the composite water treatment agent production equipment end during the execution of the composite water treatment agent production task, and obtaining the composite water treatment agent production real-time data;

[0011] Step S104, comparing the real-time data of composite water treatment agent production with the predicted composite water treatment agent production equipment-side operation data to obtain the composite water treatment agent production data comparison result, the composite water treatment agent production data comparison result includes that the real-time data of composite water treatment agent production is consistent with the predicted composite water treatment agent production equipment-side operation data, and there is an error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data. If the error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data exceeds the error range between the predicted composite water treatment agent production equipment-side operation data, then entering step S105;

[0012] Step S105, for the task flow nodes in the composite water treatment agent production task, establish a node database corresponding to the task flow nodes, obtain the equipment information in the composite water treatment agent production equipment end, build an operation simulation model corresponding to each device in the composite water treatment agent production equipment end, and substitute the node database corresponding to the task flow node into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

[0013] Furthermore, in the method for producing the composite water treatment agent of the present invention, the step S101 comprises:

[0014] The composite water treatment agent production control data includes raw material data, production condition data, catalyst data, product quality data, production efficiency data, equipment operation data, equipment energy consumption data, equipment environment data and operation records. The composite water treatment agent production control data is preprocessed to obtain the preprocessed composite water treatment agent production control data;

[0015] Data categories are collected for composite water treatment agent production control data to obtain composite water treatment agent production control data categories, data are classified according to composite water treatment agent production control data categories to obtain classified composite water treatment agent production control data, features are extracted for the classified composite water treatment agent production control data to obtain features of the classified composite water treatment agent production control data.

[0016] Furthermore, in the method for producing the composite water treatment agent of the present invention, the step S101 comprises:

[0017] The composite water treatment agent production history data includes raw material use records, production condition records, catalyst use records, product quality inspection history data, production efficiency history data, equipment operation history data, environmental history data, and operator history data. The composite water treatment agent production history data is preprocessed to obtain the preprocessed composite water treatment agent production history data;

[0018] Performing data category collection on the composite water treatment agent production history data to obtain the composite water treatment agent production history data category, classifying the data according to the composite water treatment agent production history data category to obtain the classified composite water treatment agent production history data, and performing feature extraction on the classified composite water treatment agent production history data to obtain the classified composite water treatment agent production history data features;

[0019] The classified production control data features of the composite water treatment agent are used as input data to train the random forest model, and the classified production history data features of the composite water treatment agent are used as verification data to verify the random forest model, thus obtaining a production prediction model for the composite water treatment agent.

[0020] Furthermore, in the production method of the composite water treatment agent of the present invention, the step S102 comprises:

[0021] Receive a production task for a composite water treatment agent, which includes information on target products, production quantity, and production time;

[0022] Extract data features from the production task of composite water treatment agents to obtain product specification features, production batch features, and expected production time features;

[0023] Substitute the composite water treatment agent production task into the composite water treatment agent production prediction model, trigger the composite water treatment agent production prediction model to run according to the input composite water treatment agent production task, and predict the predicted raw material ratio value, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data and predicted production equipment operation data in the production process of the composite water treatment agent.

[0024] Furthermore, in the production method of the composite water treatment agent of the present invention, the step S103 comprises:

[0025] The composite water treatment agent production task is sent to the composite water treatment agent production equipment end, and the composite water treatment agent production equipment end is started. During the process of the production equipment executing the production task, the production operation data is collected in real time using sensors, data recorders or monitoring equipment to obtain the real-time production data of the composite water treatment agent;

[0026] The real-time production data of the composite water treatment agent is preprocessed, and the preprocessed real-time production data of the composite water treatment agent is used for comparison with the predicted production data of the composite water treatment agent.

[0027] Furthermore, in the method for producing the composite water treatment agent of the present invention, the step S104 comprises:

[0028] Time-align the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent, and compare each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent item by item. Each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent includes raw material ratio, production conditions, catalyst use information, product quality data, production efficiency data, and operation data of the production equipment end;

[0029] For each parameter, the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is calculated, and it is determined whether the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is within the error range value of composite water treatment agent production task.

[0030] Furthermore, in the production method of the composite water treatment agent of the present invention, the step S105 comprises:

[0031] For the task flow nodes in the composite water treatment agent production task, a node database corresponding to the task flow nodes is established, the data in the node database is sorted according to the time when the data is generated, and data labels are established for the sorted data in the node database. Cluster analysis is performed on the comparison results of the composite water treatment agent production data to obtain the real-time production data of the composite water treatment agent that exceeds the error range. According to the data attributes of the real-time production data of the composite water treatment agent that exceeds the error range and the time information of the data generation, the corresponding data is retrieved from the node database to obtain the data information to be detected, and the equipment information in the composite water treatment agent production equipment end is obtained. According to the equipment information in the composite water treatment agent production equipment end, the historical data of each device in the composite water treatment agent production equipment end is collected. Based on The neural network model is trained based on the historical data of each device in the composite water treatment agent production equipment end and the operation data of the composite water treatment agent production equipment end to obtain the operation simulation model corresponding to each device in the composite water treatment agent production equipment end. According to the equipment information in the composite water treatment agent production equipment end, a correspondence is established between the data information to be detected and the operation simulation model corresponding to each device in the composite water treatment agent production equipment end to obtain the correspondence between the operation simulation model corresponding to the equipment and the data information to be detected. The data information to be detected is substituted into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

[0032] Beneficial effects of the present invention:

[0033] The present invention constructs a random forest model to make production predictions and compares them with actual production data, which can timely discover and correct errors in the data collection process, thereby improving the accuracy of production data. Based on accurate production data and prediction models, parameters such as raw material ratios, production conditions, and catalyst use can be optimized to improve production efficiency and product quality.

[0034] During the production process, data is collected in real time and compared with the predicted data. Once abnormal data is found, the early warning mechanism is triggered immediately, so that production managers can respond quickly and take corresponding measures to avoid quality problems or production accidents. By building an equipment operation simulation model, the equipment status is monitored and predicted in real time, potential faults are discovered in advance and maintenance is carried out, the service life of the equipment is extended, and the maintenance cost is reduced.

[0035] The production forecast data and real-time data comparison and analysis provided by the present invention provide a scientific basis for production decision-making and help managers make more reasonable and efficient production plan arrangements. By optimizing production parameters and raw material ratios, resource waste is reduced, production costs are reduced, and the economic benefits of the enterprise are improved. When facing different production tasks, production parameters and processes can be quickly adjusted to meet the needs of different customers and improve the market competitiveness of the enterprise.

[0036] In summary, the present invention realizes intelligent management of the production of composite water treatment agents by introducing data prediction, real-time monitoring and early warning mechanisms, which not only improves production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0038] Figure 1 A schematic diagram of the process steps of a method for producing a composite water treatment agent provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0040] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0041] The present invention provides a method for producing a composite water treatment agent, comprising:

[0042] Step S101, obtaining composite water treatment agent production control data and composite water treatment agent production history data, using the composite water treatment agent production control data and composite water treatment agent production history data to train a random forest model, and obtaining a composite water treatment agent production prediction model;

[0043] Obtain the production control data and production history data of the composite water treatment agent. The production control data mainly includes raw material data, production condition data, catalyst data, product quality data, production efficiency data, equipment operation data, equipment energy consumption data, equipment environment data and operation records. These data are obtained in real time and are used to reflect various states and conditions in the production process. The production history data includes past raw material usage records, production condition records, catalyst usage records, product quality inspection history data, production efficiency history data, equipment operation history records, environmental history data, operator history data, etc. These data are used to provide historical experience in the production process.

[0044] The acquired production control data and production history data are preprocessed. The main purpose of preprocessing is to clean the data, remove noise and outliers, and format and normalize the data to make it more suitable for subsequent model training and analysis.

[0045] Perform feature extraction on the preprocessed data. Feature extraction is the process of extracting information useful for model training from raw data. For production control data, raw material ratio features, production condition features, catalyst usage features, etc. can be extracted; for production history data, product quality history features, production efficiency history features, etc. can be extracted. These features will be used as input data to train the random forest model.

[0046] The extracted production control data features are used as input data, and the production history data features are used as verification data to train and verify the random forest model. The random forest model is an integrated learning method that improves the accuracy of the model by building multiple decision trees and integrating their output results. During the training process, the model will learn the laws and patterns in the data and predict future production conditions based on these laws and patterns. After training and verification, a production prediction model for composite water treatment agents is finally obtained.

[0047] The production prediction model of composite water treatment agents is used to predict future production conditions, including predicting raw material ratios, predicting production conditions, predicting catalyst usage information, predicting product quality data, predicting production efficiency data, and predicting the operation data of the composite water treatment agent production equipment.

[0048] Step S102, receiving a composite water treatment agent production task, substituting the composite water treatment agent production task into a composite water treatment agent production prediction model, and obtaining composite water treatment agent production prediction data, the composite water treatment agent production prediction data including predicted raw material ratio, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data, and predicted composite water treatment agent production equipment end operation data;

[0049] In step S102, the production tasks of the composite water treatment agent are received, and the task information is substituted into the previously trained composite water treatment agent production prediction model to obtain a series of production prediction data. The specific process is as follows:

[0050] The system receives composite water treatment agent production tasks from users or upper-level systems. These tasks usually include information such as target product specifications, production quantity, and expected production time.

[0051] Extract data features from the received production tasks. For example, extract product specification features (such as product type, performance requirements, etc.), production batch features (such as the quantity required to be produced), and expected production time features from the production tasks.

[0052] Substitute the extracted production task features into the composite water treatment agent production prediction model previously trained in step S101. The production prediction model runs and outputs a series of prediction data based on the input production task features. These prediction data include but are not limited to:

[0053] Predict raw material ratio: predict the proportion and amount of various raw materials required for production.

[0054] Predict production conditions: such as reaction temperature, pressure, pH value and other parameters in the production process.

[0055] Predict catalyst usage information: predict the type, amount and timing of catalyst required.

[0056] Predict product quality data: Predict product quality indicators such as purity, activity, etc.

[0057] Predict production efficiency data: predict production volume or production efficiency per unit time.

[0058] Predict the operating data of production equipment: predict various parameters of production equipment during operation, such as motor speed, temperature, flow, etc.

[0059] Through this step, the system can provide comprehensive forecast data support for upcoming production tasks, which will help to prepare for production in advance, optimize resource allocation and improve production efficiency.

[0060] Step S103, receiving the composite water treatment agent production task and the composite water treatment agent production task error range value, sending the composite water treatment agent production task to the composite water treatment agent production equipment end, collecting the operation data generated by the composite water treatment agent production equipment end during the execution of the composite water treatment agent production task, and obtaining the composite water treatment agent production real-time data;

[0061] Receive the production task of composite water treatment agent, which includes detailed information such as product specifications and quantity to be produced. At the same time, the system also receives the error range value associated with this task, which is used to measure the acceptable degree of deviation between actual production data and predicted data, so as to ensure the stability of the production process and the controllability of product quality.

[0062] The production task is sent to the production equipment of the composite water treatment agent. The production equipment starts to prepare and execute the production task according to the received task instruction. During the execution of the production task, the system collects the operation data of the production equipment in real time through sensors, data recorders or monitoring equipment. The operation data collected from the production equipment includes production parameters such as the input amount of raw materials, reaction temperature, pressure, pH value, stirring speed, as well as the operation status and energy consumption of the equipment.

[0063] The collected real-time production data is preprocessed, including data cleaning, format conversion, outlier processing, etc., to ensure the accuracy and consistency of the data. The preprocessed data is the real-time production data of the composite water treatment agent, which reflects the actual situation during the execution of the production task.

[0064] Acquiring data during the execution of production tasks provides a basis for subsequent data comparison, analysis, and anomaly detection. At the same time, setting the error range value also helps the system to more accurately evaluate the stability of the production process and product quality, so that production activities can proceed as expected.

[0065] Step S104, comparing the real-time data of composite water treatment agent production with the predicted composite water treatment agent production equipment-side operation data to obtain the composite water treatment agent production data comparison result, the composite water treatment agent production data comparison result includes that the real-time data of composite water treatment agent production is consistent with the predicted composite water treatment agent production equipment-side operation data, and there is an error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data. If the error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data exceeds the error range between the predicted composite water treatment agent production equipment-side operation data, then entering step S105;

[0066] In step S104, the real-time production data of the composite water treatment agent is compared with the predicted data to determine whether the actual production process is consistent with the expected process. The specific operations are as follows:

[0067] Align the real-time production data with the forecast data to ensure that the data are compared within the same time period. Compare each parameter in the real-time production data with the forecast data item by item, including but not limited to raw material ratio, production conditions (such as temperature, pressure), catalyst usage information, product quality data, production efficiency data, and operation data of the production equipment.

[0068] For each parameter, the difference between the real-time production data and the predicted data is calculated. Then, it is determined whether these differences are within a preset error range. This error range is given when the production task is received in step S103, and is used to measure the acceptable degree of deviation between the actual production data and the predicted data.

[0069] According to the results of the difference judgment, a composite water treatment agent production data comparison report is generated. The comparison results include two situations: If the differences of all parameters are within the error range, it is considered that the real-time production data is consistent with the predicted data, and the production process is proceeding normally. If the difference of one or some parameters exceeds the error range, it is considered that there is an error between the real-time production data and the predicted data, and further analysis of the cause is required.

[0070] If the data comparison results show that there are errors, and these errors exceed the preset error range value, the system will automatically enter step S105 to perform more in-depth data analysis and anomaly detection to determine the cause of the error and take corresponding corrective measures.

[0071] Step S105, for the task flow nodes in the composite water treatment agent production task, establish a node database corresponding to the task flow nodes, obtain the equipment information in the composite water treatment agent production equipment end, build an operation simulation model corresponding to each device in the composite water treatment agent production equipment end, and substitute the node database corresponding to the task flow node into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

[0072] In step S105, by comparing the actual production data with the predicted data, the data anomaly is identified and the source of the anomaly is determined. The specific execution steps are as follows:

[0073] Analyze each task flow node in the composite water treatment agent production task in detail, and establish a node database corresponding to each node. These databases contain all data and parameters related to the task flow node. Obtain detailed equipment information from the composite water treatment agent production equipment, including equipment model, specifications, technical parameters, historical operation records, etc.

[0074] Based on the acquired equipment information, machine learning or simulation technology is used to build corresponding operation simulation models for each device. These models can simulate the operation status of the equipment under normal and abnormal conditions.

[0075] Substitute the data in the node database corresponding to the task flow node into the operation simulation model corresponding to the equipment to simulate the operation status of the equipment in the actual production task. Observe and analyze the output results of the equipment operation simulation model. If the model can run normally and output reasonable simulation data, it is considered that the data collected by the equipment under normal conditions should be reliable. If the model cannot run normally or the output simulation data is significantly different from the actual data, it is considered that there is a problem with the equipment and the data collected is abnormal.

[0076] For devices that are judged to have abnormal data, the system automatically generates data abnormality warning information. These warning information include the identification of abnormal equipment, the time when the abnormality occurred, the specific manifestation of the abnormality, and the cause analysis. Based on the data abnormality warning information, production managers can take timely measures, such as checking equipment status, adjusting production parameters, replacing faulty parts, etc., to eliminate data abnormalities and ensure the stability of the production process and the controllability of product quality.

[0077] By executing step S105, the present invention can effectively identify data anomalies existing in the production process of the composite water treatment agent, and issue early warnings in a timely manner, thereby providing powerful decision-making support for production managers.

[0078] Specifically, the production method of the composite water treatment agent of the present invention, step S101, comprises:

[0079] The composite water treatment agent production control data includes raw material data, production condition data, catalyst data, product quality data, production efficiency data, equipment operation data, equipment energy consumption data, equipment environment data and operation records. The composite water treatment agent production control data is preprocessed to obtain the preprocessed composite water treatment agent production control data;

[0080] Data categories are collected for composite water treatment agent production control data to obtain composite water treatment agent production control data categories, data are classified according to composite water treatment agent production control data categories to obtain classified composite water treatment agent production control data, features are extracted for the classified composite water treatment agent production control data to obtain features of the classified composite water treatment agent production control data.

[0081] The production control data of composite water treatment agents include:

[0082] Raw material data: record the type, quantity, purity and other information of raw materials used in each production.

[0083] Production condition data: such as temperature, pressure, pH value, stirring speed, etc.

[0084] Catalyst data: record the type, amount, addition time, etc. of the catalyst.

[0085] Product quality data: quality test results of each batch of products, such as purity, impurity content, etc.

[0086] Production efficiency data: such as production cycle, production capacity, etc.

[0087] Equipment operation data: working status, operating time, fault records, etc. of production equipment.

[0088] Equipment energy consumption data: energy consumption during the production process.

[0089] Equipment environment data: temperature, humidity, air cleanliness, etc. in the production workshop.

[0090] Operation records: operator's operating steps, time, parameter adjustments, etc.

[0091] Clean and organize the acquired raw data to eliminate noise, missing values ​​and outliers. Perform category collection on the preprocessed data to identify data items of different categories, such as numerical data, categorical data, time series data, etc.

[0092] Classify the data according to the data category and group the same type of data into one category to facilitate subsequent feature extraction and model training.

[0093] Feature extraction is performed on the classified data to extract valuable information for model training. Feature extraction includes the calculation of statistical features (such as mean, variance, maximum, minimum, etc.), vector representation of text data, transformation of time series data, etc. The purpose of feature extraction is to convert the original data into a format that the model can understand and process.

[0094] Specifically, the production method of the composite water treatment agent of the present invention, step S101, comprises:

[0095] The composite water treatment agent production history data includes raw material use records, production condition records, catalyst use records, product quality inspection history data, production efficiency history data, equipment operation history data, environmental history data, and operator history data. The composite water treatment agent production history data is preprocessed to obtain the preprocessed composite water treatment agent production history data;

[0096] Performing data category collection on the composite water treatment agent production history data to obtain the composite water treatment agent production history data category, classifying the data according to the composite water treatment agent production history data category to obtain the classified composite water treatment agent production history data, and performing feature extraction on the classified composite water treatment agent production history data to obtain the classified composite water treatment agent production history data features;

[0097] The classified production control data features of the composite water treatment agent are used as input data to train the random forest model, and the classified production history data features of the composite water treatment agent are used as verification data to verify the random forest model, thus obtaining a production prediction model for the composite water treatment agent.

[0098] In the production method of the composite water treatment agent, step S101 is the link of building a production prediction model, which involves data acquisition, preprocessing, feature extraction, and model training and verification. The following is a detailed description of step S101:

[0099] Obtaining the production history data of composite water treatment agents includes:

[0100] Raw material usage records: Detailed records of the type, quantity, batch, and other information of raw materials used in each production.

[0101] Production condition records: such as reaction temperature, pressure, stirring speed, reaction time and other conditions during the production process.

[0102] Catalyst usage records: including catalyst type, dosage, time of addition, etc.

[0103] Product quality inspection historical data: quality inspection report for each batch of products, including purity, activity, stability and other indicators.

[0104] Historical data on production efficiency: such as production cycle, output per unit time, etc.

[0105] Equipment operation history records: operating status, fault records, maintenance records, etc. of production equipment.

[0106] Environmental historical data: environmental parameters such as temperature, humidity, and air cleanliness in the production workshop.

[0107] Operator historical data: operator’s identity information, operating experience, training records, etc.

[0108] Clean the acquired production history data to remove invalid, missing or abnormal data. Unify and standardize the data format to ensure data consistency and comparability. Perform category analysis on the preprocessed data to identify different types of data such as numerical data, categorical data, and time series data. Classify the data according to data category to facilitate subsequent feature extraction and model processing.

[0109] Perform feature extraction on the classified data to extract features that are valuable for production prediction. Feature extraction includes statistical features (such as mean, variance, maximum, minimum, etc.), time series features (such as trend, periodicity, etc.), text features (such as word extraction, text vectorization, etc.), etc.

[0110] The random forest model is trained using the classified composite water treatment agent production control data features as input data. Random forest is an ensemble learning method that improves the prediction accuracy of the model by building multiple decision trees and integrating their prediction results. The trained random forest model is verified using the classified composite water treatment agent production history data features as validation data. The prediction performance and generalization ability of the model are evaluated by comparing the model's prediction results with the actual production data. After repeated training and verification, the optimized composite water treatment agent production prediction model is obtained. The model can predict the parameters and product quality in the production process based on the input production tasks and data features.

[0111] Specifically, the production method of the composite water treatment agent of the present invention, step S102, comprises:

[0112] Receive a production task for a composite water treatment agent, which includes information on target products, production quantity, and production time;

[0113] Extract data features from the production task of composite water treatment agents to obtain product specification features, production batch features, and expected production time features;

[0114] Substitute the composite water treatment agent production task into the composite water treatment agent production prediction model, trigger the composite water treatment agent production prediction model to run according to the input composite water treatment agent production task, and predict the predicted raw material ratio value, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data and predicted production equipment operation data in the production process of the composite water treatment agent.

[0115] Receive the production task of compound water treatment agent from the production plan or customer order. The production task should clearly include the target product (i.e. the specific type of water treatment agent), production quantity (i.e. the total amount to be produced) and production time information (such as the start and end time of production, delivery date, etc.).

[0116] Perform detailed data feature extraction on the received production tasks to identify information for production prediction.

[0117] Product specification characteristics: describe the physical and chemical properties of the target product, such as purity, active ingredient content, color, form, etc.

[0118] Production batch characteristics: indicates the size of the production scale, that is, the number of production.

[0119] Expected production time characteristics: including time-related requirements such as production cycle and delivery period.

[0120] Substitute into the production forecast model:

[0121] The extracted production task features are used as input data and substituted into the previously constructed and trained production prediction model of the composite water treatment agent.

[0122] Trigger model execution and prediction:

[0123] The production forecasting model automatically runs and makes forecasts based on the input production task characteristics.

[0124] The forecast includes:

[0125] Predict raw material ratio values: Predict the types and ratios of raw materials required based on the specifications and production quantity of the target product.

[0126] Predict production conditions: such as reaction temperature, pressure, stirring speed and other conditions that need to be controlled during the production process.

[0127] Predict catalyst usage information: including catalyst type, dosage and addition time, etc.

[0128] Predict product quality data: predict the quality indicators achieved by the products produced.

[0129] Predict production efficiency data: such as production cycle, output per unit time, etc., to evaluate production efficiency and capacity.

[0130] Predict the operating data of production equipment: predict the state parameters of production equipment during operation, such as equipment load, energy consumption, etc.

[0131] Output forecast results: Organize the output results of the production forecast model into a forecast report.

[0132] The forecast report lists various forecast data and attaches necessary explanations and descriptions so that production managers can make decisions and plan arrangements based on the forecast results. Through the above steps, step S102 completes the forecast of the production task of the composite water treatment agent, providing strong data support for subsequent production planning and execution.

[0133] Specifically, the production method of the composite water treatment agent of the present invention, step S103, comprises:

[0134] The composite water treatment agent production task is sent to the composite water treatment agent production equipment end, and the composite water treatment agent production equipment end is started. During the process of the production equipment executing the production task, the production operation data is collected in real time using sensors, data recorders or monitoring equipment to obtain the real-time production data of the composite water treatment agent;

[0135] The real-time production data of the composite water treatment agent is preprocessed, and the preprocessed real-time production data of the composite water treatment agent is used for comparison with the predicted production data of the composite water treatment agent.

[0136] In the production method of the composite water treatment agent, step S103 is a link of real-time data collection and preprocessing. The following is a detailed description of step S103:

[0137] Send production tasks to production equipment: Send the received and predicted composite water treatment agent production tasks to the actual production equipment. The production equipment starts the corresponding production process and equipment according to the received production tasks.

[0138] When production equipment is performing production tasks, real-time production operation data is collected using sensors, data recorders or monitoring equipment. The collected data covers all aspects of the production process, including but not limited to raw material input, reaction process, catalyst use, product quality testing, equipment operation status, energy consumption, etc.

[0139] The collected real-time data of composite water treatment agent production is preprocessed. Preprocessing includes data cleaning (removing invalid, erroneous or abnormal data), data format conversion (converting raw data into a format acceptable to the model), data standardization (normalizing or standardizing numerical data), etc.

[0140] The pre-processed real-time data of composite water treatment agent production will be used to compare with the composite water treatment agent production forecast data obtained in the previous step S102. The purpose of the comparison is to evaluate the consistency between the actual production situation and the forecast results, to promptly discover deviations or problems in the production process, and to provide a basis for subsequent production adjustments and optimizations.

[0141] The real-time production data of the pre-treated composite water treatment agent and the comparison results are stored in the database for subsequent data analysis and query.

[0142] Through the above steps, step S103 completes the collection and preprocessing of real-time data of composite water treatment agent production, providing a reliable data basis for subsequent data comparison, production adjustment and optimization. At the same time, the collection and comparison of real-time data also helps to timely discover problems in the production process and improve production efficiency.

[0143] Specifically, the method for producing the composite water treatment agent of the present invention, step S104, comprises:

[0144] Time-align the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent, and compare each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent item by item. Each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent includes raw material ratio, production conditions, catalyst use information, product quality data, production efficiency data, and operation data of the production equipment end;

[0145] For each parameter, the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is calculated, and it is determined whether the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is within the error range value of composite water treatment agent production task.

[0146] In the production method of the composite water treatment agent, step S104 is a link of data comparison and difference analysis. The following is a detailed description of step S104:

[0147] The real-time data of composite water treatment agent production is time-aligned with the forecast data of composite water treatment agent production. This is because the real-time data and the forecast data are collected and generated at different time points or time periods. In order to ensure the accuracy of the comparison, the two need to be aligned to the same timeline.

[0148] Compare each parameter in the real-time data and predicted data of composite water treatment agent production item by item. These parameters include but are not limited to raw material ratio, production conditions (such as temperature, pressure, stirring speed, etc.), catalyst usage information (such as type, dosage, addition time, etc.), product quality data (such as purity, activity, stability, etc.), production efficiency data (such as production cycle, unit time output, etc.) and production equipment operation data (such as equipment load, energy consumption, failure rate, etc.).

[0149] For each parameter being compared, the difference between the real-time data and the predicted data is calculated. This can be done by simple numerical subtraction or more complex statistical methods, depending on the nature of the data and the purpose of the comparison.

[0150] Next, determine whether the calculated difference is within the error range allowed by the composite water treatment agent production task. This error range should be set in advance based on production requirements, equipment accuracy, raw material stability and other factors.

[0151] If the difference is within the error range, it means that the actual production situation is basically consistent with the predicted results and the production process is in normal state.

[0152] If the difference exceeds the error range, it means that there is a significant deviation between the actual production situation and the predicted results, and it is necessary to further analyze the reasons and take corresponding adjustment measures.

[0153] Record the comparison results and difference analysis to form a detailed comparison report. The report should include the comparison results of each parameter, the difference value, and whether it is within the error range.

[0154] Specifically, the method for producing the composite water treatment agent of the present invention, step S105, comprises:

[0155] For the task flow nodes in the composite water treatment agent production task, a node database corresponding to the task flow nodes is established, the data in the node database is sorted according to the time when the data is generated, and data labels are established for the sorted data in the node database. Cluster analysis is performed on the comparison results of the composite water treatment agent production data to obtain the real-time production data of the composite water treatment agent that exceeds the error range. According to the data attributes of the real-time production data of the composite water treatment agent that exceeds the error range and the time information of the data generation, the corresponding data is retrieved from the node database to obtain the data information to be detected, and the equipment information in the composite water treatment agent production equipment end is obtained. According to the equipment information in the composite water treatment agent production equipment end, the historical data of each device in the composite water treatment agent production equipment end is collected. Based on The neural network model is trained based on the historical data of each device in the composite water treatment agent production equipment end and the operation data of the composite water treatment agent production equipment end to obtain the operation simulation model corresponding to each device in the composite water treatment agent production equipment end. According to the equipment information in the composite water treatment agent production equipment end, a correspondence is established between the data information to be detected and the operation simulation model corresponding to each device in the composite water treatment agent production equipment end to obtain the correspondence between the operation simulation model corresponding to the equipment and the data information to be detected. The data information to be detected is substituted into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

[0156] In the production method of the composite water treatment agent, step S105 is a comprehensive data analysis, model training and anomaly detection link. The following is a detailed description of step S105:

[0157] Identify each task process node in the production task of composite water treatment agent, and establish a corresponding node database for each node. The node database stores all production data related to the node, including but not limited to raw material ratio, production conditions, catalyst usage information, product quality data, production efficiency data, etc.

[0158] Sort the data in the node database according to the time when the data was generated, and create data tags for the sorted data to facilitate subsequent data retrieval and analysis. Perform cluster analysis on the comparison results of the production data of the composite water treatment agent to identify the real-time production data that exceeds the error range. Cluster analysis locates abnormal points or abnormal areas in the production process.

[0159] According to the results of cluster analysis, obtain the real-time data of composite water treatment agent production that exceeds the error range. Analyze the attributes of these abnormal data (such as raw material ratio, production conditions, etc.) and the time information when the data was generated. According to the attributes and time information of the abnormal data, retrieve the corresponding data in the node database as the data information to be tested.

[0160] Obtain equipment information from the composite water treatment agent production equipment, including equipment type, equipment number, equipment status, etc.

[0161] Collect the historical data of each device in the composite water treatment agent production equipment. Based on these historical data and the current equipment operation data, train the neural network model to obtain the operation simulation model corresponding to each device.

[0162] According to the equipment information, a corresponding relationship is established between the data information to be detected and the operation simulation model corresponding to each equipment.

[0163] Substitute the data information to be tested into the corresponding operation simulation model of the device. If the model can run normally and give reasonable output results, it means that the data collected by the device is normal. If the model cannot run or gives unreasonable results, it means that the data collected by the device is abnormal.

[0164] When data anomalies are detected, data anomaly warning information is generated. The warning information should include detailed information of the abnormal data, the time when the anomaly occurred, the cause of the anomaly, and recommended countermeasures, etc. The warning information is sent to relevant personnel in a timely manner so that they can respond quickly and handle the anomaly.

[0165] The technical solution of the present invention solves the problem that if the collected data is not found to be wrong in time, the production quality of the composite water treatment agent will be affected by the following steps:

[0166] Obtain production control data and production history data of composite water treatment agents. These data include real-time data such as raw material input, reaction temperature, pressure, pH value, stirring speed, as well as past production records, quality inspection data, etc. Preprocess the acquired data, including cleaning data, removing noise and outliers, formatting data, etc., to ensure data accuracy and consistency.

[0167] The random forest model is trained and validated using the preprocessed production control data features as input data and the production history data features as validation data. The random forest model improves the accuracy of prediction by building multiple decision trees and synthesizing their output results.

[0168] Receive the production task of composite water treatment agent, substitute the task into the trained production prediction model, and obtain the predicted raw material ratio, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data and predicted production equipment operation data.

[0169] During the execution of production tasks, the operation data of the production equipment is collected in real time to obtain the real-time production data of the composite water treatment agent. The real-time production data is compared with the predicted operation data of the production equipment to check whether there is any error.

[0170] If the error between real-time production data and predicted data exceeds the set error range, the data anomaly detection mechanism is triggered.

[0171] Establish a node database for the task flow nodes in the production task and obtain the equipment information of the production equipment. Build the operation simulation model corresponding to each device and substitute the node database into the model for verification. If the operation simulation model corresponding to the device cannot run normally, it means that the data collected by the device is abnormal, and data abnormality warning information is generated.

[0172] Through the above steps, the technical solution of the present invention can timely discover and warn errors in the data collection process, thereby avoiding production quality problems caused by data collection errors. At the same time, by comparing the prediction model with real-time data, the production process can also be monitored to further improve production efficiency.

Claims

1. A method for producing a composite water treatment agent, characterized in that: include: Step S101, obtaining composite water treatment agent production control data and composite water treatment agent production history data, using the composite water treatment agent production control data and composite water treatment agent production history data to train a random forest model, and obtaining a composite water treatment agent production prediction model; Step S102, receiving a composite water treatment agent production task, substituting the composite water treatment agent production task into a composite water treatment agent production prediction model, and obtaining composite water treatment agent production prediction data, the composite water treatment agent production prediction data including predicted raw material ratio, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data, and predicted composite water treatment agent production equipment end operation data; Step S103, receiving the composite water treatment agent production task and the composite water treatment agent production task error range value, sending the composite water treatment agent production task to the composite water treatment agent production equipment end, collecting the operation data generated by the composite water treatment agent production equipment end during the execution of the composite water treatment agent production task, and obtaining the composite water treatment agent production real-time data; Step S104, comparing the real-time data of composite water treatment agent production with the predicted composite water treatment agent production equipment-side operation data to obtain the composite water treatment agent production data comparison result, the composite water treatment agent production data comparison result includes that the real-time data of composite water treatment agent production is consistent with the predicted composite water treatment agent production equipment-side operation data, and there is an error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data. If the error between the real-time data of composite water treatment agent production and the predicted composite water treatment agent production equipment-side operation data exceeds the error range between the predicted composite water treatment agent production equipment-side operation data, then entering step S105; Step S105, for the task flow nodes in the composite water treatment agent production task, establish a node database corresponding to the task flow nodes, obtain the equipment information in the composite water treatment agent production equipment end, build an operation simulation model corresponding to each device in the composite water treatment agent production equipment end, and substitute the node database corresponding to the task flow node into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

2. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S101 includes: The composite water treatment agent production control data includes raw material data, production condition data, catalyst data, product quality data, production efficiency data, equipment operation data, equipment energy consumption data, equipment environment data and operation records. The composite water treatment agent production control data is preprocessed to obtain the preprocessed composite water treatment agent production control data; Data categories are collected for composite water treatment agent production control data to obtain composite water treatment agent production control data categories, data are classified according to composite water treatment agent production control data categories to obtain classified composite water treatment agent production control data, features are extracted for the classified composite water treatment agent production control data to obtain features of the classified composite water treatment agent production control data.

3. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S101 includes: The composite water treatment agent production history data includes raw material use records, production condition records, catalyst use records, product quality inspection history data, production efficiency history data, equipment operation history data, environmental history data, and operator history data. The composite water treatment agent production history data is preprocessed to obtain the preprocessed composite water treatment agent production history data; Performing data category collection on the composite water treatment agent production history data to obtain the composite water treatment agent production history data category, classifying the data according to the composite water treatment agent production history data category to obtain the classified composite water treatment agent production history data, and performing feature extraction on the classified composite water treatment agent production history data to obtain the classified composite water treatment agent production history data features; The classified production control data features of the composite water treatment agent are used as input data to train the random forest model, and the classified production history data features of the composite water treatment agent are used as verification data to verify the random forest model, thus obtaining a production prediction model for the composite water treatment agent.

4. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S102 includes: Receive a production task for a composite water treatment agent, which includes information on target products, production quantity, and production time; Extract data features from the production task of composite water treatment agents to obtain product specification features, production batch features, and expected production time features; Substitute the composite water treatment agent production task into the composite water treatment agent production prediction model, trigger the composite water treatment agent production prediction model to run according to the input composite water treatment agent production task, and predict the predicted raw material ratio value, predicted production conditions, predicted catalyst usage information, predicted product quality data, predicted production efficiency data and predicted production equipment operation data in the production process of the composite water treatment agent.

5. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S103 includes: The composite water treatment agent production task is sent to the composite water treatment agent production equipment end, and the composite water treatment agent production equipment end is started. During the process of the production equipment executing the production task, the production operation data is collected in real time using sensors, data recorders or monitoring equipment to obtain the real-time production data of the composite water treatment agent; The real-time production data of the composite water treatment agent is preprocessed, and the preprocessed real-time production data of the composite water treatment agent is used for comparison with the predicted production data of the composite water treatment agent.

6. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S104 includes: Time-align the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent, and compare each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent item by item. Each parameter in the real-time production data of the composite water treatment agent with the forecast production data of the composite water treatment agent includes raw material ratio, production conditions, catalyst use information, product quality data, production efficiency data, and operation data of the production equipment end; For each parameter, the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is calculated, and it is determined whether the difference between the real-time data of composite water treatment agent production and the predicted data of composite water treatment agent production is within the error range value of composite water treatment agent production task.

7. The method for producing a composite water treatment agent according to claim 1, characterized in that: The step S105 includes: For the task flow nodes in the composite water treatment agent production task, a node database corresponding to the task flow nodes is established, the data in the node database is sorted according to the time when the data is generated, and data labels are established for the sorted data in the node database. Cluster analysis is performed on the comparison results of the composite water treatment agent production data to obtain the real-time production data of the composite water treatment agent that exceeds the error range. According to the data attributes of the real-time production data of the composite water treatment agent that exceeds the error range and the time information of the data generation, the corresponding data is retrieved from the node database to obtain the data information to be detected, and the equipment information in the composite water treatment agent production equipment end is obtained. According to the equipment information in the composite water treatment agent production equipment end, the historical data of each device in the composite water treatment agent production equipment end is collected. Based on The neural network model is trained based on the historical data of each device in the composite water treatment agent production equipment end and the operation data of the composite water treatment agent production equipment end to obtain the operation simulation model corresponding to each device in the composite water treatment agent production equipment end. According to the equipment information in the composite water treatment agent production equipment end, a correspondence is established between the data information to be detected and the operation simulation model corresponding to each device in the composite water treatment agent production equipment end to obtain the correspondence between the operation simulation model corresponding to the equipment and the data information to be detected. The data information to be detected is substituted into the operation simulation model corresponding to the equipment. If the operation simulation model corresponding to the equipment is normal, the data collected by the equipment is normal. If the operation simulation model corresponding to the equipment cannot run, the data collected by the equipment is abnormal, and data abnormality warning information is generated.

Images