Different component material gas adsorption and thermodynamics intelligent sensing test system and method

Abstract

The invention discloses a gas adsorption and thermodynamic intelligent sensing test system and method for different component materials, and belongs to the technical field of material adsorption and thermodynamic experiments, the system comprises a data set module, an experiment sensing module and an adsorption-desorption and thermodynamic coupling model unit, and the method comprises the following steps: S1, material element analysis; s2, constructing an adsorption-desorption and thermodynamic coupling model; s3, acquiring field experiment parameters; s4, obtaining an experiment result; s5, obtaining a data table, a data graph and data fitting in combination with an experiment result obtained in the step S4, and exporting data; according to the invention, the adsorption and desorption and thermodynamic experiment of the multi-component material can be completed, and an idea is provided for exploring the adsorption and desorption and thermodynamic experiment rule of the multi-component material.

Description

Technical Field

[0001] The present invention belongs to the technical field of material adsorption and thermodynamic experiments, and specifically relates to a system and method for testing gas adsorption and thermodynamic intelligent perception of materials with different components. Background Art

[0002] A lot of work has been done at home and abroad on the development of gas adsorption-desorption experimental equipment, which has promoted the development of soil mechanics, rock mechanics, seepage mechanics and geology to a certain extent. There are many existing coal sample gas adsorption equipment and thermodynamic testing equipment and methods, but there are still some problems. On the one hand, the existing experimental equipment is mainly concentrated on gas adsorption experimental equipment, or only conducts adsorption-desorption experiments on one or two gases, and there is still little research and development on experimental equipment for adsorption and desorption experiments of multi-component materials. On the other hand, there are many studies on the thermodynamics of coal and coal products, but most equipment requires multiple experiments and post-calculations. The process of obtaining the thermodynamic parameters of coal samples is still relatively complicated and the degree of intelligence is low.

[0003] At present, there is no technology or method that can simultaneously perform adsorption-desorption and thermodynamic experiments on different component materials, and intelligently obtain their experimental processes and parameters. The results of adsorption-desorption and thermodynamic parameters of multi-component coal samples are of great significance for the study of the pore structure characteristics of coal bodies and their adsorption-desorption laws, and for a more in-depth analysis of the gas flow laws in gas coal. Based on this, the present invention proposes a gas adsorption and thermodynamic intelligent perception test system and method for different component materials. Summary of the invention

[0004] The present invention aims to provide a gas adsorption and thermodynamic intelligent perception testing system and method for different component materials to solve the problems existing in the prior art, so that it can complete the adsorption, desorption and thermodynamic experiments of multi-component materials, and intelligently control the experimental process, intelligently obtain the experimental results, reduce human interference in the experimental process, enhance the experimental efficiency, and more easily obtain the adsorption, desorption and thermodynamic experimental laws, reduce the experimental cost, and provide ideas for exploring the adsorption, desorption and thermodynamic experimental laws of multi-component materials.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides a gas adsorption and thermodynamic intelligent perception test system and method for different component materials, comprising the following steps:

[0007] A gas adsorption and thermodynamic intelligent perception test system for materials with different components, including a data set module, an experimental sensing module, and an adsorption-desorption and thermodynamic coupling model unit;

[0008] The data set module consists of a data collection unit, a data analysis unit, a data verification unit, and a recursive neural network unit;

[0009] The data collection unit is used to receive all data information, cluster and collect the information, and form a data map;

[0010] The data analysis unit is used to clean the data to be analyzed, and then standardize the data and process it according to the required processing method, cluster the processed data, and return it to the data storage unit;

[0011] The data verification unit is used to verify the processed data by using differential analysis and other analysis methods to ensure the accuracy of the data;

[0012] Recursive neural network, updates weight parameters through adaptive boosting algorithm;

[0013] The experimental sensing module consists of an instrument operation data acquisition unit, an adsorption and desorption data acquisition unit, a thermodynamic data acquisition unit, a data receiving unit and a general control unit;

[0014] The instrument operation data acquisition unit is used to collect the operation parameters of the instrument in the experiment and send the collected data to the data receiving unit;

[0015] The main control unit is used to make parameter determinations to ensure the normal operation of the experiment;

[0016] The adsorption and desorption data acquisition unit and the thermodynamic data acquisition unit collect the experimental data and send the data to the data receiving unit;

[0017] The data receiving unit is used to collect and send data to the data storage unit and provide parameters for the master control unit;

[0018] The adsorption, desorption and thermodynamic model unit has a preset coupling model and receives the algorithm optimized by the recursive neural network. By updating and collecting data, the system can update the weight parameters of the existing coupling model. The optimal model can be obtained through the received parameters, and the predicted experimental results of the material can be obtained, which are finally transmitted to the data storage unit.

[0019] Preferably, the coupling model includes the classical Langmuir model, the Freundlich adsorption model, the Sips adsorption model, the BET adsorption model and the classical thermodynamic parameter calculation model.

[0020] Preferably, a method for testing different component materials by a gas adsorption and thermodynamic intelligent perception testing system for different component materials comprises the following steps:

[0021] S1: Material element analysis: Determine the specific surface area, industrial analysis, fiber component and vitrinite reflection parameter, element analysis, and determine the basic physical parameters of the material;

[0022] S2: Construction of adsorption-desorption and thermodynamic coupling model: Based on the basic physical parameters of the material, a coupling model with a higher degree of fit is selected to predict the material adsorption-desorption and thermodynamic parameters;

[0023] S3: On-site experimental parameter acquisition: obtain material experimental parameters through on-site experiments;

[0024] S4: Acquisition of experimental results: Acquisition of adsorption, desorption and thermodynamic experimental results through field experimental parameters combined with coupling models;

[0025] S5: Combine the experimental results obtained in step S4 to obtain a data table, a data graph and a data fit, and export the data.

[0026] Preferably, the experiment in step S3 includes six parts in total, namely, air tightness test, blank experiment, pretreatment experiment, buoyancy experiment, adsorption experiment and desorption experiment.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] (1) The present invention proposes a gas adsorption and thermodynamic intelligent perception test system and method for different component materials, which realizes the problem of integrated experimental research on the adsorption and desorption characteristics of multi-component gases by different component materials, can ensure the safety and stability of the adsorption and desorption process, and improve the overall experimental efficiency.

[0029] (2) Compared with the traditional adsorption isotherm experiment, the present invention can predict the adsorption and desorption experimental results of the experimental components, and can complete the prediction of the thermodynamic parameters of the experimental components. At the same time, the experimental process is optimized. The experimental process adopts intelligent monitoring to reduce the manual operation in the experimental process. The experimental results are desorbed and analyzed using an optimized model to reduce experimental errors.

[0030] (3) The present invention adopts recursive neural networks, multiple data units, multiple processing methods, intelligent closed loops and other methods to construct an intelligent testing system, which reduces the tedious manual calculation links in existing experiments and has scalability, and has great application value in the fields of adsorption, desorption and thermodynamic experiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of a gas adsorption and thermodynamic intelligent perception testing system and method for different component materials of the present invention;

[0032] Figure 2A schematic diagram of a data receiving unit of a gas adsorption and thermodynamic intelligent perception testing system and method for different component materials of the present invention; DETAILED DESCRIPTION

[0033] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments:

[0034] like Figure 1-2 As shown, a gas adsorption and thermodynamic intelligent perception test system for materials with different components includes a data set module, an experimental sensing module, and an adsorption-desorption and thermodynamic coupling model unit;

[0035] The data set module consists of a data collection unit, a data analysis unit, a data verification unit, and a recursive neural network unit. It selects the optimal model through experience through input parameters, and obtains the predicted experimental results of the material.

[0036] The data collection unit is used to receive all data information, cluster and collect the information, and form a data map;

[0037] The data analysis unit is used to clean the data to be analyzed, and then standardize the data and process it according to the required processing method, cluster the processed data, and return it to the data storage unit;

[0038] The data verification unit is used to verify the processed data by using differential analysis and other analysis methods to ensure the accuracy of the data;

[0039] Recursive neural network, updates weight parameters through adaptive boosting algorithm;

[0040] The experimental sensing module consists of an instrument operation data acquisition unit, an adsorption and desorption data acquisition unit, a thermodynamic data acquisition unit, a data receiving unit and a general control unit;

[0041] The instrument operation data acquisition unit is used to collect the operation parameters of the instrument in the experiment and send the collected data to the data receiving unit;

[0042] The master control unit is used to determine parameters, ensure the normal operation of the experiment, and further reduce external interference in the experimental process;

[0043] The adsorption and desorption data acquisition unit and the thermodynamic data acquisition unit collect the experimental data and send the data to the data receiving unit;

[0044] The data receiving unit is used to collect and send data to the data storage unit and provide parameters for the master control unit;

[0045] The adsorption, desorption and thermodynamic model unit has a preset coupling model and receives the algorithm optimized by the recursive neural network. Through the update and collection of data, the system can update the weight parameters of the existing coupling model, obtain the optimal model through the received parameters, and obtain the predicted experimental results of the material, and finally transmit them to the data collection unit. The coupling model includes the classic Langmuir model, Freundlich adsorption model, Sips adsorption model, BET adsorption model and the classic thermodynamic parameter calculation model. To further improve the accuracy of the model, the data can be used to calibrate the model through the neural network model set in the data set module, and adjust the model parameter structure to gradually obtain more model optimization solutions.

[0046] A method for testing materials of different components by a gas adsorption and thermodynamic intelligent perception testing system for materials of different components, comprising the following steps:

[0047] S1: Material element analysis: Determine the specific surface area, industrial analysis, fiber component and vitrinite reflection parameter, element analysis, and determine the basic physical parameters of the material;

[0048] Material element analysis is selected according to the experimental materials. The selection principle is to select parameters that can be used to characterize the experimental materials, with basic physical parameters being given priority.

[0049] S2: Construction of adsorption-desorption and thermodynamic coupling model: Based on the basic physical parameters of the material, a coupling model with a higher degree of fit is selected to predict the material adsorption-desorption and thermodynamic parameters;

[0050] In order to adapt to the characteristics of different materials and improve the accuracy of the model, the data receiving unit is preset with a coupling model containing multiple adsorption models and thermodynamic parameter calculation models. The data collected in the public literature are preset, and the data obtained in the experiment are included. The data can be used to calibrate the model through the neural network set in the data receiving unit, adjust the model parameter structure, and gradually obtain more model optimization solutions. The parameters input in step S1 are empirically selected to obtain the optimal model and the predicted experimental results of the material.

[0051] S3: On-site experimental parameter acquisition: obtain material experimental parameters through on-site experiments;

[0052] S4: Acquisition of experimental results: Acquisition of adsorption, desorption and thermodynamic experimental results through field experimental parameters combined with coupling models;

[0053] S5: Combine the experimental results obtained in step S4 to obtain a data table, a data graph and a data fit, and export the data.

[0054] The experiment in step S3 includes six parts, namely, air tightness test, blank experiment, pretreatment experiment, buoyancy experiment, adsorption experiment and desorption experiment.

[0055] The specific implementation process is as follows:

[0056] Embodiment 1:

[0057] A gas adsorption and thermodynamic intelligent perception test system and method for materials with different components, including adsorption, desorption and thermodynamic model construction, wherein a data receiving unit is provided with a data storage unit, a data analysis unit and a data verification unit, the data storage collection unit collects all data information, clusters and collects the information, and forms a data map, the data analysis unit first cleans the data to be analyzed, and then standardizes the data and processes it according to the required processing method, and finally clusters the processed data and returns it to the data storage unit, the data verification unit uses difference analysis on the processed data to verify the data to ensure the accuracy of the data, inputs initial data, inputs initial model, updates literature data and experimental data, and the recursive neural network set in the model adaptive learning data receiving unit can update weight parameters through an adaptive boosting algorithm, through the update and collection of information data, the system can update the weight parameters of the existing model, and obtain the model through empirical selection of the parameters input in step S1, and obtain the predicted experimental results of the material.

[0058] First, the initial data is the experimental data and public literature data available when the method is established, and the data is input as the initial data. The initial model is the common formula available when the method is established, most of which are publicly recognized common formulas, such as the Langmuir model, Freundlich adsorption model, Sips adsorption model, BET adsorption model, classical thermodynamic parameter calculation model, etc. In order to further improve the accuracy of the coupling model in S4, the coupling model is equipped with a data receiving unit, and the data obtained from the public literature is preset in the data collection, and the data obtained from the experiment is included. The data can be used to calibrate the model through the neural network model set in the data receiving unit, adjust the model parameter structure, and gradually obtain more model optimization solutions.

[0059] Second, update the data. The updated data are the data of each experiment and the data of public literature. The experimental data are the data obtained by the system during the experiment, and the public literature data need to be manually screened and input. The updated data will be incorporated into the data collection and participate in the model adaptive learning together with the initial data.

[0060] Third, model adaptive learning, based on the update of the database, the recursive neural network set by the method is trained based on a large amount of data to set up a training model, automatically extract features from the data through learning, and adjust the weights during the training process to improve performance. The adaptive boosting algorithm set by the method is an integrated model. The algorithm will determine how much the next weight will change based on the error size of the last learner and determine the weight of the next sample. In the first iteration, the weights of all samples are the same. In subsequent iterations, the weights of learners with larger errors will be gradually increased, while the weights of learners with smaller errors will be reduced. The weights are assigned to the lower-level learners for the next round of training. After the iteration is completed, the median of the weighted weak learners is taken to finally obtain a strong regressor.

[0061] Fourth, in the prediction of experimental materials and calculation of experimental results, the coupling model will select the model based on the basic data obtained from elemental material analysis and use R 2 , RMSE and MAE are used as evaluation indicators. 2 The larger the value, the better the model prediction performance. The smaller the RMSE and MAE, the smaller the model error, which means the higher the accuracy.

[0062] The specific calculation formula is as follows:

[0063]

[0064] Embodiment 2:

[0065] A coal gas adsorption and thermodynamic intelligent perception test comprises: coal material element analysis, coal material adsorption, desorption and thermodynamic coupling model construction, wherein the basic physical parameters of the material, a coupling model with a higher fitting degree is selected to predict the material adsorption, desorption and thermodynamic parameters, and the experimental results are obtained through the coal material adsorption and desorption experiment. In order to reduce the interference of external factors in the experimental process, a data set module is provided in the system, and the data set module comprises an instrument operation data acquisition unit, an adsorption and desorption data acquisition unit, a thermodynamic data acquisition unit, a data receiving unit and a general control unit. The instrument operation data acquisition unit collects the operation parameters of the instrument in the experiment, and sends the collected data to the data receiving unit. The general control unit performs parameter judgment to ensure the normal operation of the experiment. The adsorption and desorption data acquisition unit and the thermodynamic data acquisition unit collect the experimental data, and send the collected data to the data receiving unit, which is recorded in the data receiving unit in step S2 for data cleaning and analysis steps.

[0066] First: Conduct specific surface area measurement, industrial analysis, fiber component and vitrinite reflection parameter measurement, determine the molar ratio of H, O, N and S elements to C element in coal, conduct industrial analysis of coal, fiber component and vitrinite reflection parameter measurement, and determine the basic physical parameters of the material.

[0067] Second: Input the obtained parameters into the system. If only predicted parameters are needed, the system can directly select an appropriate model to calculate and output the predicted parameters. Taking coal as an example, based on the molecular formula obtained by coal element determination, the chemical equation obtained from the complete combustion reaction obtained from the coal molecular formula is selected to predict the thermodynamic parameters, and the adsorption and desorption parameters are calculated based on the adsorption and desorption model corresponding to the coal pore structure.

[0068] Assuming the element ratios are a1, a2, a3, and a4, the molecular formula of coal is CH a1 O a2 N a3 S a4 , the combustion reaction can be constructed as:

[0069] CH a1 O a2 N a3 S a4 +O a4 =CO2+a1 / 2H2O+a3 / 2N2+a4SO2

[0070] Thermodynamic parameters are calculated according to the combustion reaction equation, including combustion reaction heat, standard molar formation enthalpy, standard molar formation entropy, standard molar formation Gibbs free energy, etc.

[0071] Since the adsorption form of coal gas is the coexistence of surface monolayer adsorption and micropore filling, the Langmuir equation and DA equation can be used together to characterize the adsorption and desorption laws of coalbed gas.

[0072] The pore size and its distribution affect the proportion of pore volume and pore specific surface area. Therefore, the system will correct the impact of different pore size distribution ratios on the isothermal adsorption model and optimize the gas density fitting of the DA model.

[0073] The system will use the optimized and corrected model for prediction calculations, and the data verification unit will perform fitting calculations to obtain the degree of fit between the prediction results and the coupling model.

[0074] Third: Conduct field experiments, and conduct experiments on experimental materials in accordance with national standards. The experimental steps include six processes: airtightness test, blank experiment, pretreatment experiment, buoyancy experiment, adsorption experiment, desorption experiment, etc. The experimental results are collected by the adsorption and desorption data acquisition unit and the thermodynamic data acquisition unit, and input into the data storage unit.

[0075] The data collection unit transmits the experimental data to the data analysis unit, and at the same time the data is screened and cleaned, and then the corresponding model selection and weight analysis are performed to obtain a new weight model. The data verification unit uses the new weight model to fit and verify the data.

[0076] The analysis results and the new weight model are simultaneously entered into the adsorption-desorption and thermodynamic model units as one of the bases for future calculations.

[0077] Fourth: perform data output. In step S4, multiple data export modes are built in to obtain data table mode, data graph mode and data fitting mode. The required mode is selected for data export, and the verified data is exported. Different modes can be selected for data export, such as data table, data graph, data fitting and other different modes can be selected. The data can be selected in various different modes such as original experimental results, experimental analysis results and so on.

[0078] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A gas adsorption and thermodynamic intelligent perception test system for materials with different components, characterized by: It includes data set module, experimental sensing module and adsorption-desorption and thermodynamic coupling model unit; The data set module consists of a data collection unit, a data analysis unit, a data verification unit, and a recursive neural network unit; The data collection unit is used to receive all data information, cluster and collect the information, and form a data map; The data analysis unit is used to clean the data to be analyzed, and then standardize the data and process it according to the required processing method, cluster the processed data, and return it to the data storage unit; The data verification unit is used to verify the processed data by using differential analysis and other analysis methods to ensure the accuracy of the data; Recursive neural network, which updates weight parameters through adaptive boosting algorithm; The experimental sensing module consists of an instrument operation data acquisition unit, an adsorption and desorption data acquisition unit, a thermodynamic data acquisition unit, a data receiving unit and a general control unit; The instrument operation data acquisition unit is used to collect the operation parameters of the instrument in the experiment and send the collected data to the data receiving unit; The main control unit is used to make parameter determinations to ensure the normal operation of the experiment; The adsorption-desorption data acquisition unit and the thermodynamic data acquisition unit collect the experimental data and send the data to the data receiving unit; The data receiving unit is used to collect and send data to the data storage unit and provide parameters for the master control unit; The adsorption, desorption and thermodynamic model unit has a preset coupling model and receives the algorithm optimized by the recursive neural network. By updating and collecting data, the system can update the weight parameters of the existing coupling model. The optimal model can be obtained through the received parameters, and the predicted experimental results of the material can be obtained, which are finally transmitted to the data storage unit.

2. According to claim 1, a gas adsorption and thermodynamic intelligent perception test system for materials with different components, characterized in that: The coupling model includes the classical Langmuir model, the Freundlich adsorption model, the Sips adsorption model, the BET adsorption model and the classical thermodynamic parameter calculation model.

3. The method for testing different component materials by a gas adsorption and thermodynamic intelligent perception testing system for different component materials according to claim 1, characterized in that: The following steps are involved: S1: Material element analysis: Determine the specific surface area, industrial analysis, fiber component and vitrinite reflection parameter, element analysis, and determine the basic physical parameters of the material; S2: Construction of adsorption-desorption and thermodynamic coupling model: Based on the basic physical parameters of the material, a coupling model with a higher degree of fit is selected to predict the material adsorption-desorption and thermodynamic parameters; S3: On-site experimental parameter acquisition: obtain material experimental parameters through on-site experiments; S4: Acquisition of experimental results: Acquisition of adsorption, desorption and thermodynamic experimental results through field experimental parameters combined with coupling models; S5: Combine the experimental results obtained in step S4 to obtain a data table, a data graph and a data fit, and export the data.

4. The method for testing different component materials by a gas adsorption and thermodynamic intelligent perception testing system for different component materials according to claim 3 is characterized in that: The experiment in step S3 includes six parts in total, namely, air tightness test, blank experiment, pretreatment experiment, buoyancy experiment, adsorption experiment and desorption experiment.

Images