Automatic water-rock interaction experimental data acquisition system in geothermal recharge process

Abstract

The invention provides a system for automatically acquiring experimental data of water-rock interaction in a geothermal recharge process, which is characterized in that an indoor simulation experiment of the water-rock interaction in the geothermal recharge process is carried out through a constant-temperature water bath oscillator, and in the experiment process, the ion concentration, TDS (total dissolved solids) and pH (potential of hydrogen) of a solution are measured through an ion concentration sensor, a TDS sensor and a pH sensor respectively; an ion concentration sensor, a TDS sensor and a pH sensor are connected to an ESP8266 through a serial communication protocol, data are uploaded to a cloud end through a built-in Wi-Fi module of the ESP8266, and real-time subscription is achieved through the cloud end constructed through the. NET technology. The problem that monitoring of water-rock interaction experimental data in a traditional coal mine goaf is not real-time or accurate enough is solved. Belongs to the geothermal energy exploitation field.

Description

Technical Field

[0001] The invention relates to an automatic data acquisition system for water-rock interaction experiments in a geothermal recharge process, and belongs to the field of geothermal energy exploitation. Background Art

[0002] Geothermal recharge in abandoned coal mine goafs is a promising method for geothermal energy exploration, but it can reduce the efficiency of heat extraction due to corrosion or scaling of pipes caused by the interaction of water and rock. During the geothermal recharge process, when the reinjected water comes into contact with the surrounding rock, a strong chemical imbalance occurs, causing minerals to dissolve and precipitate. These mineral precipitates (such as silica) can clog pipes, geothermal wells and heat exchangers, thereby reducing the efficiency of geothermal systems. The dissolution of minerals can cause an imbalance in the concentration of ions in the solution. The chemical composition of mine water can also reduce the efficiency of geothermal systems because certain minerals in the water can easily precipitate in pipes, causing scaling and clogging, which in turn affects the efficiency of geothermal energy extraction.

[0003] At present, most of the research on water-rock interaction is focused on conventional groundwater and artificial recharge areas. The monitoring of water-rock interaction experimental data in coal mine goaf is not real-time and accurate enough. Therefore, the present invention provides an automatic data collection system for water-rock interaction experiment in geothermal recharge process. Summary of the invention

[0004] The present invention provides an automatic data collection system for water-rock interaction experiments in a geothermal recharge process, so as to solve the problems that the monitoring of water-rock interaction experiments in traditional coal mine goafs is not yet real-time and accurate enough.

[0005] In order to solve the above problems, an automatic data acquisition system for water-rock interaction experiments during geothermal recharge is proposed. An indoor simulation experiment of water-rock interaction during geothermal recharge is carried out through a constant temperature water bath oscillator. During the experiment, the ion concentration sensor, TDS sensor and pH sensor are used to measure the solution ion concentration, TDS and pH respectively. The ion concentration sensor, TDS sensor and pH sensor are connected to ESP8266 using a serial communication protocol. The data is uploaded to the cloud through the built-in Wi-Fi module of ESP8266, and real-time subscription is realized through the cloud built with .NET technology.

[0006] The aforementioned experimental data automatic collection system also includes cloud-based data processing and real-time subscription: Data processing: Using data analysis algorithms in the cloud to identify patterns and trends in the water-rock interaction process; Real-time subscription: Optimizing the real-time subscription function implemented by .NET technology, using more efficient data stream processing frameworks, such as Apache Kafka or RabbitMQ, to improve the scalability and responsiveness of the system.

[0007] The above-mentioned cloud data processing specifically includes:

[0008] 1) Data preprocessing and feature engineering

[0009] Use Python's Pandas library to clean the data, including filling missing values, deleting duplicate values, and handling outliers; non-dimensionalize the data to facilitate model processing; and perform feature selection;

[0010] 2) Model training and verification

[0011] Select a suitable model based on data characteristics; divide the data set into a training set and a test set; train the model using the training set data, and adjust the model parameters through cross-validation to optimize performance;

[0012] 3) Model deployment and real-time data processing

[0013] Use cloud services to deploy models; use Apache Kafka or Amazon Kinesis to collect and process real-time data streams, and then use PySpark for data analysis and model prediction;

[0014] 4) Upload data to the cloud

[0015] Use cloud storage services to upload data; choose appropriate cloud storage services to store and back up data, and ensure data security and accessibility;

[0016] 5) Data visualization and monitoring

[0017] Use Tableau, Power BI, or D3.js to create real-time data visualization dashboards to display dynamic changes in data and model prediction results; monitor the performance indicators of the model in the production environment, including prediction accuracy, latency, and throughput, to ensure that the model continues to work;

[0018] 6) Model optimization and iteration

[0019] Evaluate model performance using confusion matrix, precision, and recall, and optimize the model based on feedback; regularly update the model to maintain its accuracy and relevance based on business needs and data changes.

[0020] In the aforementioned automatic experimental data collection system, the experimental samples were collected from the abandoned coal mine goaf, which were mine water and mudstone. The mudstone samples were mechanically crushed, and the crushed mudstone samples were placed on a 200-mesh sieve for screening;

[0021] In the above-mentioned automatic experimental data acquisition system, the parameters during the experiment, the settings of time, temperature, and speed restore the corresponding parameters of water-rock interaction during the reinjection process of abandoned coal mine goaf. The experimental time period is 1 day, 5 days, 10 days, 15 days, 20 days, and 25 days; the experimental temperature is 30℃, 50℃, and 70℃; the experimental oscillation speed is 100r / min.

[0022] In the aforementioned automatic experimental data acquisition system, the ion concentration sensor is used to measure the concentration of specific ions in the solution, including sodium ions, potassium ions, calcium ions, magnesium ions, bicarbonate ions, chloride ions, and sulfate ions, and the concentration unit is mg / L;

[0023] In the aforementioned automatic experimental data collection system, the TDS sensor is used to measure the amount of dissolved solids in the solution, including salts, minerals and organic matter. The working principle is to use conductivity measurement technology to calculate the TDS value by measuring the ability of electrolytes to propagate current in water;

[0024] In the aforementioned experimental data automatic acquisition system, the serial communication protocol is I 2 C, has the function of supporting multiple sensors to be used simultaneously, I 2 C can be used to connect a variety of sensors, allowing a single-chip microcomputer to communicate with multiple sensors via a simple two-wire connection.

[0025] In the aforementioned automatic experimental data collection system, the cloud built with .NET technology implements the real-time subscription function, which can obtain the changing trend and real-time status of the experimental data. The process includes the following steps:

[0026] Create a new ASP.NET Core Web API project using the .NET CLI; create a SensorData model to store sensor data; create a MongoDB context to handle database operations; create a controller to handle the upload and real-time features of sensor data;

[0027] Use SignalR to add real-time functions (install SignalR package; create SignalR Hub; configure Startup; trigger events when sensor data is uploaded);

[0028] Deploy .NET applications to the cloud and configure environment variables, database connections and other settings on the cloud service to ensure normal operation.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] Real-time: The automatic data acquisition system for water-rock interaction experiments during geothermal recharge can monitor the ion concentration, TDS, pH and other parameters in the water-rock interaction process in real time, provide timely experimental data, and help to make quick responses and adjust experimental parameters.

[0031] Accuracy and reliability: The automatic data collection system can reduce errors caused by human operation and improve the accuracy and consistency of data collection. In addition, the automatic data collection system can continuously monitor and record data without human intervention to ensure the integrity and reliability of the data.

[0032] Improve experimental efficiency: The automatic acquisition system can save labor costs and time, improve experimental data, and the research team can invest more time in data analysis and model building. At the same time, multiple sensors can work simultaneously to achieve synchronous monitoring of multiple parameters, which is convenient for data correlation analysis.

[0033] Long-term monitoring and data recording: The system is capable of long-term operation and continuous data collection, providing a rich data basis for subsequent long-term research and helping to understand the dynamic characteristics of water-rock interaction over time.

[0034] Cloud data processing and analysis: With powerful data processing capabilities, the cloud uses data analysis algorithms to identify patterns and trends in the water-rock interaction process, improving the depth and breadth of data processing; the real-time subscription function implemented through .NET technology improves the system's response speed and scalability; the use of the Pandas library for data cleaning and feature extraction improves data quality and provides accurate input for model training; the model parameters are adjusted through cross-validation to optimize model performance and improve prediction accuracy; the use of cloud services for model deployment improves the scalability and reliability of the system; the use of tools such as Apache Kafka or Amazon Kinesis to collect and process real-time data streams improves the real-time nature of data processing; the use of tools such as Tableau, Power BI or D3.js to create real-time data visualization dashboards improves the intuitiveness and comprehensibility of data; the performance indicators of the model in the production environment are monitored to ensure the continued effectiveness and stability of the model; the confusion matrix, precision, recall and other indicators are used to evaluate the model performance, and the model is optimized based on feedback to improve the adaptability and accuracy of the model.

[0035] Cloud data storage: Experimental data can be stored in the cloud, facilitating sharing and collaboration among team members.

[0036] In summary, this technical solution provides a comprehensive, efficient and real-time automatic data collection and analysis system for water-rock interaction experiments in the geothermal reinjection process by integrating multiple sensors, automated data collection, cloud data processing and analysis, real-time data stream processing, data visualization and monitoring technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a simple framework diagram of the system. DETAILED DESCRIPTION

[0038] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0039] Example

[0040] This paper provides an indoor simulation experiment scheme for water-rock interaction during reinjection:

[0041] ①. Mix 20g of crushed mudstone and 250ml of mine water and put them into a 300ml conical flask;

[0042] The mudstone was mechanically crushed by an ore crusher, and then the crushed mudstone sample was put into a 200-mesh stainless steel sieve for manual screening to obtain crushed mudstone.

[0043] ②. Place the conical flask in the water bath of a constant temperature water bath oscillator to conduct a rock interaction simulation experiment;

[0044] Among them, the speed of the constant temperature water bath oscillator was set to 100r / min; the temperatures were set to 30°C, 50°C, and 70°C respectively; and the experimental time periods were 1 day, 5 days, 10 days, 15 days, 20 days, and 25 days respectively.

[0045] Among them, since the experimental cycles are 1 day, 5 days, 10 days, 15 days, 20 days, and 25 days, 6 conical flasks are required in a set of temperature experiments.

[0046] ③. After the above steps, the indoor simulation experiment of water-rock interaction during the reinjection process can be completed.

[0047] Automatic data collection part

[0048] This implementation provides a technology for automatically collecting experimental data on water-rock interaction during reinjection:

[0049] ①. Place the ion concentration sensor, TDS sensor, and pH sensor into the experimental solution to measure the sodium ion, potassium ion, calcium ion, magnesium ion, bicarbonate ion, chloride ion, sulfate ion, TDS, and pH value of the experimental solution, ensuring that there is appropriate space between the sensors to avoid mutual interference;

[0050] Among them, the sensing parts of the ion concentration sensor, TDS sensor, and pH sensor need to be completely immersed in the experimental solution, but avoid the power supply or connecting parts of the sensor from contacting the liquid to prevent damage.

[0051] Among them, according to the experimental cycle (1 day, 5 days, 10 days, 15 days, 20 days, 25 days), the time intervals for sensor data collection are set to 1 day, 5 days, 10 days, 15 days, 20 days, and 25 days.

[0052] ②. Use I 2 C. Connect the ion concentration sensor, TDS sensor, and pH sensor to the analog pins of ESP8266;

[0053] Wherein, the ion concentration sensor, TDS sensor and pH sensor are used at the same time, and the reading method is through frequency division or alternation.

[0054] ESP8266 integrates Wi-Fi module, processor and other necessary peripherals, simplifies the design and can easily connect to wireless network; ESP8266 supports multiple communication interfaces, including UART, SPI, I 2 C, etc., can transmit data to various external sensors; ESP8266 is designed for low-power applications, supports multiple energy-saving modes, and is suitable for battery-powered devices.

[0055] ③. Use Google Cloud platform to upload the experimental data collected by ESP8266 to the cloud;

[0056] ④. Use API.NET Core combined with SignalR to implement real-time data subscription function in the cloud. The following are the steps.

[0057] First, create an ASP.NET Core project. Create a new ASP.NET Core application in Visual Studio, select Web Application (Model-View-Controller), and then install SighnalR using the NuGet package manager.

[0058] Next, set up the SignalR Hub. Create a new class in the project named MonitoringHub.cs, which is used to manage connections and messages. Then configure Http and WebSocket support in Startup.cs.

[0059] Then, the front-end is implemented. Reference the SignalR client library in the HTML file.

[0060] Finally, deployment and monitoring: Deploy the ASP.NET Core application to the cloud service; use application monitoring tools to monitor the application's performance and real-time experimental data stream.

[0061] The automatic data collection technology also includes cloud data processing:

[0062] 1) Data preprocessing and feature engineering

[0063] Data cleaning: Use Python's Pandas library to clean the data, including filling missing values ​​with the `fillna()` function, removing duplicate values ​​with `drop_duplicates()`, and handling outliers with `replace()` or `clip()`;

[0064] Specifically, it includes: using Pandas's `isnull()` or `isna()` functions to identify missing values ​​in the data, and using the `fillna()` function to fill missing values ​​by forward filling (`ffill`), backward filling (`bfill`), or using the mean (`mean()`), median (`median()`), etc.; using the `describe()` function to view the statistical description of the data and identify possible outliers, using the `clip()` function to limit the range of the data, or using the `zscore` or `iqr` (interquartile range) method to identify and handle outliers; using the `duplicated()` function to identify duplicate rows in the data; using the `drop_duplicates()` function to delete duplicate values, you can keep the first or last occurrence of the record as needed.

[0065] Feature extraction: Use `MinMaxScaler` or `StandardScaler` in the Scikit-learn library to non-dimensionalize the data for model processing; specifically: use Scikit-learn's `MinMaxScaler` to scale the features to the [0,1] interval, which is suitable for range-sensitive algorithms; use `StandardScaler` to standardize the features to have a mean of 0 and a standard deviation of 1, which is suitable for distance-sensitive algorithms; for each feature, use the `fit_transform()` method to scale it and save the result as a new dataset.

[0066] Feature Selection: Use Scikit-learn’s `SelectKBest` or `RFE` (Recursive Feature Elimination) for feature selection to reduce feature dimensionality and improve model performance.

[0067] Feature selection using SelectKBest: Use `SelectKBest` in combination with different scoring functions (such as `chi2`, `f_classif`, `mutual_info_classif`, etc.) to select the most important K features; determine the number of features to select by setting the `k` parameter; use the `fit_transform()` method to select features and apply them to the training set and test set.

[0068] Use Recursive Feature Elimination (RFE): Use `RFE` to select features by recursively reducing the feature set, removing the feature with the smallest weight in each iteration until the specified number of features is reached; choose a base model (such as `RandomForest`, `SVM`, etc.) and set the number of features to retain; train the model using the `fit()` method and use the `transform()` method to select features.

[0069] 2) Model training and verification

[0070] Model selection: Choose an appropriate model based on the characteristics of the data. For classification problems, use the random forest `RandomForestClassifier` in the Scikit-learn library, the support vector machine `SVC`, or the neural network model in the deep learning framework such as TensorFlow or PyTorch.

[0071] Training and test set division: Use Scikit-learn's `train_test_split` function to split the dataset into training and test sets, usually in a ratio of 80:20.

[0072] Model training: Use the training set data to train the model and adjust the model parameters to optimize performance through methods such as cross-validation. Use Scikit-learn's `GridSearchCV` or `RandomizedSearchCV` for parameter tuning.

[0073] 3) Model deployment and real-time data processing

[0074] Cloud deployment: Use cloud services such as AWS SageMaker, Google AIPlatform, or Azure Machine Learning to deploy and automate model management. These platforms provide a full range of solutions for model deployment, monitoring, and expansion.

[0075] Real-time data stream processing: Use tools such as Apache Kafka or Amazon Kinesis to collect and process real-time data streams, and then use PySpark for data analysis and model prediction.

[0076] 4) Upload data to the cloud

[0077] Data transfer: Upload data using cloud storage services such as AWS S3, Google Cloud Storage, or Azure Blob Storage. You can write scripts to use their APIs to perform batch uploads of data.

[0078] Data storage: Choose a suitable cloud storage service to store and back up data and ensure data security and accessibility.

[0079] 5) Data visualization and monitoring

[0080] Data visualization: Use tools such as Tableau, Power BI, or D3.js to create real-time data visualization dashboards to show dynamic changes in data and model prediction results.

[0081] Model monitoring: Monitor the performance indicators of the model in the production environment, including prediction accuracy, latency, and throughput, to ensure that the model continues to be effective. You can use monitoring tools on the cloud service platform or open source tools such as Prometheus and Grafana.

[0082] 6) Model optimization and iteration

[0083] Performance evaluation: Use confusion matrix, precision, recall and other indicators to evaluate model performance, and optimize the model based on the feedback. You can refer to the detailed introduction of machine learning model evaluation in the CSDN blog.

[0084] Model iteration: Update the model regularly to maintain its accuracy and relevance based on business needs and data changes. Model optimization can be performed using methods such as policy iteration, such as the policy iteration process mentioned in the CSDN blog.

[0085] Through the above steps, a complete machine learning or deep learning solution is implemented from data preprocessing to model deployment to identify and analyze patterns and trends in the water-rock interaction process.

[0086] The automatic data collection system for water-rock interaction experiments during geothermal reinjection provided in this embodiment has the ability to efficiently, accurately and in real time collect water-rock interaction experimental data during the reinjection process. The system not only helps to improve the efficiency of the experiment and the reliability of the data, but also promotes the in-depth and extensive scientific research, promotes the effective development and utilization of geothermal energy, and provides important support for environmental protection and sustainable development.

[0087] Obviously, the above-mentioned embodiments and comparative examples are only a part of the present invention, rather than all examples. The preferred embodiments of the present invention disclosed above are only used to help explain the present invention, without describing all the details in detail. According to the content of this specification, more modifications and changes can be made. Based on the embodiments and comparative examples in the present invention, all other embodiments and various modifications or supplements obtained by ordinary technicians in the field without making creative work, as long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, belong to the scope of protection of the present invention.

[0088] In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiments and comparative examples. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments in a suitable manner.

Claims

1. An automatic data collection system for water-rock interaction experiment during geothermal recharge, which uses a constant temperature water bath oscillator to conduct indoor simulation experiments on water-rock interaction during geothermal recharge, and is characterized by: During the experiment, the ion concentration sensor, TDS sensor and pH sensor were used to measure the solution ion concentration, TDS and pH respectively. The ion concentration sensor, TDS sensor and pH sensor were connected to ESP8266 using the serial communication protocol. The data was uploaded to the cloud through the built-in Wi-Fi module of ESP8266, and real-time subscription was achieved through the cloud built with .NET technology.

2. According to claim 1, the automatic data collection system for water-rock interaction experiment in geothermal recharge process is characterized by: It also includes cloud-based data processing: using data analysis algorithms in the cloud to identify patterns and trends in the water-rock interaction process; Real-time subscription: Optimize the real-time subscription function implemented by .NET technology and use Apache Kafka or RabbitMQ to improve the scalability and responsiveness of the system.

3. According to claim 2, the automatic data collection system for water-rock interaction experiments during geothermal recharge is characterized in that: Cloud data processing includes: 1) Data preprocessing and feature engineering Use Python's Pandas library to clean the data, including filling missing values, deleting duplicate values, and handling outliers; non-dimensionalize the data to facilitate model processing; and perform feature selection; 2) Model training and verification Select a suitable model based on data characteristics; divide the data set into a training set and a test set; train the model using the training set data, and adjust the model parameters through cross-validation to optimize performance; 3) Model deployment and real-time data processing Use cloud services to deploy models; use Apache Kafka or Amazon Kinesis to collect and process real-time data streams, and then use PySpark for data analysis and model prediction; 4) Upload data to the cloud Use cloud storage services to upload data; choose appropriate cloud storage services to store and back up data, and ensure data security and accessibility; 5) Data visualization and monitoring Use Tableau, Power BI, or D3.js to create real-time data visualization dashboards to display dynamic changes in data and model prediction results; monitor the performance indicators of the model in the production environment, including prediction accuracy, latency, and throughput, to ensure that the model continues to work; 6) Model optimization and iteration Evaluate model performance using confusion matrix, precision, and recall, and optimize the model based on feedback; regularly update the model to maintain its accuracy and relevance based on business needs and data changes.

4. According to claim 1, the automatic data collection system for water-rock interaction experiment in geothermal recharge process is characterized by: The experimental samples were collected from the abandoned coal mine goaf, which were mine water and mudstone. The mudstone samples were mechanically crushed and placed on a 200-mesh sieve for screening.

5. According to claim 1, the automatic data collection system for water-rock interaction experiment in geothermal recharge process is characterized by: The parameters in the experimental process, such as time, temperature and speed, were set to restore the corresponding parameters of water-rock interaction during the reinjection process of abandoned coal mine goaf. The experimental time periods were 1 day, 5 days, 10 days, 15 days, 20 days and 25 days; the experimental temperatures were 30°C, 50°C and 70°C; and the experimental oscillation speed was 100r / min.

6. The automatic data acquisition system for water-rock interaction experiment in geothermal recharge process according to claim 1 is characterized by: The ion concentration sensor is used to measure the concentration of specific ions in a solution, including sodium ions, potassium ions, calcium ions, magnesium ions, bicarbonate ions, chloride ions, and sulfate ions, and the concentration unit is mg / L.

7. The automatic data collection system for water-rock interaction experiment in geothermal recharge process according to claim 1 is characterized by: The TDS sensor is used to measure the amount of dissolved solids in a solution, including salts, minerals, and organic matter. The working principle is to use conductivity measurement technology to calculate the TDS value by measuring the ability of electrolytes to transmit electric current in water.

8. The automatic data collection system for water-rock interaction experiment in geothermal recharge process according to claim 1, characterized in that: The serial communication protocol is 2 C, has the function of supporting multiple sensors to be used simultaneously, I 2 C is used to connect the sensors, allowing a single-chip microcomputer to communicate with multiple sensors through simple two wires.

9. The automatic data collection system for water-rock interaction experiment in geothermal recharge process according to claim 1, characterized in that: The cloud built with .NET technology implements the real-time subscription function to obtain the changing trend and real-time status of experimental data. The process includes the following steps: Create a new ASP.NET Core Web API project using the .NET CLI; create a SensorData model to store sensor data; create a MongoDB context to handle database operations; create a controller to handle the upload and real-time features of sensor data; Use SignalR to add real-time functions, install the SignalR package; create a SignalR Hub; configure Startup; trigger events when sensor data is uploaded; Deploy the .NET application to the cloud, configure environment variables and database connections on the cloud service to ensure normal operation.

Images