Biogas engineering big data collection and analysis system

Abstract

The application discloses a biogas engineering big data collection and analysis system and belongs to the technical field of renewable energy engineering informatization. The system comprises the following modules: a mobile terminal data collection module; a cloud terminal data management module, which is arranged on a cloud server and is used for receiving, storing and statistically summarizing research data uploaded by each project; a data comparison and verification module; a regional gas production analysis module; an AI big data prediction module, which is used for deep analysis of biogas engineering data in the whole country by using a machine learning algorithm, and is used for forming gas production potential prediction and engineering benefit evaluation; and a project archive management module, which is used for archiving and filing project data such as investment schemes, equipment lists, field photos, meeting minutes and the like and for permission management. The application realizes mobile collection, cloud centralized management and intelligent analysis of biogas engineering research data, significantly improves project research efficiency and data accuracy, and provides technical support for big data and standardization of the biogas engineering industry in the whole country.

Description

Technical Field

[0001] This invention belongs to the field of information technology for renewable energy engineering, and specifically relates to a biogas engineering big data acquisition and analysis system. Background Technology

[0002] Biogas projects, as an important form of biomass energy utilization, play a vital role in promoting the resource recovery of organic waste, reducing greenhouse gas emissions, and providing clean energy. With my country's increasing policy support for the renewable energy industry, biogas purification and biomethane projects are experiencing accelerated development nationwide.

[0003] However, the following prominent problems still exist in the research, construction, and operation management of biogas projects:

[0004] First, data collection methods are outdated. Currently, on-site surveys for biogas purification projects generally rely on paper forms or Excel spreadsheets for data recording. Survey personnel need to handwrite a large number of technical parameters on-site and then enter them into the computer back in the office. This method is not only inefficient but also prone to data entry errors. There is a lack of effective correlation between on-site photos and table data, making later traceability difficult.

[0005] Second, the data is scattered and difficult to aggregate. The survey data of each project is stored in a scattered manner on personal computers or departmental servers. There is a lack of unified data format standards and centralized storage platforms. It is difficult for headquarters managers to keep track of the survey progress and data quality of projects across the country in real time, let alone conduct horizontal comparisons and in-depth analysis of a large amount of project data.

[0006] Third, the accuracy of the data is difficult to verify. The gas production of biogas projects is affected by various factors such as the type of raw materials, climate conditions, and process parameters. The gas production rate data of the same raw materials can vary in different projects. Due to the lack of an automatic comparison mechanism for similar data, abnormal data is often difficult to detect in a timely manner, leading to risks in investment decisions based on inaccurate data.

[0007] Fourth, there is a lack of systematic research on regional gas production patterns. The gas production characteristics of raw materials such as food waste and landfill leachate are closely related to regional population density, dietary habits, and climate conditions. However, a regional gas production analysis model based on large-scale measured data has not yet been established, which cannot provide a reliable basis for capacity planning of new projects.

[0008] Fifth, project data management is disorganized. Key project documents such as investment plans, equipment lists, site photos, and meeting minutes are usually stored as scattered files on different people's computers. There is a lack of a unified archiving and version management mechanism, resulting in low team collaboration efficiency and the risk of data gaps when personnel change.

[0009] In recent years, the rapid development of mobile internet, cloud computing, and artificial intelligence technologies has provided new technological paths for solving the aforementioned problems. WeChat mini-programs, with their no-installation and instant-use features, are ideally suited as data collection entry points for on-site surveys; cloud servers offer elastic and scalable storage and computing capabilities; and machine learning algorithms can extract patterns and build predictive models from massive amounts of data. However, there is currently a lack of technical solutions for organically combining these three technologies and specifically applying them to the biogas engineering field to form a complete big data collection and analysis system. Summary of the Invention

[0010] The purpose of this invention is to provide a biogas engineering big data collection and analysis system, which realizes mobile collection, cloud-based centralized management and intelligent analysis of biogas engineering survey data, significantly improves project survey efficiency and data accuracy, and provides technical support for the big data and standardization of the national biogas engineering industry, thereby solving the problems in the prior art mentioned in the background.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] A biogas engineering big data acquisition and analysis system, comprising:

[0013] The mobile data acquisition module uses WeChat Mini Program to achieve structured input of on-site project data. The data includes raw material source information, wastewater parameters, fermentation type, project scale parameters, biogas slurry disposal method, biogas component data, utilization method parameters, site facility parameters, and safety and compliance parameters. It also supports on-site photo shooting and uploading, GPS positioning marking, and offline caching functions.

[0014] The cloud-based data management module, deployed on a cloud server, is used to receive and store survey data uploaded by various projects through the mobile data collection module, classify and summarize the data, generate multi-dimensional reports, and provide data query and export interfaces.

[0015] The data comparison and verification module is used to perform horizontal comparison of gas production rate data from similar raw material sources, establish a standard reference range for raw material type and gas production rate, and automatically trigger an anomaly flag and send a verification reminder to the management personnel when the uploaded data exceeds the standard reference range.

[0016] The regional gas production analysis module, based on the collected data of regional raw materials such as kitchen waste and landfill leachate, and combined with parameters such as regional climate characteristics, population density, and economic development level, establishes a correlation analysis model between the tonnage of waste and the amount of gas produced in each region.

[0017] The AI ​​big data prediction module uses machine learning algorithms to conduct in-depth analysis of biogas project data collected nationwide, forming a heat map of biogas project distribution, a gas production potential prediction model, and a project benefit evaluation system.

[0018] The project file management module is used to classify and archive project materials such as investment plans, equipment lists, site photos, meeting minutes, and contract texts, and supports multi-user permission management and collaborative editing functions.

[0019] Preferably, the mobile data acquisition module includes:

[0020] The structured form engine transforms the three major form systems in the biogas purification project survey checklist—raw material and process parameters, biogas utilization and pipeline matching, and site facilities and safety compliance—into a standardized input interface on the WeChat mini program.

[0021] The intelligent photo-taking component automatically activates the camera function at the survey items that need to be recorded on-site, and the photos taken are automatically associated with the corresponding survey items and have timestamps and GPS coordinate watermarks added.

[0022] The offline data caching mechanism temporarily stores survey data in a local cache queue when the on-site network environment is unstable, and automatically uploads it to the cloud in order after the network is restored.

[0023] Preferably, the horizontal comparison logic of the data comparison and verification module includes:

[0024] A standard gas production rate reference database is established according to the type of raw materials. The reference database includes the baseline values ​​of 6.2 m³ / ton for domestic waste, 55 m³ / ton for kitchen waste, 75 m³ / ton for food waste, 350 m³ / ton for dried straw, 90 m³ / ton for chicken manure, 50 m³ / ton for duck manure, 5 m³ / ton for pig manure, and 25 m³ / ton for cow manure.

[0025] Calculate the mean and standard deviation of gas production rate of the same type of raw material within a preset time window, and mark data that exceed the mean ± 2 times the standard deviation as outliers;

[0026] For marked abnormal data, a review process is triggered, a verification reminder message is sent to the project research personnel, and the verification results and correction values ​​are recorded.

[0027] Preferably, the method for establishing the regional gas production analysis module includes:

[0028] Collect basic data on the amount of food waste generated, leachate production, scale of existing treatment facilities, and geographical and climatic conditions in each region.

[0029] Using the amount of waste as the independent variable and the measured amount of gas produced as the dependent variable, a regional linear regression model was constructed. The model is in the form of: Q=α×M+β×T+γ×P+ε, where Q is the amount of gas produced, M is the amount of waste, T is the fermentation temperature, P is the population density, α, β, and γ are the regression coefficients, and ε is the error term.

[0030] Based on the model output, a distribution map of gas production per ton of waste in each region is drawn to identify areas with high gas production potential and areas with inefficient operation.

[0031] Preferably, the AI ​​big data prediction module includes:

[0032] The data preprocessing submodule is used to clean, normalize, and perform feature engineering on the raw data of biogas projects across the country.

[0033] The gas production potential prediction submodule, based on random forest or gradient boosting tree algorithm, takes as input the feed type, processing scale, climate region, and process type characteristics, and outputs the predicted gas production value and confidence interval.

[0034] The project benefit evaluation submodule calculates the investment payback period, net present value, and internal rate of return for each project by combining investment costs, operating costs, biogas sales prices, and subsidy policy parameters.

[0035] The national biogas project big data visualization submodule uses geographic information maps as a carrier to overlay and display multi-dimensional information such as gas production, operating status, and economic benefits of each project.

[0036] Preferably, the collaborative functions of the project file management module include:

[0037] A role-based access control system distinguishes the operational permissions of project administrators, researchers, technical reviewers, and visitors.

[0038] The version control mechanism saves historical versions of every modification to the investment plan and equipment list documents, and supports version rollback and difference comparison.

[0039] The meeting minutes structure extraction function uses natural language processing technology to automatically extract decision-making items, responsible persons, and time nodes from meeting minutes text, and generate a list of tasks to be done.

[0040] Preferably, it also includes a method for collecting and analyzing big data from biogas projects, comprising the following steps:

[0041] S1. Using the structured form engine on the WeChat mini program, collect data on the source of raw materials, wastewater parameters, fermentation type, project scale, biogas slurry disposal method, biogas composition, utilization method, site facilities and safety compliance at the biogas purification project site. At the same time, collect on-site photos and record GPS location information.

[0042] S2. The collected data is uploaded to the cloud server via HTTPS encrypted transmission protocol. The cloud server performs integrity verification and format standardization processing on the received data and stores it in the classification database.

[0043] S3. Perform a horizontal comparison of gas production rate data for similar raw materials. When the detection data deviates from the standard reference range, trigger an anomaly marker and a review reminder.

[0044] S4. Based on the collected data of kitchen waste and leachate in each region, and combined with regional characteristic parameters, establish a correlation analysis model between waste tonnage and gas production.

[0045] S5. Utilize machine learning algorithms to conduct in-depth analysis of biogas engineering data nationwide, and generate gas production potential prediction and engineering benefit evaluation results.

[0046] S6 archives project materials such as investment plans, equipment lists, site photos, and meeting minutes to a cloud archive by project number, supporting multi-user collaborative management and access control.

[0047] Technical effects and advantages of the present invention: The biogas engineering big data acquisition and analysis system proposed in this invention has the following advantages compared with the prior art:

[0048] 1. Significantly improves data collection efficiency. Researchers can complete structured data entry and photo taking on-site via WeChat mini-program, eliminating the need for paper forms and secondary computer entry, reducing the research time for a single project by more than 50%.

[0049] 2. Ensure data authenticity and traceability. On-site photos are timestamped and GPS watermarked, and data is automatically linked to photos, effectively preventing data fraud; an offline caching mechanism ensures normal operation in environments without network access.

[0050] 3. Improve data accuracy. The automatic horizontal comparison mechanism of gas production rates of similar raw materials can promptly detect abnormal data and trigger a review process, avoiding investment decisions based on erroneous data.

[0051] 4. Establish understanding of regional gas production patterns. Based on a regression analysis model using large-scale measured data, this study systematically reveals for the first time the quantitative relationship between waste tonnage and gas production in various regions, providing a scientific basis for project site selection and capacity planning.

[0052] 5. Achieve nationwide big data analytics for biogas projects. The AI ​​analysis module will aggregate biogas project data scattered across the country into a unified knowledge asset, supporting both macro-level industry decision-making and micro-level project optimization through visualized maps and predictive models.

[0053] 6. Improve project collaboration efficiency. A unified cloud-based file management platform breaks down information silos, and version control and closed-loop task management ensure project data security and orderly team collaboration. Attached Figure Description

[0054] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] This invention provides, for example Figure 1 The biogas engineering big data acquisition and analysis system shown includes:

[0057] The mobile data acquisition module enables structured input of on-site project data via WeChat mini-program. The data includes raw material source information, wastewater parameters, fermentation type, project scale parameters, biogas slurry disposal method, biogas composition data, utilization method parameters, site facility parameters, and safety and compliance parameters. It also supports on-site photo capture and upload, GPS location marking, and offline caching functions.

[0058] The cloud-based data management module, deployed on a cloud server, is used to receive and store survey data uploaded by various projects through the mobile data collection module, classify and summarize the data, generate multi-dimensional reports, and provide data query and export interfaces.

[0059] The data comparison and verification module is used to compare the gas production rate data of similar raw material sources horizontally, establish a standard reference range between raw material type and gas production rate, and automatically trigger an anomaly mark and send a verification reminder to the management personnel when the uploaded data exceeds the standard reference range.

[0060] The regional gas production analysis module, based on the collected data of regional raw materials such as kitchen waste and landfill leachate, and combined with parameters such as regional climate characteristics, population density, and economic development level, establishes a correlation analysis model between the tonnage of waste and the amount of gas produced in each region.

[0061] The AI ​​big data prediction module uses machine learning algorithms to conduct in-depth analysis of biogas project data collected nationwide, forming a heat map of biogas project distribution, a gas production potential prediction model, and a project benefit evaluation system.

[0062] The project file management module is used to classify and archive project materials such as investment plans, equipment lists, site photos, meeting minutes, and contract texts, and supports multi-user permission management and collaborative editing functions.

[0063] Please see Figure 1 :

[0064] The mobile data acquisition module includes:

[0065] The structured form engine transforms the three major form systems in the biogas purification project survey checklist—raw material and process parameters, biogas utilization and pipeline matching, and site facilities and safety compliance—into a standardized input interface on the WeChat mini program.

[0066] The intelligent photo-taking component automatically activates the camera function at the survey items that need to be recorded on-site, and the photos taken are automatically associated with the corresponding survey items and have timestamps and GPS coordinate watermarks added.

[0067] The offline data caching mechanism temporarily stores survey data in a local cache queue when the on-site network environment is unstable, and automatically uploads it to the cloud in order after the network is restored.

[0068] More specifically, the mobile data acquisition module uses WeChat Mini Programs to achieve structured data entry from project sites. This module has a built-in structured form engine that transforms the three major form systems in the biogas purification project survey checklist—raw materials and process parameters, biogas utilization and pipeline matching, and site facilities and safety compliance—into a standardized data entry interface on the WeChat Mini Program.

[0069] Regarding raw materials and process parameters, researchers can record data such as raw material type (food processing plant waste, livestock and poultry manure, kitchen waste, straw, etc.), daily / monthly / yearly processing volume, anaerobic system influent water volume and COD content, fermentation temperature type (high temperature / medium temperature), fermentation tank volume and number, heat preservation and heating methods and parameters, and biogas slurry disposal methods through drop-down selection or manual input.

[0070] Regarding biogas utilization and pipeline network support, the system supports the input of biogas component data (methane content, carbon dioxide content, hydrogen sulfide content, oxygen content, ammonia content), current utilization mode (power generation / boiler combustion / flare venting / purification and external sales) and its detailed parameters, electricity consumption, end-point gas pipeline pressure and distance, and information on surrounding biogas projects that can be co-processed.

[0071] Regarding site facilities and safety compliance, the system provides functions such as available land area, geological conditions, land nature, parameters of biogas desulfurization and pretreatment facilities, distances to supporting facilities (biogas pipeline network, power distribution facilities, water supply pipes, drainage pipes, exhaust gas pipelines, fire protection facilities, flare system), electrical support requirements, water price and sewage discharge fees, by-product disposal methods, commercial contract terms, historical equipment reuse status, fire protection and safety system configuration, equipment brand and compliance requirements, and comparison of safety distance measurement values ​​with standard requirements.

[0072] The intelligent photo-taking component automatically activates the camera function at survey sites requiring on-site documentation. Photos taken are automatically linked to the corresponding survey item and time-stamped with GPS coordinates. For example, when recording wastewater flow, the system prompts the user to take a photo of the wastewater flow meter; when recording land area, it prompts the user to attach site dimensions / panoramic view / plan view; and when recording desulfurization facilities, it prompts the user to take a photo to record the current status of the facilities.

[0073] The offline data caching mechanism temporarily stores survey data in a local cache queue when the network environment is unstable. Once the network is restored, the data is automatically uploaded to the cloud in sequence, ensuring that the field survey work is not limited by network conditions.

[0074] Please see Figure 1 :

[0075] More specifically, the cloud data management module is deployed on a cloud server and adopts a microservice architecture design, including data receiving service, storage service, statistical summary service and query export service.

[0076] The data receiving service receives survey data uploaded by mobile devices via HTTPS encrypted transmission protocol. It performs integrity verification (field integrity, numerical range verification, and photo file integrity verification) and format standardization processing on the received data. After converting unstructured or semi-structured data into a unified database schema, it stores the data in the classification database.

[0077] The storage service uses relational databases (such as MySQL or PostgreSQL) to store structured survey data, object storage (such as MinIO or AWS S3) to store on-site photos, test reports and other files, and time-series databases (such as InfluxDB) to store continuous monitoring data (such as flow, temperature and pressure).

[0078] The statistical aggregation service aggregates and calculates data according to project, region, raw material, and time dimensions, and automatically generates multi-dimensional reports, including: project basic information summary table, raw material type distribution statistics table, gas production rate interval statistics table, regional project density statistics table, and survey progress tracking table.

[0079] The query and export service provides a web-based management backend, allowing administrators to filter and query project data using combined criteria, and export data to formats such as Excel and PDF.

[0080] Please see Figure 1 :

[0081] The horizontal comparison logic of the data comparison and verification module includes:

[0082] A standard gas production rate reference database was established based on the type of raw materials. The reference database includes baseline values ​​of 6.2 m³ / ton for municipal solid waste, 55 m³ / ton for kitchen waste, 75 m³ / ton for food waste, 350 m³ / ton for dried straw, 90 m³ / ton for chicken manure, 50 m³ / ton for duck manure, 5 m³ / ton for pig manure, and 25 m³ / ton for cow manure.

[0083] Calculate the mean and standard deviation of gas production rate of the same type of raw material within a preset time window, and mark data that exceed the mean ± 2 times the standard deviation as outliers;

[0084] For marked abnormal data, a review process is triggered, a verification reminder message is sent to the project research personnel, and the verification results and correction values ​​are recorded.

[0085] It should be noted that the data comparison and verification module is used to perform horizontal comparison of gas production data from similar raw material sources, establish a standard reference range for raw material type and gas production rate, and automatically trigger an anomaly flag and send a verification reminder to the management personnel when the uploaded data exceeds the standard reference range.

[0086] This module first establishes a standard gas production rate reference database according to the type of raw materials. The benchmark values ​​include: 6.2 m³ / ton for domestic waste, 55 m³ / ton for kitchen waste, 75 m³ / ton for food waste, 350 m³ / ton for dried straw, 90 m³ / ton for chicken manure (21% solids content), 50 m³ / ton for duck manure (12% solids content), 5 m³ / ton for pig manure (0.5% solids content), and 25 m³ / ton for cow manure.

[0087] The system continuously collects actual gas production data from each project and calculates the mean and standard deviation of gas production rates for similar raw materials within a preset time window (e.g., the most recent 12 months). For newly uploaded data, the system calculates its gas production rate and determines whether it falls within the mean ± 2 standard deviations. If it exceeds this range, the data is marked as an outlier, and a verification reminder message is sent to the project survey personnel, requesting confirmation of whether the raw material type selection is correct, whether the processing volume statistics are accurate, and whether the metering instruments are functioning properly. The verification results and correction values ​​are recorded in the data audit log as the basis for updating the subsequent comparison model.

[0088] In addition, the module also supports cross-project comparison. When there are multiple similar raw material projects in the same area, the system automatically generates a horizontal comparison chart to intuitively show the differences in gas production rates of each project, helping technicians to identify efficient and inefficient projects and uncover best practices.

[0089] Furthermore, the system adopts Statistical discrimination algorithm calculates the average gas production rate of similar raw materials within a preset time window. Standard deviation The calculation formula is: Through the interval formula To determine the validity of the data, data that exceed the mean ± 2 standard deviations are marked as outliers.

[0090] In the formula: =Average gas production rate of similar raw materials within a preset time window; =Total number of samples; =No. Actual gas production rate of the project (m³) 3 / ton); =Sample standard deviation of gas production rate, characterizing the degree of data dispersion.

[0091] Please see Figure 1 :

[0092] The methods for establishing the regional gas production analysis module include:

[0093] Collect basic data on the amount of food waste generated, leachate production, scale of existing treatment facilities, and geographical and climatic conditions in each region.

[0094] Using the amount of waste as the independent variable and the measured amount of gas produced as the dependent variable, a regional linear regression model was constructed. The model is in the form of: Q=α×M+β×T+γ×P+ε, where Q is the amount of gas produced, M is the amount of waste, T is the fermentation temperature, P is the population density, α, β, and γ are the regression coefficients, and ε is the error term.

[0095] Based on the model output, a distribution map of gas production per ton of waste in each region is drawn to identify areas with high gas production potential and areas with inefficient operation.

[0096] It should be noted that the regional gas production analysis module is based on the collected data of regional raw materials such as kitchen waste and landfill leachate, and combines regional climate characteristics, population density, and economic development level parameters to establish a correlation analysis model between the tonnage of waste and the amount of gas produced in each region.

[0097] The module first collects basic data on the amount of food waste generated, leachate production, scale of existing treatment facilities, and geographical and climatic conditions in each region. Among these, the amount of food waste generated is positively correlated with the regional population and the level of development of the catering industry; the amount of leachate production is related to the scale of the landfill and rainfall conditions; and climatic conditions (average annual temperature and humidity) directly affect the efficiency of anaerobic fermentation.

[0098] A regional linear regression model was constructed using waste tonnage as the independent variable and measured gas production as the dependent variable. The model form is: Q = α × M + β × T + γ × P + ε, where Q is gas production (m³ / day), M is waste tonnage (tons / day), T is fermentation temperature (°C), P is population density (people / km²), α, β, and γ are regression coefficients, and ε is the error term. The coefficients were determined by least squares fitting, and the generalization ability was evaluated through cross-validation.

[0099] Based on the model output, the system automatically generates a map showing the distribution of gas production per ton of waste in each region, using color intensity to indicate gas production efficiency and identifying areas with high gas production potential and inefficient operation. Managers can then optimize project layout by prioritizing new projects in high-potential areas and implementing process modifications or operational optimizations in inefficient areas.

[0100] Furthermore, based on the least squares multiple linear regression algorithm, a regional gas production correlation model is constructed with waste tonnage as the independent variable and measured gas production as the dependent variable. The model form is as follows: Using loss function Iterative solution of regression coefficients And through goodness of fit Verify the reliability of the model.

[0101] In the formula, Regional average daily gas production (m³) 3 / day); M: Average daily organic waste processing capacity in the region (tons / day); T: Average annual fermentation baseline temperature in the region (°C); P: Population density in the region (people / km²) 2 ); : Model regression coefficients; Random error term; To predict gas production for the model, This represents the average gas production per sample. A value >0.85 indicates the model is effective and can be used for regional gas production potential assessment.

[0102] Please see Figure 1 :

[0103] The AI ​​big data prediction module includes:

[0104] The data preprocessing submodule is used to clean, normalize, and perform feature engineering on the raw data of biogas projects across the country.

[0105] The gas production potential prediction submodule, based on random forest or gradient boosting tree algorithm, takes as input the feed type, processing scale, climate region, and process type characteristics, and outputs the predicted gas production value and confidence interval.

[0106] The project benefit evaluation submodule calculates the investment payback period, net present value, and internal rate of return for each project by combining investment costs, operating costs, biogas sales prices, and subsidy policy parameters.

[0107] The national biogas project big data visualization submodule uses geographic information maps as a carrier to overlay and display multi-dimensional information such as gas production, operating status, and economic benefits of each project.

[0108] More specifically, the AI ​​big data prediction module uses machine learning algorithms to conduct in-depth analysis of biogas project data collected nationwide, forming a heat map of biogas project distribution across the country, a gas production potential prediction model, and an engineering benefit evaluation system.

[0109] The module consists of four sub-modules:

[0110] The data preprocessing submodule is responsible for cleaning (removing null values ​​and outliers), normalizing (converting parameters of different dimensions to a unified scale), and performing feature engineering processing (extracting feature variables such as raw material type codes, seasonal characteristics, regional codes, and process type codes) on the raw data of biogas projects across the country.

[0111] The gas production potential prediction submodule is based on a random forest or gradient boosting tree (XGBoost / LightGBM) algorithm. It takes as input features such as feedstock type, processing scale, climate region, process type, and insulation method, and outputs a predicted gas production value and a 95% confidence interval. The model training data includes feedstock parameters and actual gas production records from historical survey data, and the optimal hyperparameters are determined through grid search and cross-validation.

[0112] The project benefit assessment submodule combines parameters such as investment costs (equipment procurement, civil construction, installation and commissioning), operating costs (energy consumption, labor, maintenance), biogas sales prices, and government subsidy policies to calculate the payback period (PBP), net present value (NPV), and internal rate of return (IRR) for each project. The module supports sensitivity analysis to assess the impact of factors such as raw material price fluctuations, changes in subsidy policies, and equipment failure rates on project profitability.

[0113] The National Biogas Project Big Data Visualization Submodule uses a geographic information map as a carrier to overlay and display multi-dimensional information such as gas production (bubble size), operating status (color: green for normal / yellow for warning / red for fault), and economic benefits (IRR value displayed when hovering). It supports filtering and drilling analysis by region, by raw material type, and by project status.

[0114] Furthermore, by combining investment costs, operating costs, biogas sales prices, and subsidy policy parameters, and through static payback period... Net Present Value Internal Rate of Return Mathematical models are used to calculate the payback period, net present value, and internal rate of return for each project.

[0115] In the formula, Project annual income (biogas sales, organic fertilizer, subsidies) Annual operating costs (raw materials, energy consumption, labor, operation and maintenance), ( ) represents the net cash flow in year t. Initial total investment of the project Project operation calculation period, A value of ≥0 indicates that the project is economically feasible.

[0116] Please see Figure 1 :

[0117] The collaborative functions of the project file management module include:

[0118] A role-based access control system distinguishes the operational permissions of project administrators, researchers, technical reviewers, and visitors.

[0119] The version control mechanism saves historical versions of every modification to the investment plan and equipment list documents, and supports version rollback and difference comparison.

[0120] The meeting minutes structure extraction function uses natural language processing technology to automatically extract decision-making items, responsible persons, and time nodes from meeting minutes text, and generate a list of tasks to be done.

[0121] More specifically, the project file management module is used to classify and archive project materials such as investment plans, equipment lists, site photos, meeting minutes, and contract texts, and supports multi-user permission management and collaborative editing functions.

[0122] The module employs a role-based access control (RBAC) system, defining roles such as project administrator, researcher, technical reviewer, and visitor. Project administrators have full operational permissions; researchers can upload materials and fill out forms but cannot delete archived data; technical reviewers can approve materials and publish review comments; visitors can only view public information.

[0123] The version control mechanism saves historical versions of every modification to documents such as investment plans and equipment lists, recording the modifier, modification time, and a summary of the modifications, supporting version rollback and difference comparison. When multiple people edit the same document simultaneously, the system uses an optimistic locking mechanism to prevent data conflicts.

[0124] The structured extraction function for meeting minutes utilizes Natural Language Processing (NLP) technology to automatically extract decision-making matters, responsible persons, and time nodes from meeting minutes text, generating a to-do list and pushing it to relevant personnel. Task completion status is linked to the overall project progress, forming a closed-loop management system.

[0125] The on-site photo management system supports browsing in three views: by project, by survey date, and by shooting location (map mode). Photos are automatically associated with corresponding survey form entries, and clicking on a photo will directly jump to the relevant data record.

[0126] Please see Figure 1 :

[0127] It also includes a method for collecting and analyzing big data from biogas projects, comprising the following steps:

[0128] S1. Using the structured form engine on the WeChat mini program, collect data on the source of raw materials, wastewater parameters, fermentation type, project scale, biogas slurry disposal method, biogas composition, utilization method, site facilities and safety compliance at the biogas purification project site. At the same time, collect on-site photos and record GPS location information.

[0129] S2. The collected data is uploaded to the cloud server via HTTPS encrypted transmission protocol. The cloud server performs integrity verification and format standardization processing on the received data and stores it in the classification database.

[0130] S3. Perform a horizontal comparison of gas production rate data for similar raw materials. When the detection data deviates from the standard reference range, trigger an anomaly marker and a review reminder.

[0131] S4. Based on the collected data of kitchen waste and leachate in each region, and combined with regional characteristic parameters, establish a correlation analysis model between waste tonnage and gas production.

[0132] S5. Utilize machine learning algorithms to conduct in-depth analysis of biogas engineering data nationwide, and generate gas production potential prediction and engineering benefit evaluation results.

[0133] S6 archives project materials such as investment plans, equipment lists, site photos, and meeting minutes to a cloud archive by project number, supporting multi-user collaborative management and access control.

[0134] Example 1: System Overall Architecture

[0135] The system of this invention adopts a four-layer architecture design, from bottom to top as follows:

[0136] (1) Data Acquisition Layer: Using WeChat Mini Programs as the entry point, it supports both iOS and Android platforms. The Mini Program front-end is developed using the Taro or uni-app cross-platform framework, enabling a single codebase to run on multiple platforms. The form component is based on Tencent WeUI design specifications and adapts to different screen sizes. The camera component calls the WeChat native APIs wx.chooseImage and wx.getLocation to obtain photo and location information. Offline storage uses the WeChat local caching API wx.setStorageSync, supporting a maximum data cache of 10MB.

[0137] (2) Network Transport Layer: HTTPS / TLS 1.3 encrypted transmission protocol is adopted to ensure the security of data during transmission over the public network. For large files (such as on-site photos), a chunked upload mechanism is adopted, with each chunk being 2MB and supporting resume upload. The interface adopts RESTful design, JSON serialization, and status codes follow the HTTP standard.

[0138] (3) Cloud service layer: Deployed on Alibaba Cloud or Huawei Cloud ECS cluster, using Docker containerization and Kubernetes orchestration management. The database layer uses MySQL 8.0 master-slave architecture to ensure high availability, Redis to cache hot data, and RabbitMQ message queue to process asynchronous tasks (such as photo processing, report generation, and message push).

[0139] (4) AI Analysis Layer: Model training and inference services are built based on the Python machine learning ecosystem (scikit-learn, XGBoost, TensorFlow). Model training tasks are executed offline on GPU servers, and trained models are managed and deployed via MLflow. The online prediction service uses Flask or FastAPI to provide a REST interface, with a response time controlled within 200ms.

[0140] (5) Visualization layer: The Web management backend is developed using Vue.js 3 + Element Plus technology stack, the GIS map uses Gaode Map or Baidu Map JavaScript API, and the chart component uses ECharts 5.

[0141] Example 2: On-site data acquisition process

[0142] After arriving at the project site, the researchers opened the WeChat mini-program and followed these steps to complete the data collection:

[0143] Step 1: Create a research task. Select or search for a project name; the system will automatically load the project's historical research records and basic information. If it's a new project, fill in basic information such as the project address, owner, and contact person; the system will obtain GPS coordinates through address parsing.

[0144] Step 2: Fill in the raw material and process parameters. Guided by the form engine, fill in the following parameters in sequence: raw material source (select "food processing plant", enter daily processing capacity of 30 tons / day), wastewater parameters (water volume entering the anaerobic system of 50 tons / day, COD content of 3500mg / L), fermentation type (select "mesophilic fermentation"), project scale (high-temperature fermentation tank volume of 500m³ / unit, quantity of 2 units), heat preservation and heating method (select "steam heating", enter steam flow rate of 0.5 tons / h, temperature of 120℃, pressure of 4 kg), and biogas slurry disposal method (select "farmland irrigation").

[0145] Step 3: Take a related photo. When entering the wastewater flow rate, the system detects the photo-related marker for this field, automatically activates the camera, and the researcher takes a photo of the flow meter reading. The photo is automatically associated with this data entry.

[0146] Step 4: Fill in the biogas utilization and pipeline matching parameters. Enter the biogas composition (CH4 55%, CO2 40%, H2S 200ppm, O2 1%, NH3 0.5%), current utilization method (select "power generation", fill in the generator set gas consumption of 800m³ / day, configuration of 250kW×2 units, power generation per cubic meter of biogas 1.8kWh / m³), electricity consumption (electricity price 0.65 yuan / kWh, transformer load 500KVA), and end-point gas pipeline (distance 200 meters, gas company "XX Gas").

[0147] Step 5: Fill in the site facilities and safety compliance parameters. Enter the available land area of ​​800m², the site geological conditions as "sandy soil", the land use as "industrial land"; the desulfurization method as "biological desulfurization", and the H2S content after desulfurization as 15ppm; measure and fill in all safety distances (fire hydrant distance from the project is 30 meters, flare distance from the purification device is 40 meters); fill in the water price as 3.5 yuan / ton and the wastewater treatment price as 8 yuan / ton.

[0148] Step 6: Data Validation and Submission. The form engine performs front-end validation on required fields and numerical ranges. Once validation is successful, the data enters the local cache queue. If the network is stable, the data is immediately uploaded to the cloud; if the network is poor, the data is temporarily stored locally and automatically synchronized once the network is restored.

[0149] Example 3: Data Comparison and Verification Process

[0150] Once a project uploads data on the processing volume and gas production of food waste, the system compares and verifies the data according to the following process:

[0151] Step 1: Calculate the gas production rate. The project processes 30 tons of kitchen waste per day, with an average daily gas production of 1800 m³. The system calculates the gas production rate to be 60 m³ / ton.

[0152] Step 2: Query the standard reference range. The system shows that the standard reference value for the gas production rate of food waste is 75 m³ / ton. The average gas production rate of similar projects in the past 12 months is 70 m³ / ton, with a standard deviation of 8 m³ / ton. The normal range is 54-86 m³ / ton (mean ± 2 times standard deviation).

[0153] Step 3: Determine if it is abnormal. 60 m³ / ton is within the normal range, but it is 15 m³ / ton lower than the standard reference value. The system marks this data as "low" (yellow warning) instead of "abnormal" (red warning).

[0154] Step 4: Trigger Alert. The system automatically sends a WeChat message to the project's research manager: "The gasification rate of this project's food waste (60m³ / ton) is lower than the average level of similar projects. Please verify whether the raw material composition contains a large amount of non-degradable impurities, or whether the fermentation tank temperature meets the standard."

[0155] Step 5: Manual Verification. After receiving the notification, the investigators conducted an on-site inspection of the raw material pretreatment process and found that approximately 15% of the kitchen waste contained plastic packaging materials that had not been effectively separated. The verification result "containing 15% impurities" was entered into the system, and the system recalculated and corrected the gas production rate to 70.6 m³ / ton based on the impurity ratio, updating it to the normal value.

[0156] Step 6: Model Update. The results of this review (including 15% impurities and a corrected gas production rate of 70.6 m³ / ton) are recorded in the data audit log and used periodically to update the statistical parameters of gas production rates for similar raw materials, thereby improving the accuracy of subsequent comparisons.

[0157] Example 4: Construction of Regional Gas Production Analysis Model

[0158] Taking a province in East China as an example, the system collected operational data from 12 food waste treatment projects in the region and constructed a regional gas production analysis model:

[0159] The basic data is shown in the table below:

[0160] Project Number Processing capacity (tons / day) Gas production (m³ / day) Fermentation temperature (°C) Population density (people / km²) Gas production rate (m³ / ton) P001 25 450 35 600 18.0 P002 40 820 38 850 20.5 P003 35 650 36 720 18.6 P004 50 1100 39 920 22.0 P005 30 580 37 680 19.3 ... ... ... ... ... ...

[0161] The model Q = α × M + β × T + γ × P + ε was fitted using a multiple linear regression algorithm. The least squares method yielded the following values: α = 1.85, β = 12.3, γ = 0.08, i.e., Q = 1.85M + 12.3T + 0.08P. The model R² = 0.92, indicating a good fit.

[0162] Based on this model, when a new project is designed to process 50 tons / day, the fermentation temperature is set at 38℃, and the population density of the city is 800 people / km², the predicted gas production is:

[0163] Q = 1.85 × 50 + 12.3 × 38 + 0.08 × 800 = 92.5 + 467.4 + 64 = 623.9 m³ / day, which is equivalent to a gas production rate of 12.48 m³ / ton of waste.

[0164] Example 5: AI Big Data Prediction and Visualization

[0165] The AI ​​analysis module conducted in-depth analysis of data from over 500 biogas projects collected nationwide:

[0166] (1) Data preprocessing: 23 records with missing key fields were cleaned and removed. The gas production rate was logarithmically transformed to eliminate the right skewed distribution. The raw material type was encoded as an 8-dimensional one-hot vector and the province was encoded as a regional code.

[0167] (2) Feature engineering: Construct derivative features such as "raw material processing scale" (logarithmic transformation), "seasonal coefficient" (spring, summer, autumn and winter coding), "process complexity" (single-stage / two-stage / multi-stage anaerobic coding), and "insulation level" (with / without insulation, heat tracing method coding).

[0168] (3) Model Training: The XGBoost algorithm was adopted, with gas production as the regression target. 80% of the data was used for training and 20% for testing. The optimal hyperparameters were determined by 5-fold cross-validation and grid search: max_depth=6, learning_rate=0.05, n_estimators=300. The test set RMSE=45.2m³ / day and MAPE=8.3%, and the prediction accuracy met the requirements of engineering applications.

[0169] (4) Visualization: The location and scale of each project are displayed in the form of bubbles on the national map. The size of the bubble corresponds to the processing volume, and the color depth corresponds to the ratio of actual gas production to predicted gas production (actual / predicted). Areas with a ratio greater than 1.1 (such as some projects in South China) are marked as high-efficiency areas, and areas with a ratio less than 0.9 (such as some projects in Northeast China) are marked as low-efficiency areas, providing direction for targeted technical improvements.

[0170] Example 6: Collaborative Management of Project Files

[0171] Taking the "XX City Kitchen Waste Biogas Purification Project" as an example, this demonstrates the collaborative functions of the file management module:

[0172] (1) Document archiving: The project leader uploads the investment plan PDF (with the file name automatically adding a timestamp), the equipment list Excel (including manufacturer, model, quantity, and price), 120 on-site photos (automatically grouped by survey date and shooting location), and 5 meeting minutes. The system automatically generates a project archive directory tree.

[0173] (2) Access control: The project administrator authorizes 3 research personnel to have "upload" permission, 2 technical reviewers to have "approval" permission, and finance department personnel to have "view investment plan" permission but not to view technical parameter details.

[0174] (3) Version management: The technical reviewer makes suggestions for modification of the equipment selection in the investment plan. The project manager updates the equipment list to version V2. The system automatically retains version V1 and records the modification log: "On March 10, 2026, Mr. Zhang changed the brand of biogas compressor from manufacturer A to manufacturer B. Reason: Manufacturer B's price is reduced by 12% and the warranty period is extended by 1 year."

[0175] (4) The system automatically extracted the following from the meeting minutes of the project progress meeting on March 15: "Decision item: Construction to commence on June 1", responsible person: "Manager Li", and deadline: "Complete the construction permit application by April 25, 2026". This task was automatically added to the to-do list, and the system sent a reminder message to Manager Li on April 20.

[0176] In summary, this invention organically integrates mobile WeChat mini-programs, cloud data management, intelligent comparison and verification, regional gas production modeling, AI big data prediction, and project file collaborative management to construct an information solution covering the entire lifecycle of biogas projects. This significantly improves the efficiency of data collection, management, and analysis in the industry, and provides strong technical support for the standardization and big data development of the biogas engineering industry nationwide.

[0177] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A biogas engineering big data acquisition and analysis system, characterized in that, include: The mobile data acquisition module uses WeChat Mini Program to achieve structured input of on-site project data. The data includes raw material source information, wastewater parameters, fermentation type, project scale parameters, biogas slurry disposal method, biogas component data, utilization method parameters, site facility parameters, and safety and compliance parameters. It also supports on-site photo shooting and uploading, GPS positioning marking, and offline caching functions. The cloud-based data management module, deployed on a cloud server, is used to receive and store survey data uploaded by various projects through the mobile data collection module, classify and summarize the data, generate multi-dimensional reports, and provide data query and export interfaces. The data comparison and verification module is used to perform horizontal comparison of gas production rate data from similar raw material sources, establish a standard reference range for raw material type and gas production rate, and automatically trigger an anomaly flag and send a verification reminder to the management personnel when the uploaded data exceeds the standard reference range. The regional gas production analysis module, based on the collected data of regional raw materials such as kitchen waste and landfill leachate, and combined with parameters such as regional climate characteristics, population density, and economic development level, establishes a correlation analysis model between the tonnage of waste and the amount of gas produced in each region. The AI ​​big data prediction module uses machine learning algorithms to conduct in-depth analysis of biogas project data collected nationwide, forming a heat map of biogas project distribution, a gas production potential prediction model, and a project benefit evaluation system. The project file management module is used to classify and archive project materials such as investment plans, equipment lists, site photos, meeting minutes, and contract texts, and supports multi-user permission management and collaborative editing functions.

2. The biogas engineering big data acquisition and analysis system according to claim 1, characterized in that: The mobile data acquisition module includes: The structured form engine transforms the three major form systems in the biogas purification project survey checklist—raw material and process parameters, biogas utilization and pipeline matching, and site facilities and safety compliance—into a standardized input interface on the WeChat mini program. The intelligent photo-taking component automatically activates the camera function at the survey items that need to be recorded on-site, and the photos taken are automatically associated with the corresponding survey items and have timestamps and GPS coordinate watermarks added. The offline data caching mechanism temporarily stores survey data in a local cache queue when the on-site network environment is unstable, and automatically uploads it to the cloud in order after the network is restored.

3. The biogas engineering big data acquisition and analysis system according to claim 1, characterized in that: The horizontal comparison logic of the data comparison and verification module includes: A standard gas production rate reference database is established according to the type of raw materials. The reference database includes the baseline values ​​of 6.2 m³ / ton for domestic waste, 55 m³ / ton for kitchen waste, 75 m³ / ton for food waste, 350 m³ / ton for dried straw, 90 m³ / ton for chicken manure, 50 m³ / ton for duck manure, 5 m³ / ton for pig manure, and 25 m³ / ton for cow manure. Calculate the mean and standard deviation of gas production rate of the same type of raw material within a preset time window, and mark data that exceed the mean ± 2 times the standard deviation as outliers; For marked abnormal data, a review process is triggered, a verification reminder message is sent to the project research personnel, and the verification results and correction values ​​are recorded.

4. The biogas engineering big data acquisition and analysis system according to claim 1, characterized in that: The method for establishing the regional gas production analysis module includes: Collect basic data on the amount of food waste generated, leachate production, scale of existing treatment facilities, and geographical and climatic conditions in each region. Using the amount of waste as the independent variable and the measured amount of gas produced as the dependent variable, a regional linear regression model was constructed. The model is in the form of: Q=α×M+β×T+γ×P+ε, where Q is the amount of gas produced, M is the amount of waste, T is the fermentation temperature, P is the population density, α, β, and γ are the regression coefficients, and ε is the error term. Based on the model output, a distribution map of gas production per ton of waste in each region is drawn to identify areas with high gas production potential and areas with inefficient operation.

5. The biogas engineering big data acquisition and analysis system according to claim 1, characterized in that: The AI ​​big data prediction module includes: The data preprocessing submodule is used to clean, normalize, and perform feature engineering on the raw data of biogas projects across the country. The gas production potential prediction submodule, based on random forest or gradient boosting tree algorithm, takes as input the feed type, processing scale, climate region, and process type characteristics, and outputs the predicted gas production value and confidence interval. The project benefit evaluation submodule calculates the investment payback period, net present value, and internal rate of return for each project by combining investment costs, operating costs, biogas sales prices, and subsidy policy parameters. The national biogas project big data visualization submodule uses geographic information maps as a carrier to overlay and display multi-dimensional information such as gas production, operating status, and economic benefits of each project.

6. The biogas engineering big data acquisition and analysis system according to claim 1, characterized in that: The collaborative functions of the project file management module include: A role-based access control system distinguishes the operational permissions of project administrators, researchers, technical reviewers, and visitors. The version control mechanism saves historical versions of every modification to the investment plan and equipment list documents, and supports version rollback and difference comparison. The meeting minutes structure extraction function uses natural language processing technology to automatically extract decision-making items, responsible persons, and time nodes from meeting minutes text, and generate a list of tasks to be done.

7. The biogas engineering big data acquisition and analysis system according to any one of claims 1-6 further includes a biogas engineering big data acquisition and analysis method, characterized in that, Includes the following steps: S1. Using the structured form engine on the WeChat mini program, collect data on the source of raw materials, wastewater parameters, fermentation type, project scale, biogas slurry disposal method, biogas composition, utilization method, site facilities and safety compliance at the biogas purification project site. At the same time, collect on-site photos and record GPS location information. S2. The collected data is uploaded to the cloud server via HTTPS encrypted transmission protocol. The cloud server performs integrity verification and format standardization processing on the received data and stores it in the classification database. S3. Perform a horizontal comparison of gas production rate data for similar raw materials. When the detection data deviates from the standard reference range, trigger an anomaly marker and a review reminder. S4. Based on the collected data of kitchen waste and leachate in each region, and combined with regional characteristic parameters, establish a correlation analysis model between waste tonnage and gas production. S5. Utilize machine learning algorithms to conduct in-depth analysis of biogas engineering data nationwide, and generate gas production potential prediction and engineering benefit evaluation results. S6 archives project materials such as investment plans, equipment lists, site photos, and meeting minutes to a cloud archive by project number, supporting multi-user collaborative management and access control.

Images