Straw carbonization intelligent prediction control method based on big data analysis and numerical algorithm

Abstract

The application discloses a straw carbonization intelligent prediction control method based on big data analysis and numerical algorithm, relates to the technical field of prediction and control, and is used for solving the problem of large production cycle prediction deviation. The carbonization prediction duration is called, the carbonization furnace cavity quality is monitored in real time, and the stop duration is recorded. The carbonization allowable duration is calculated based on the stop duration and the prediction duration. The historical cooling termination coefficient and the production cycle are acquired by accessing a cooling record database. The cavity temperature and the biochar quality are collected to calculate the cooling termination coefficient and compare the historical data to obtain the cooling time of different cooling modes. The prediction duration and the cooling time are used to calculate the actual production cycle and compare the actual production cycle with the production cycle to obtain the over-limit cooling number. The optimized production cycle is calculated based on the carbonization allowable duration, and the cooling mode is screened. The cooling start time point is set. The temperature drop amplitude and the furnace pressure fluctuation rate under each cooling mode simulation are collected to determine the decision result of the terminal cooling mode, and the carbonization cycle prediction precision is improved.

Description

Technical Field

[0001] This invention relates to the field of prediction and control technology, and more specifically, to an intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms. Background Technology

[0002] In the process of straw carbonization, the carbonization reaction has significant nonlinear and multivariate coupling characteristics. Its production cycle and cooling efficiency are affected by a variety of factors such as furnace temperature distribution, cavity pressure, and straw moisture content. In the existing technology, the carbonization process control mainly relies on fixed time settings or empirical models for regulation, lacking the ability to comprehensively analyze and predict multi-source data.

[0003] The existing technology has the following shortcomings:

[0004] Currently, existing technologies mainly rely on fixed time settings or manual experience models to control the carbonization process. They lack a unified integration and analysis mechanism for multi-source information such as furnace temperature, cavity pressure, straw moisture content, and historical cooling data. This makes it impossible to perform adaptive carbonization cycle prediction and dynamic control parameter optimization based on real-time monitoring data and historical data. As a result, the production cycle prediction deviation is large, the selection of cooling methods is lagging, and the carbonization process response is not timely. Therefore, a smart predictive control method for straw carbonization based on big data analysis and numerical algorithms is proposed. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms. This method collects real-time operating data of the carbonization furnace and associates it with historical production data to form a dataset. It then uses algorithms such as statistical analysis and multinomial regression to construct a prediction and calculation model, and performs closed-loop optimization of carbonization cycle prediction and cooling control to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms, comprising the following steps:

[0007] Step S1: When straw is fed into the carbonization furnace, the carbonization prediction time is called up, the quality of the carbonization furnace cavity is monitored in real time, and the stop time is recorded when the quality of the carbonization furnace cavity stops changing. The allowable carbonization time is calculated based on the stop time and the carbonization prediction time.

[0008] Step S2: Access the historical cooling termination coefficient and production cycle of the cooling record library, collect the temperature of the inner cavity of the carbonization furnace and the quality of biochar to calculate the cooling termination coefficient, and compare the cooling termination coefficient with the historical cooling termination coefficient to obtain the cooling time of different cooling methods.

[0009] Step S3: Calculate the actual production cycle by combining the predicted carbonization time with the cooling time of different cooling methods and compare it with the production cycle to obtain the number of times the cooling exceeds the limit. Analyze whether to activate the end-of-line cooling treatment of carbonization based on the number of times the cooling exceeds the limit.

[0010] Step S4: Calculate the optimized production cycle by combining the actual production cycle with the allowable carbonization time. Compare the optimized production cycle with the production cycle to select the cooling method. Set the cooling activation time point and collect the simulated temperature drop and furnace pressure fluctuation rate of each cooling method to determine the decision result of the end cooling method.

[0011] In a preferred embodiment, in step S1, when straw is fed into the carbonization furnace, the carbonization time prediction model is used to call the carbonization time prediction.

[0012] Set a monitoring time to monitor the quality of the carbonization furnace cavity in real time. When the quality stops changing within the monitoring time, start recording the number of subsequent monitoring times to obtain the total number of monitoring times.

[0013] When recording the number of subsequent monitoring times, the total number of monitoring times is compared and analyzed with the preset monitoring threshold.

[0014] If the total number of monitored times exceeds the monitoring threshold, the current time point will be marked as the stop time point;

[0015] If the total number of monitoring times is less than the monitoring threshold, then continue recording subsequent monitoring times.

[0016] In a preferred embodiment, in step S1, the starting time of the carbonization process when the straw was input into the carbonization furnace is retrieved from the carbonization record library, and the difference between the stopping time and the starting time of the carbonization process is calculated to obtain the stopping time.

[0017] The permissible carbonization time is obtained by calculating the difference between the predicted carbonization time and the stopping time.

[0018] In a preferred embodiment, in step S2, the cooling record library is accessed, and the historical cooling termination coefficient and production cycle of the cooling record library are retrieved;

[0019] The temperature of the carbonization furnace cavity and the quality of biochar were collected at the stop time point.

[0020] Infrared temperature sensors are installed at various detection points inside the carbonization furnace to collect the temperature of each detection point in real time. The temperature of each detection point is accumulated and the ratio is calculated with the total number of detection points to obtain the temperature of the cavity inside the carbonization furnace.

[0021] The mass of biochar produced from straw carbonization is obtained by measuring the mass of the biochar produced by carbonization using a weighing sensor installed at the bottom of the carbonization reaction vessel in the carbonization furnace.

[0022] In a preferred embodiment, in step S2, the temperature of the inner cavity of the carbonization furnace and the quality of biochar are standardized and weighted summation is performed to obtain the cooling termination coefficient.

[0023] Compare and analyze the cooling termination coefficient with historical cooling termination coefficients;

[0024] Select the cooling time of the cooling method corresponding to the historical cooling termination coefficient that is consistent with the cooling termination coefficient value, and calculate the average cooling time of each cooling method to obtain the cooling time of different cooling methods.

[0025] In a preferred embodiment, in step S3, the carbonization prediction time is added to the cooling time of different cooling methods to calculate the actual production cycle corresponding to different cooling methods;

[0026] Compare and analyze the actual production cycle with the production cycle;

[0027] If the actual production cycle exceeds the production cycle, the corresponding cooling method will be counted as one over-limit cooling.

[0028] If the actual production cycle is shorter than the production cycle, the corresponding cooling method will not be counted as an over-limit cooling operation.

[0029] The number of times cooling methods were marked as exceeding the limit was obtained.

[0030] In a preferred embodiment, in step S3, the number of cooling cycles exceeding the limit is compared and analyzed with a preset exceeding threshold.

[0031] If the number of cooling cycles exceeds the limit threshold, then end-of-line cooling treatment for carbonization will be activated.

[0032] If the number of over-limit cooling cycles is less than the over-limit threshold, the carbonization end cooling process will not be activated.

[0033] In a preferred embodiment, in step S4, the difference between the actual production cycle corresponding to different cooling methods and the allowable carbonization time is calculated to obtain the optimized production cycle;

[0034] The optimized production cycle will be compared and analyzed with the original production cycle.

[0035] If the optimized production cycle exceeds the production cycle, the corresponding cooling method will be deleted.

[0036] If the optimized production cycle is shorter than the production cycle, the corresponding cooling method will be retained.

[0037] The stop time point is taken as the cooling start time point. After the cooling start time point, the temperature drop rate and furnace pressure fluctuation rate of each cooling method are collected.

[0038] In a preferred embodiment, in step S4, simulated temperature detection points are preset in the carbonization furnace cavity. The same initial temperature conditions are set for each cooling method, and the temperature change value of each detection point is calculated according to the set simulation period. The temperature changes of all points are accumulated, and the average value is taken to obtain the average temperature of the cavity during the simulation period. After all simulation periods are completed, the difference between the initial temperature and the average value of the average cavity temperature during all simulation periods is calculated to obtain the temperature drop amplitude simulated by each cooling method.

[0039] Set the total number of simulated pressure detections during all simulation periods, preset multiple pressure detection points and capture the pressure change value of each pressure detection point, compare the pressure change value with the preset pressure threshold, count the pressure change value exceeding the pressure threshold as the pressure change number and calculate the ratio with the total number of simulated pressure detections to obtain the furnace pressure fluctuation rate simulated by each cooling method.

[0040] The temperature drop rate and the pressure fluctuation rate inside the furnace are standardized and substituted into the polynomial regression formula to calculate the end cooling coefficient.

[0041] The terminal cooling coefficient of each cooling method was calculated by substituting the simulated temperature drop and furnace pressure fluctuation rate into the polynomial regression formula.

[0042] In a preferred embodiment, in step S4, the end cooling coefficient of each cooling method is compared and analyzed with a preset end threshold.

[0043] If the terminal cooling coefficient exceeds the terminal threshold, the corresponding cooling method is selected as the terminal cooling method, and the decision result of the terminal cooling method is obtained;

[0044] If the terminal cooling coefficient is lower than the terminal threshold, the corresponding cooling method will be deleted.

[0045] If there are multiple decision results for end-cooling methods, the cooling time of each end-cooling method is sorted, and the end-cooling method with the shortest time is selected as the end-cooling method to be executed.

[0046] The technical effects and advantages of this invention are as follows:

[0047] This invention monitors the carbonization furnace cavity quality in real time by calling the carbonization prediction duration. When the furnace cavity quality stops changing, the duration of the stoppage is recorded. Based on the stoppage duration and the carbonization prediction duration, the permissible carbonization duration is calculated. Historical cooling termination coefficients and production cycles are accessed from the cooling record library. The furnace cavity temperature and biochar quality are collected to calculate the cooling termination coefficient. This coefficient is compared with historical coefficients to obtain the cooling time for different cooling methods. The actual production cycle is calculated using the carbonization prediction duration and the cooling time for different methods, and compared with the actual production cycle to obtain the number of times cooling exceeds the limit. The number of times cooling exceeds the limit is used to analyze whether to activate the system. The end-of-line cooling process in carbonization calculates the optimized production cycle by comparing the actual production cycle with the allowable carbonization time. The optimized production cycle is then compared with the actual production cycle to select the appropriate cooling method. The cooling activation time is set, and the simulated temperature drop and furnace pressure fluctuation rate of each cooling method are collected to determine the final cooling method. This data-driven intelligent prediction and control system for carbonization processes improves the accuracy of carbonization cycle prediction, automates the selection of cooling methods, shortens control decision response time, and builds closed-loop data control capabilities. This reduces production cycle fluctuations and cooling energy consumption caused by prediction errors, achieving intelligent and highly reliable production processes. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the implementation of the intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to the present invention.

[0049] Figure 2 This is a schematic diagram illustrating the steps of the intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms of the present invention. Detailed Implementation

[0050] 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. 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.

[0051] This invention improves the accuracy of carbonization cycle prediction by calling the carbonization prediction time and monitoring the quality of the carbonization furnace cavity in real time. When the quality of the carbonization furnace cavity stops changing, the stopping time is recorded. Based on the stopping time and the carbonization prediction time, the allowable carbonization time is calculated. Historical cooling termination coefficients and production cycles are accessed from the cooling record library. The temperature of the carbonization furnace cavity and the quality of biochar are collected to calculate the cooling termination coefficient. The cooling termination coefficient is compared with the historical cooling termination coefficient to obtain the cooling time for different cooling methods. The actual production cycle is calculated by comparing the carbonization prediction time with the cooling time of different cooling methods and compared with the production cycle to obtain the number of times of exceeding the limit cooling. Based on the number of times of exceeding the limit cooling, it is analyzed whether to activate the end-of-line cooling treatment of carbonization. The optimized production cycle is calculated by comparing the actual production cycle with the allowable carbonization time. The optimized production cycle is compared with the production cycle to select the cooling method. The cooling activation time point is set, and the simulated temperature drop and furnace pressure fluctuation rate of each cooling method are collected to determine the decision result of the end-of-line cooling method. This improves the accuracy of carbonization cycle prediction and reduces the production cycle fluctuation and cooling energy consumption caused by prediction deviation.

[0052] Example 1

[0053] Please see Figures 1 to 2 The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms has the following specific operation process:

[0054] Step S1: When straw is fed into the carbonization furnace, the carbonization prediction time is called up, the quality of the carbonization furnace cavity is monitored in real time, and the stop time is recorded when the quality of the carbonization furnace cavity stops changing. The allowable carbonization time is calculated based on the stop time and the carbonization prediction time.

[0055] Step S2: Access the historical cooling termination coefficient and production cycle of the cooling record library, collect the temperature of the inner cavity of the carbonization furnace and the quality of biochar to calculate the cooling termination coefficient, and compare the cooling termination coefficient with the historical cooling termination coefficient to obtain the cooling time of different cooling methods.

[0056] Step S3: Calculate the actual production cycle by combining the predicted carbonization time with the cooling time of different cooling methods and compare it with the production cycle to obtain the number of times the cooling exceeds the limit. Analyze whether to activate the end-of-line cooling treatment of carbonization based on the number of times the cooling exceeds the limit.

[0057] Step S4: Calculate the optimized production cycle by combining the actual production cycle with the allowable carbonization time. Compare the optimized production cycle with the production cycle to select the cooling method. Set the cooling activation time point and collect the simulated temperature drop and furnace pressure fluctuation rate of each cooling method to determine the decision result of the end cooling method.

[0058] The specific process is as follows:

[0059] In step S1, when straw is fed into the carbonization furnace, the carbonization time prediction model is used to call the carbonization time prediction.

[0060] Among them, the carbonization time prediction model is a prediction model built based on historical carbonization batches and real-time furnace cavity characteristic parameters. Its prediction process is to extract features from the furnace cavity temperature change curve, cavity pressure change curve, input straw mass, moisture content and corresponding actual carbonization completion time recorded in historical carbonization batches, and obtain the carbonization prediction time through multiple regression and time series prediction algorithms. The specific parameter collection and calculation process required for prediction are common knowledge of the experimenters and will not be elaborated here.

[0061] Set a monitoring time to monitor the quality of the carbonization furnace cavity in real time. When the quality stops changing within the monitoring time, start recording the number of subsequent monitoring times to obtain the total number of monitoring times.

[0062] It should be noted that the monitoring time was set by the researchers based on the thermal stability time of the carbonization furnace cavity. Specifically, it is a continuous and short monitoring time. When the mass of the carbonization furnace cavity stops changing, the monitoring time is still executed and the number of monitoring times is recorded. For example, if the monitoring time is 8 seconds, when 200 monitoring times have passed, the mass stops changing in the 201st monitoring time, and the number of subsequent monitoring times is recorded. For example, if the mass does not change in the next 20 monitoring times, the total number of monitoring times is 20.

[0063] When recording the number of subsequent monitoring times, the total number of monitoring times is compared and analyzed with the preset monitoring threshold.

[0064] If the total number of monitored times exceeds the monitoring threshold, the current time point will be marked as the stop time point;

[0065] If the total number of monitoring times is lower than the monitoring threshold, the number of subsequent monitoring times will continue to be recorded as the total number of monitoring times and compared with the monitoring threshold again, until the monitoring threshold is exceeded.

[0066] It should be noted that the monitoring threshold was set by the researchers based on the thermal inertia parameters of the carbonization furnace cavity and the gas evolution rate during straw carbonization. This ensures that the monitoring system has a sufficient time window to identify the stable state of carbonization quality, so as to eliminate false stability caused by instantaneous gas evolution or combustion disturbance, thereby ensuring the accuracy of the stop time point judgment. This will not be elaborated here.

[0067] Understandably, the total number of monitoring times increases as the monitoring time progresses. Generally, the carbonization process ends when the mass of the carbonization furnace cavity stops changing.

[0068] Furthermore, if the cumulative value of the monitoring time exceeds the carbonization prediction time, the prediction is considered accurate, and the method process ends.

[0069] The starting time of the carbonization process when straw is input into the carbonization furnace is retrieved from the carbonization record library. The difference between the stopping time and the starting time of the carbonization process is calculated to obtain the stopping time.

[0070] Specifically, the carbonization record library is a historical dataset used to store process data of the entire straw carbonization process, including process data such as the start time of the carbonization process. It also serves as the input source for the carbonization prediction model, providing multiple batches of historical data to support carbonization duration prediction and allowable duration optimization analysis.

[0071] The permissible carbonization time is obtained by calculating the difference between the predicted carbonization time and the stopping time.

[0072] Among them, the permissible carbonization time is the length of time during which the carbonization process is allowed to be cooled in advance, which is determined based on the difference between the predicted carbonization time and the stopping time.

[0073] In step S2, the cooling record library is accessed, and the historical cooling termination coefficient and production cycle are retrieved from the cooling record library;

[0074] Among them, the cooling record library is a database used to store the historical operating parameters, cooling performance indicators and corresponding production cycle data of the carbonization furnace under different cooling methods;

[0075] It should be noted that the historical cooling termination coefficient is the result of a comprehensive analysis of the temperature of the carbonization furnace cavity and the quality of biochar collected at the historical stop time. Different cooling methods are used for cooling based on the analysis results. The production cycle is the standard production time required for the entire straw carbonization process (including the carbonization reaction stage and the cooling stage).

[0076] The temperature of the carbonization furnace cavity and the quality of biochar were collected at the stop time point.

[0077] The logic for obtaining the temperature of the inner cavity of the carbonization furnace is to collect the temperature of each detection point in the cavity in real time by using infrared temperature sensors installed at various detection points in the carbonization furnace. The temperature of each detection point is accumulated and the ratio is calculated with the total number of detection points to obtain the temperature of the inner cavity of the carbonization furnace.

[0078] Specifically, the infrared temperature sensor is a high-temperature temperature measuring device based on the principle of non-contact infrared radiation. It detects the infrared radiation intensity of the inner wall of the carbonization furnace cavity and the surface of the material and converts it into a corresponding temperature signal to achieve real-time monitoring of the cavity temperature. The setting of the detection points is determined by the researchers based on the thermal field distribution of the carbonization furnace cavity structure and the straw accumulation density. There is no limit to the number of detection points, which will not be elaborated here.

[0079] The biochar quality is obtained by measuring the weight of the biochar produced after the straw has been carbonized using a weighing sensor installed at the bottom of the carbonization reaction vessel in the carbonization furnace.

[0080] Among them, the carbonization reaction vessel is a high-temperature resistant container used to support straw raw materials and complete the pyrolysis reaction in the carbonization furnace; the weighing sensor is a measuring element used to convert the force signal of the carbonization reaction vessel and the materials therein into an electrical signal; and biochar is a carbon-rich solid product generated by the pyrolysis reaction (carbonization treatment) of straw under oxygen-isolated conditions.

[0081] The temperature inside the carbonization furnace and the quality of biochar are standardized so that the temperature inside the carbonization furnace and the quality of biochar are under the same dimension and the values ​​are expressed between 0 and 1.

[0082] It should be noted that the standardization methods include, but are not limited to, standard linear transformation based on interval scaling, statistical Z-Score standardization method, or normalization method based on nonlinear mapping function. The application methods of standardization will not be elaborated here.

[0083] The cooling termination coefficient is obtained by weighted summation of the temperature inside the carbonization furnace cavity and the biochar mass after standardization.

[0084] The specific calculation formula is as follows:

[0085] ;

[0086] In the formula, The cooling termination coefficient, The temperature of the carbonization furnace cavity after standardization treatment. To standardize the quality of biochar after processing, and These are weight parameters;

[0087] It should be noted that when the temperature inside the carbonization furnace cavity and the mass of biochar are higher, it means that the furnace cavity heat is higher and the volume of carbonized material is larger, which increases the difficulty of cooling. This means that the required cooling time is longer and the cooling efficiency requirement is higher. The larger the cooling termination coefficient, the longer and more efficient the cooling method is needed to cool the biochar inside the carbonization furnace and the entire furnace cavity.

[0088] Compare and analyze the cooling termination coefficient with historical cooling termination coefficients;

[0089] Select the cooling time of the cooling method corresponding to the historical cooling termination coefficient that is consistent with the cooling termination coefficient value, and calculate the average cooling time of each cooling method to obtain the cooling time of different cooling methods.

[0090] A specific example is as follows: For instance, there are six historical cooling termination coefficients that are consistent with the cooling termination coefficient value, namely A1, A2, A3, A4, A5, and A6. The cooling methods of the six are natural cooling, water cooling, and inert gas cooling, respectively. Among them, A1 and A2 are water cooling, A3, A4, and A5 are inert gas cooling, and A6 is natural cooling. Then, the cooling times corresponding to A1 and A2 are added together and the ratio is calculated to obtain the cooling time of water cooling. Then, the cooling times corresponding to A3, A4, and A5 are added together and the ratio is calculated to obtain the cooling time of inert gas cooling. Since natural cooling has one and only one historical cooling termination coefficient, the cooling time of natural cooling is consistent with the cooling time corresponding to A6.

[0091] Furthermore, if there is no historical cooling termination coefficient that matches the cooling termination coefficient value, then the current cooling termination coefficient does not appear in the historical records, and there is a lack of reference cooling time data. The message "No matching item" can be generated and transmitted to the visualization port, and the method flow can be terminated.

[0092] In step S3, the predicted carbonization time is added to the cooling time of different cooling methods to calculate the actual production cycle corresponding to different cooling methods;

[0093] Specifically, the actual production cycle is the total production time obtained by adding the predicted carbonization time to the cooling time of the selected cooling method. It is used to reflect the actual time required to complete a complete carbonization to cooling process. By comparing it with the production cycle, it is possible to determine whether there is a time limit exceeding the limit, thereby ensuring production efficiency and process safety.

[0094] Compare and analyze the actual production cycle with the production cycle;

[0095] If the actual production cycle exceeds the production cycle, the corresponding cooling method will be counted as one over-limit cooling.

[0096] If the actual production cycle is shorter than the production cycle, the corresponding cooling method will not be counted as an over-limit cooling operation.

[0097] The number of times cooling methods were marked as exceeding the limit was obtained.

[0098] It should be noted that the number of times cooling exceeds the limit is the number of cooling methods whose actual production cycle exceeds the production cycle after comparing and analyzing the actual production cycle corresponding to different cooling methods. It is used to reflect the number of cooling schemes that have exceeded the production cycle limit under the current carbonization batch conditions. The larger the number, the more likely it is that under the current straw carbonization conditions, the actual production cycle of most cooling methods exceeds the preset production cycle, which poses a high risk of production delay. It is necessary to use end-of-carbonization cooling treatment, that is, to carry out cooling treatment within the allowable carbonization time.

[0099] The number of cooling cycles exceeding the limit is compared and analyzed with the preset over-limit threshold.

[0100] If the number of cooling cycles exceeds the limit threshold, then end-of-line cooling treatment for carbonization will be activated.

[0101] If the number of over-limit cooling cycles is less than the over-limit threshold, the carbonization end cooling treatment will not be activated.

[0102] It should be noted that the preset over-limit threshold was set by the researchers based on the production cycle tolerance range and the process response rate of the cooling equipment, and will not be elaborated here.

[0103] Furthermore, when the end-of-carbonization cooling treatment is not activated, the corresponding cooling method (i.e., the conventional cooling method) can be selected after the carbonization prediction time is completed. The corresponding cooling method is the cooling method that is not counted as an over-limit cooling.

[0104] In step S4, the difference between the actual production cycle corresponding to different cooling methods and the allowable carbonization time is calculated to obtain the optimized production cycle;

[0105] Among them, the optimized production cycle refers to the production cycle required if cooling is performed at the stop time. Specifically, the optimized production cycle is shorter than the actual production cycle (i.e., cooling is performed in advance by utilizing the allowable carbonization time). This indicates that by intervening in the cooling operation at the end of carbonization, the overall production cycle can be effectively shortened while ensuring sufficient carbonization, thereby meeting the production rate requirements of the production cycle.

[0106] The optimized production cycle will be compared and analyzed with the original production cycle.

[0107] If the optimized production cycle exceeds the production cycle, the corresponding cooling method will be deleted.

[0108] If the optimized production cycle is shorter than the production cycle, the corresponding cooling method will be retained.

[0109] The stop time point is taken as the cooling start time point. After the cooling start time point, the temperature drop rate and furnace pressure fluctuation rate of each cooling method are collected.

[0110] It is understandable that a single cooling method is usually selected during the cooling process of the carbonization furnace. Due to the instability of the pressure inside the furnace, using multiple cooling methods at the same time can easily cause superimposed gas pressure disturbances in the furnace cavity, uneven distribution of cooling medium, and distortion of heat exchange, resulting in distorted cooling data and unstable stress in the carbonization furnace structure.

[0111] Therefore, in order to avoid experimental errors and equipment safety risks caused by parallel cooling of multiple cooling methods, this invention simulates the cooling process of each cooling method under the same initial conditions to obtain the temperature drop range and furnace pressure fluctuation rate corresponding to each method. Thus, using virtual simulation to collect cooling data is a well-known alternative method for those skilled in the art when it is impossible to conduct actual measurements of multiple cooling methods, and will not be elaborated here.

[0112] The simulated temperature drop for each cooling method refers to the temperature drop in the furnace during the simulated cooling process for different cooling methods (such as natural cooling, water cooling, and inert gas cooling) in a virtual carbonization furnace environment. The acquisition logic is to preset simulation temperature detection points in the carbonization furnace cavity, set the same initial temperature conditions for each cooling method, calculate the temperature change value of each detection point according to the set simulation period, and accumulate the temperature changes of all points to obtain the average cavity temperature during the simulation period. After all simulation periods are completed, the difference between the initial temperature and the average cavity temperature of all simulation periods is calculated to obtain the simulated temperature drop for each cooling method.

[0113] It should be noted that the simulation of the cooling method is based on the heat conduction and convection heat transfer model of the carbonization furnace. The model type, boundary conditions and thermal property parameters are not limited and will not be elaborated here.

[0114] Furthermore, the preset simulation temperature detection points inside the carbonization furnace cavity are set by the researchers based on the furnace structure and heat distribution intensity. The simulation period is set according to the total simulation time. The specific number of simulation periods is not limited and will not be elaborated here.

[0115] Furthermore, a specific calculation example is as follows: If there are three temperature detection points in the simulation, namely Ea, Eb, and Ec, and the initial temperature is Ta, and the simulation period is three, then the temperature change values ​​of each detection point in the first, second, and third simulation periods are Ea1, Eb1, and Ec1, Ea2, Eb2, and Ec2, and Ea3, Eb3, and Ec3, respectively. The average temperature change values ​​in the first, second, and third simulation periods are (Ea1+Eb1+Ec1) / 3, (Ea2+Eb2+Ec2) / 3, and (Ea3+Eb3+Ec3) / 3. Therefore, the temperature drop is Ta - [(Ea1+Eb1+Ec1) / 3 + (Ea2+Eb2+Ec2) / 3 + (Ea3+Eb3+Ec3) / 3] / 3;

[0116] The logic for obtaining the simulated furnace pressure fluctuation rate for each cooling method is as follows: set the total number of simulated pressure detections during all simulation periods, preset multiple pressure detection points and capture the pressure change value of each pressure detection point, compare the pressure change value with the preset pressure threshold, count the pressure change value exceeding the pressure threshold as the pressure change number, and calculate the ratio with the total number of simulated pressure detections to obtain the simulated furnace pressure fluctuation rate for each cooling method.

[0117] It should be noted that the total number of simulated pressure detection points was set based on the furnace cavity volume and the thermal conductivity of the cooling medium. The pressure detection points were set by dividing the furnace cavity geometric model into equal parts in space. The preset pressure threshold was set by the researchers based on the allowable safe pressure range inside the furnace, which will not be elaborated here.

[0118] Specifically, the time interval between two adjacent simulated pressure tests is not limited and is set by the researchers based on the steady-state duration and transient response rate of the cooling process, which will not be elaborated here.

[0119] The temperature drop rate and the furnace pressure fluctuation rate are standardized so that the temperature drop rate and the furnace pressure fluctuation rate are on the same dimension and the numerical expression range is between 0 and 1.

[0120] It should be noted that the standardization process has been described in the above embodiments and will not be repeated here;

[0121] The standardized temperature drop and furnace pressure fluctuation rate are substituted into a multinomial regression formula to calculate the terminal cooling coefficient, as shown in the following formula:

[0122] ;

[0123] In the formula, The terminal cooling coefficient, The temperature drop after standardization treatment. The standardized furnace pressure fluctuation rate. To adjust the parameters, as well as The weighting coefficients corresponding to the temperature drop rate and the furnace pressure fluctuation rate after standardization.

[0124] It should be noted that the greater the temperature drop, the stronger the current cooling method's ability to conduct and dissipate heat in the furnace cavity, the higher the cooling efficiency, and the better the heat exchange stability. In this case, the terminal cooling coefficient is larger, and the current cooling method is more likely to be selected as the terminal cooling method. Conversely, the greater the pressure fluctuation rate inside the furnace, the more obvious the uneven gas flow, thermal stress disturbance, or transient pressure change generated by the current cooling method during the cooling process, and the worse the cooling stability. In this case, the terminal cooling coefficient is smaller, and the current cooling method is less suitable as the terminal cooling method.

[0125] Furthermore, the simulated temperature drop and furnace pressure fluctuation rate of each cooling method were substituted into the polynomial regression formula to calculate the terminal cooling coefficient of each cooling method.

[0126] The terminal cooling coefficients of each cooling method are compared and analyzed with the preset terminal threshold.

[0127] If the terminal cooling coefficient exceeds the terminal threshold, the corresponding cooling method is selected as the terminal cooling method, and the decision result of the terminal cooling method is obtained;

[0128] If the terminal cooling coefficient is lower than the terminal threshold, the corresponding cooling method will be deleted.

[0129] It should be explained that the end threshold was set by our researchers based on the thermal inertia characteristics of the carbonization furnace cavity and the historical cooling time, which will not be elaborated here.

[0130] Furthermore, if there are multiple decision results for end cooling methods, the cooling time of each end cooling method is sorted, and the end cooling method with the shortest time is selected as the end cooling method to be executed.

[0131] Furthermore, if there is no decision result regarding the end-cooling method, a message stating "The current production cycle cannot be met" will be generated and sent to the visualization port to prompt the operators.

[0132] It should be explained that the end-of-line cooling method is the end-of-line cooling method executed at the cooling activation time point, which is used to meet the needs of the current production cycle and reduce the frequency of production cycle adjustment. Specifically, in actual application scenarios, due to calculation errors, the researchers in this experiment can correct the cooling activation time point by setting the cooling activation time point correction factor and the temperature drop rate correction coefficient (this example assumes an ideal state, that is, the calculation is completed when the stop time point is received, and the default stop time point is the cooling activation time point), so that it meets the actual application adjustment.

[0133] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0134] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0135] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0136] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0137] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent predictive control of straw carbonization based on big data analysis and numerical algorithms, characterized by: Includes the following steps: Step S1: When straw is fed into the carbonization furnace, the carbonization prediction time is called up, the quality of the carbonization furnace cavity is monitored in real time, and the stop time is recorded when the quality of the carbonization furnace cavity stops changing. The allowable carbonization time is calculated based on the stop time and the carbonization prediction time. Step S2: Access the historical cooling termination coefficient and production cycle of the cooling record library, collect the temperature of the inner cavity of the carbonization furnace and the quality of biochar to calculate the cooling termination coefficient, and compare the cooling termination coefficient with the historical cooling termination coefficient to obtain the cooling time of different cooling methods. Step S3: Calculate the actual production cycle by combining the predicted carbonization time with the cooling time of different cooling methods and compare it with the production cycle to obtain the number of times the cooling exceeds the limit. Analyze whether to activate the end-of-line cooling treatment of carbonization based on the number of times the cooling exceeds the limit. Step S4: Calculate the optimized production cycle by combining the actual production cycle with the allowable carbonization time. Compare the optimized production cycle with the production cycle to select the cooling method. Set the cooling activation time point and collect the simulated temperature drop and furnace pressure fluctuation rate of each cooling method to determine the decision result of the end cooling method.

2. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 1, characterized in that: In step S1, when straw is fed into the carbonization furnace, the carbonization time prediction model is used to call the carbonization time prediction. Set a monitoring time to monitor the quality of the carbonization furnace cavity in real time. When the quality stops changing within the monitoring time, start recording the number of subsequent monitoring times to obtain the total number of monitoring times. When recording the number of subsequent monitoring times, the total number of monitoring times is compared and analyzed with the preset monitoring threshold. If the total number of monitored times exceeds the monitoring threshold, the current time point will be marked as the stop time point; If the total number of monitoring times is less than the monitoring threshold, then continue recording subsequent monitoring times.

3. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 2, characterized in that: In step S1, the starting time of the carbonization process when the straw was input into the carbonization furnace is retrieved from the carbonization record library, and the difference between the stopping time and the starting time of the carbonization process is calculated to obtain the stopping time. The permissible carbonization time is obtained by calculating the difference between the predicted carbonization time and the stopping time.

4. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 1, characterized in that: In step S2, the cooling record library is accessed, and the historical cooling termination coefficient and production cycle are retrieved from the cooling record library; The temperature of the carbonization furnace cavity and the quality of biochar were collected at the stop time point. Infrared temperature sensors are installed at various detection points inside the carbonization furnace to collect the temperature of each detection point in real time. The temperature of each detection point is accumulated and the ratio is calculated with the total number of detection points to obtain the temperature of the cavity inside the carbonization furnace. The mass of biochar produced from straw carbonization is obtained by measuring the mass of the biochar produced by carbonization using a weighing sensor installed at the bottom of the carbonization reaction vessel in the carbonization furnace.

5. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 4, characterized in that: In step S2, the temperature of the inner cavity of the carbonization furnace and the quality of biochar are standardized and weighted summation is performed to obtain the cooling termination coefficient. Compare and analyze the cooling termination coefficient with historical cooling termination coefficients; Select the cooling time of the cooling method corresponding to the historical cooling termination coefficient that is consistent with the cooling termination coefficient value, and calculate the average cooling time of each cooling method to obtain the cooling time of different cooling methods.

6. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 1, characterized in that: In step S3, the predicted carbonization time is added to the cooling time of different cooling methods to calculate the actual production cycle corresponding to different cooling methods; Compare and analyze the actual production cycle with the production cycle; If the actual production cycle exceeds the production cycle, the corresponding cooling method will be counted as one over-limit cooling. If the actual production cycle is shorter than the production cycle, the corresponding cooling method will not be counted as an over-limit cooling operation. The number of times cooling methods were marked as exceeding the limit was obtained.

7. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 6, characterized in that: In step S3, the number of cooling cycles exceeding the limit is compared and analyzed with the preset threshold for exceeding the limit. If the number of cooling cycles exceeds the limit threshold, then end-of-line cooling treatment for carbonization will be activated. If the number of over-limit cooling cycles is less than the over-limit threshold, the carbonization end cooling process will not be activated.

8. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 1, characterized in that: In step S4, the difference between the actual production cycle corresponding to different cooling methods and the allowable carbonization time is calculated to obtain the optimized production cycle; The optimized production cycle will be compared and analyzed with the original production cycle. If the optimized production cycle exceeds the production cycle, the corresponding cooling method will be deleted. If the optimized production cycle is shorter than the production cycle, the corresponding cooling method will be retained. The stop time point is taken as the cooling start time point. After the cooling start time point, the temperature drop rate and furnace pressure fluctuation rate of each cooling method are collected.

9. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 8, characterized in that: In step S4, simulated temperature detection points are preset in the carbonization furnace cavity. The same initial temperature conditions are set for each cooling method, and the temperature change value of each detection point is calculated according to the set simulation period. The temperature changes of all points are accumulated, and the average value is taken to obtain the average temperature of the cavity during the simulation period. After all simulation periods are completed, the difference between the initial temperature and the average value of the average cavity temperature during all simulation periods is calculated to obtain the temperature drop of each cooling method. Set the total number of simulated pressure detections during all simulation periods, preset multiple pressure detection points and capture the pressure change value of each pressure detection point, compare the pressure change value with the preset pressure threshold, count the pressure change value exceeding the pressure threshold as the pressure change number and calculate the ratio with the total number of simulated pressure detections to obtain the furnace pressure fluctuation rate simulated by each cooling method. The temperature drop rate and the pressure fluctuation rate inside the furnace are standardized and substituted into the polynomial regression formula to calculate the end cooling coefficient. The terminal cooling coefficient of each cooling method was calculated by substituting the simulated temperature drop and furnace pressure fluctuation rate into the polynomial regression formula.

10. The intelligent predictive control method for straw carbonization based on big data analysis and numerical algorithms according to claim 9, characterized in that: In step S4, the terminal cooling coefficient of each cooling method is compared and analyzed with the preset terminal threshold. If the terminal cooling coefficient exceeds the terminal threshold, the corresponding cooling method is selected as the terminal cooling method, and the decision result of the terminal cooling method is obtained; If the terminal cooling coefficient is lower than the terminal threshold, the corresponding cooling method will be deleted. If there are multiple decision results for end cooling methods, the cooling time of each end cooling method is sorted, and the end cooling method with the shortest time is selected as the end cooling method to be executed.

Images