A method and system for data acquisition and accuracy calibration of guide cover adjustment

Abstract

This invention provides a method and system for data acquisition and precision calibration of a guide shield adjustment process, relating to the field of automation control technology. The method includes: acquiring real-time height and position data of the guide shield during the saccharification and boiling process; combining this data with synchronously acquired process parameters to generate an initial state vector with process identifiers; comparing this initial state vector with a target state vector in a preset process formula to extract initial deviation values ​​in each dimension, forming a control deviation vector; determining the positions of multiple key sections of the guide shield based on the control deviation vector, and acquiring flow rate and temperature parameters at each key section to form a set of control base point data with spatial coordinates; and mapping the control base point data to a three-dimensional feature space composed of height, flow rate, and temperature to construct multiple dynamic control prisms. This invention enables accurate data acquisition and high-precision closed-loop calibration during the guide shield adjustment process.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, and in particular to a method and system for data acquisition and accuracy calibration of guide cover adjustment. Background Technology

[0002] Wort saccharification and boiling is a core process in beer production. The guide hood, a key component within the boiling kettle for regulating wort circulation, heat exchange, and evaporation separation, directly determines boiling uniformity, off-flavor removal efficiency, and energy consumption. In automated saccharification production lines, the guide hood needs dynamic adjustment based on real-time process conditions, along with data acquisition and precision calibration, to stably match the boiling process requirements of different formulations and batches.

[0003] Existing guide vane adjustment and data calibration technologies generally employ control logic of single-point height acquisition plus one-dimensional deviation adjustment. They rely solely on the difference between the guide vane height and the target setpoint for open-loop or simple closed-loop correction, often facing the following technical shortcomings: They fail to incorporate multi-source parameters such as height, flow velocity, and temperature into a unified space for coupled modeling and quantitative evaluation. This deficiency leads to a series of related problems. The control deviation only reflects single-point height differences, failing to effectively characterize the flow field and temperature gradient distortion of wort at different cross-sections of the guide vane. The adjustment action does not match the actual process deviation, resulting in low height adjustment accuracy and lag. Furthermore, the lack of a spatially characteristic-based quantitative evaluation and closed-loop calibration mechanism for control effects means that the model and control rules cannot be self-corrected after adjustment. Over long-term operation, the cumulative errors in data acquisition and position control continue to increase, making it difficult to meet the requirements of high-precision and stable industrial boiling production. Summary of the Invention

[0004] This invention provides a method and system for data acquisition and accuracy calibration of guide cover adjustment, which can realize accurate data acquisition and high-precision closed-loop calibration during the guide cover adjustment process.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] Firstly, a method for acquiring and calibrating the accuracy of guide cover adjustment data, the method comprising:

[0007] Real-time height position data of the guide hood during the saccharification and boiling process is collected, and combined with the synchronously acquired process parameters, an initial state vector with process identifier is generated for regulation.

[0008] The initial state vector of regulation is compared with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector.

[0009] The positions of multiple key sections of the guide shroud are determined based on the control deviation vector, and flow velocity and temperature parameters are collected at each key section to form a set of control base point data with spatial coordinates;

[0010] The control base point data is mapped to a three-dimensional feature space consisting of height, flow rate and temperature to construct multiple dynamic control prisms. In this invention, they are specifically dynamic control triangular prisms, and the vertex of each dynamic control prism corresponds to the coordinates of a control base point.

[0011] Calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space to obtain the control distortion index, which characterizes the degree of fit between the current control state and the target state.

[0012] The control distortion index is compared with a preset threshold. If it exceeds the threshold, a dynamic compensation amount for the height of the guide cover is derived based on the control distortion index to correct the control deviation vector and generate an optimized control command.

[0013] The guide cover is driven to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

[0014] Secondly, the guide cover adjustment data acquisition and accuracy calibration system includes:

[0015] The status acquisition and fusion module is used to acquire real-time height position data of the guide hood during the saccharification and boiling process, and combine it with synchronously acquired process parameters to generate an initial control state vector with process identifier.

[0016] The deviation vector calculation module is used to compare the initial state vector of regulation with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector.

[0017] The critical section and base point acquisition module is used to determine the positions of multiple critical sections of the guide shroud based on the control deviation vector, and to acquire flow velocity and temperature parameters at each critical section to form a set of control base point data with spatial coordinates.

[0018] The 3D spatial mapping and prism construction module is used to map the control base point data to a 3D feature space consisting of height, flow rate and temperature, and to construct multiple dynamic control prisms, with the vertex of each dynamic control prism corresponding to the coordinates of a control base point.

[0019] The distortion index calculation module is used to calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space, so as to obtain the control distortion index that characterizes the degree of fit between the current control state and the target state.

[0020] The dynamic compensation and command generation module is used to compare the control distortion index with the preset threshold. If it exceeds the threshold, the dynamic compensation amount for the height of the guide cover is derived based on the control distortion index to correct the control deviation vector and generate the optimized control command.

[0021] The adjustment execution and accuracy calibration module is used to drive the guide cover to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

[0022] The above-described solution of the present invention has at least the following beneficial effects:

[0023] Because it employs a multi-source data acquisition and attention mechanism fusion technique, it overcomes the problems of single data and large abnormal fluctuation interference during the guide shield adjustment process, thereby improving the accuracy and stability of the initial state vector. Because it adopts a three-dimensional feature space mapping and dynamic control prism construction technique, it overcomes the problem that single-point height adjustment alone cannot characterize the flow field and temperature gradient distortion, thereby achieving coupled quantitative evaluation of height, flow velocity, and temperature. Because it adopts a control distortion index calculation and height dynamic compensation technique, it overcomes the problems of mismatch between adjustment action and actual process deviation and low adjustment accuracy, thereby improving the height control accuracy and response speed of the guide shield. Because it adopts a feedback calibration and prism construction rule self-updating technique, it overcomes the problems of large cumulative error and lack of closed-loop calibration in long-term operation, thereby ensuring long-term stable operation and batch consistency of the system. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a method for data acquisition and accuracy calibration of a guide cover adjustment according to an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of a guide cover adjustment data acquisition and accuracy calibration system provided in an embodiment of the present invention. Detailed Implementation

[0026] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0027] like Figure 1 As shown, an embodiment of the present invention proposes a method for data acquisition and accuracy calibration of guide cover adjustment, the method comprising the following steps:

[0028] Step 1: Collect real-time height position data of the guide hood during the saccharification and boiling process, and combine it with the synchronously acquired process parameters to generate an initial state vector of control with process identification.

[0029] Step 2: Compare the initial state vector of regulation with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector.

[0030] Step 3: Determine the positions of multiple key sections of the guide shroud based on the control deviation vector, and collect flow velocity and temperature parameters at each key section to form a set of control base point data with spatial coordinates;

[0031] Step 4: Map the control base point data to a three-dimensional feature space consisting of height, flow rate, and temperature to construct multiple dynamic control prisms. The vertex of each dynamic control prism corresponds to the coordinates of a control base point.

[0032] Step 5: Calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space to obtain the control distortion index, which characterizes the degree of fit between the current control state and the target state.

[0033] Step 6: Compare the control distortion index with the preset threshold. If it exceeds the threshold, derive the dynamic compensation amount for the height of the guide cover based on the control distortion index to correct the control deviation vector and generate the optimized control command.

[0034] Step 7: Drive the guide cover to perform height adjustment according to the optimized control command. After the adjustment is completed, collect feedback data again to generate a feedback state vector. Compare the feedback state vector with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

[0035] In this embodiment of the invention, it should be emphasized that the dynamically controlled prism is specifically a dynamically controlled triangular prism. By collecting the real-time height of the guide hood and synchronous process parameters, a state vector with process identification is generated, realizing the accurate fusion and characterization of multi-source data. By constructing a control deviation vector and determining the key section position, the spatial distribution data of flow field and temperature can be comprehensively obtained. Mapping the base point data to a three-dimensional feature space and constructing a dynamic prism can intuitively reflect the spatial coupling relationship of the process state. Calculating the control distortion index to quantify the degree of fit can achieve accurate evaluation and dynamic compensation of the control effect. Generating optimization instructions based on the distortion index can effectively improve the accuracy and response speed of the guide hood height adjustment. Collecting feedback data and updating the model in real time can effectively eliminate accumulated errors, ensure the long-term stable operation of the system, and ultimately realize the closed-loop control of the guide hood adjustment in the saccharification and boiling process, reasonably improving the uniformity and batch stability of the boiling process.

[0036] In a preferred embodiment of the present invention, step 1 above may include:

[0037] Step 1.1: Acquire the height position signal of the guide hood in real time according to the preset sampling frequency as the original height time series data. Specifically, this includes: setting a fixed sampling frequency in advance according to the actual needs of the saccharification and boiling process (the sampling frequency can be adjusted according to the actual needs of the saccharification and boiling process). Combining the response speed of the guide hood height adjustment and the change law of process parameters, in this embodiment, the sampling frequency is set to 10Hz, that is, the height position signal of the guide hood is collected once every 0.1 seconds; a high-precision displacement sensor is installed on the drive mechanism of the guide hood (this drive mechanism is installed on the top of the boiling pot, which can realize the smooth up and down movement of the guide hood, and has the characteristics of high positioning accuracy, low operating noise, and adaptability to high temperature conditions) to capture the height position change of the guide hood in real time throughout the saccharification and boiling process. The sensor converts the physical displacement signal into an electrical signal and transmits it to the data acquisition terminal. The data acquisition terminal performs noise reduction processing on the received electrical signal, and then arranges the processed signal in time sequence according to the acquisition time to form continuous and complete original height time series data. At the same time, the corresponding acquisition timestamp is marked for each set of height data to ensure the traceability of the data.

[0038] Step 1.2: Synchronously collect wort temperature, steam flow rate, and pressure parameters inside the boiling vessel during the saccharification and boiling process to create a synchronous process time-series dataset. Specifically, this includes: simultaneously initiating the process parameter acquisition process while starting the guide hood height position signal acquisition, ensuring complete synchronization between the height data and process parameter acquisition times; during acquisition, first, collect wort temperature data from multiple points inside the boiling vessel, selecting different depths within the vessel and taking the average value as the current wort temperature parameter; simultaneously acquire steam flow rate data from the steam input pipe and pressure data inside the boiling vessel, ensuring that all three process parameters are real-time data from the same moment; after acquisition, integrate and bind the three process parameters (wort temperature, steam flow rate, and pressure inside the vessel) at the same timestamp, ensuring that each set of process parameters accurately corresponds to the corresponding guide hood height data, forming a synchronous process time-series dataset that matches the original height time-series data one-to-one, providing a reliable time synchronization basis for the spatiotemporal feature alignment of multi-source data.

[0039] Step 1.3: Input the original height time series data and the synchronous process time series dataset into the pre-built multi-source information fusion model. The pre-built multi-source information fusion model performs spatiotemporal feature alignment of the original height time series data and the synchronous process time series dataset based on the attention mechanism to identify and remove abnormal fluctuation points. At the same time, it uses interpolation compensation to repair missing data and generate enhanced aligned multi-source monitoring data. Specifically, the pre-built multi-source information fusion model is based on the Transformer encoder and has been specially improved. The core improvement is that, on the basis of the Transformer encoder, a multi-head attention mechanism module, a spatiotemporal feature fusion module, and a data augmentation module adapted to the characteristics of saccharification and boiling process data are embedded, which are specifically used to handle the time series fusion task of guide hood height and multiple process parameters. In the model building phase, data preparation work was carried out first. Full time-series data of different production batches, different process formulas (such as pale beer and dark beer), and different equipment operating conditions were collected from the beer saccharification production line for 3 to 6 consecutive months. Specifically, time-series data of guide hood height position, wort temperature, steam flow, and pressure inside the pot were collected. In particular, abnormal fluctuation data caused by sensor instantaneous failure and missing data caused by data transmission interruption (missing time is 1 to 5 sampling periods) were collected to build an original training dataset covering normal and abnormal operating conditions.

[0040] The original training dataset is preprocessed. First, all time-series data are standardized, mapping the values ​​of each parameter to the [0,1] interval to eliminate the influence of dimensional differences and numerical spans between different parameters. Then, the time-series data is segmented by sliding window, with a sliding window length of 10 sampling periods and a step size of 5 sampling periods, converting continuous time-series data into fixed-length sample sequences. Each sample sequence is labeled with precise labels, including the standard results of spatiotemporal feature alignment, the location and type of abnormal fluctuation points (such as abrupt changes caused by sensor interference, or fluctuations caused by instantaneous process fluctuations), and the true reference values ​​of missing data (supplemented by manual verification of equipment logs from the same period). This forms a standardized training dataset and a validation dataset, with a ratio of 8:2 between the training set and the validation set.

[0041] The training process employs a phased iterative training approach. The first phase trains the multi-head attention mechanism module. By adjusting the number of attention heads (set to 8) and the initialization method of attention weights, the model can automatically capture the temporal correlations and spatial coupling relationships between different dimensions of temporal data (height, temperature, flow rate, pressure), such as identifying the synchronous change pattern of the guide hood height when the wort temperature rises sharply. The second phase trains the spatiotemporal feature fusion module, introducing a temporal convolutional layer (with a kernel size of 3) to extract local temporal features of each parameter. Combined with the global features extracted by the Transformer encoder, a weighted fusion method is used to obtain the fused features of the multi-source data. The weighting coefficients are pre-set according to the degree of influence of each parameter on the control of the guide hood (guide hood...). The weights for the boiler height (0.4), wort temperature (0.3), steam flow rate (0.15), and boiler pressure (0.15) are calculated. The third stage, the training data augmentation module, introduces cross-entropy loss (for outlier identification) and mean squared error loss (for missing data repair) for outlier identification and missing data repair tasks. The Adam optimizer is used with a learning rate of 0.001, 500 iterations, and a batch size of 32. Model performance is validated every 10 iterations. Iterative training stops when the model achieves a spatiotemporal feature alignment accuracy of over 98%, an outlier identification F1 score of over 97%, and a missing data repair mean squared error of less than 0.01 on the validation set, thus completing the construction and training of the multi-source information fusion model.

[0042] The improved Transformer encoder architecture is used to build a multi-source information fusion model. Its core advantage lies in the fact that, compared with traditional LSTM and CNN temporal fusion models, the Transformer's multi-head attention mechanism can better capture the long-range dependencies of multi-source temporal data, adapting to the characteristics of slow parameter changes and multi-parameter synergistic effects in the saccharification and boiling process. At the same time, the improved model adds a data augmentation module, which can specifically handle abnormal and missing data commonly found in industrial production. There is no need to build a separate anomaly handling model, simplifying the data processing process and improving the model's practicality and robustness. In practical saccharification and boiling production applications, after the model completes training and is put into real-time operation, the raw height time-series data collected in real time and the synchronous process time-series dataset are input into the model. The model begins to standardize the input data to ensure that the input data format is consistent with the training data. The multi-head attention mechanism module accurately aligns the time-series data of different dimensions in terms of time and space, automatically matches the collected data of different parameters at the same time, and removes invalid data such as height data mutations caused by poor sensor contact, steam interference, and abnormal temperature data fluctuations. At the same time, for missing data caused by network interruption or equipment failure during data transmission, the model uses interpolation compensation based on spatiotemporal correlation features to intelligently repair the missing data by combining the parameter change trend of adjacent time moments and the historical data patterns of the same batch, thus avoiding the impact of missing data on the subsequent state vector generation. Finally, the output is multi-source monitoring data that has been data augmented, improved in quality, and highly aligned in terms of spatiotemporal features, providing accurate and reliable data support for the generation of the initial state vector.

[0043] Step 1.4 involves performing similarity matching between the aligned multi-source monitoring data and the features of each recipe in the preset process recipe library to identify the process recipe type of the current batch. Corresponding process recipe identifiers and collection time tags are then added to the aligned multi-source monitoring data to generate labeled data with complete process identifiers. Specifically, this includes: building a preset process recipe library. This library stores standard parameter features of various common process recipes in beer production (including different categories such as pale ale, dark ale, and stout). Each recipe includes the standard time series range and variation patterns of guide hood height, wort temperature, steam flow rate, and pressure inside the brewing vessel. A unique process recipe number is assigned to each recipe for quick identification and adjustment. The enhanced multi-source monitoring data is then matched one by one with the features of each standard formula in the preset process formula library. Simultaneously, the similarity value between the monitoring data and each standard formula feature is calculated (similarity value ranges from 0 to 1, with values ​​closer to 1 indicating a higher degree of matching). A similarity threshold of 0.9 is set. When the similarity value between a standard formula and the monitoring data reaches or exceeds this threshold, the current production batch is determined to be suitable for that formula, and the formula number is recorded. Finally, the corresponding process formula number, precise collection time stamp, and current production batch number are added to the multi-source monitoring data to form complete, clearly identified, and traceable labeled data, ensuring that subsequent control processes can accurately match the requirements of the current process formula.

[0044] Step 1.5: Following a preset vector dimension order, normalize and combine the parameter values ​​contained in the labeled data to generate an initial control state vector. Specifically, this includes: pre-setting the dimension order of the initial control state vector; combining the core influencing factors of the guide shield control, determining the vector dimension order as: guide shield height, wort temperature, steam flow rate, and pressure inside the vessel. This order is consistent with the dimension order of the subsequent control deviation vector to avoid dimension misalignment during vector comparison; normalizing the four parameters in the labeled data—guide shield height, wort temperature, steam flow rate, and pressure inside the vessel—using the min-max normalization method to uniformly map the values ​​of each parameter to the [0,1] interval. The specific process is as follows: first, calculate the maximum and minimum values ​​of each parameter in the historical data; then, calculate the normalized parameter value using the formula: (current parameter value - minimum parameter value) / (maximum parameter value - minimum parameter value), thereby eliminating dimensional differences between different parameters (e.g., guide shield height in millimeters, wort temperature in degrees Celsius, and steam flow rate in meters per second). 3 The influence of the normalized parameters ( / h) and the numerical span is considered. The four parameters are combined sequentially according to the preset dimensional order to form a fixed-dimensional and numerically standardized initial control state vector. This vector can comprehensively and accurately characterize the initial control state of the current saccharification and boiling process, providing standard data for subsequent step deviation comparison.

[0045] In a preferred embodiment of the present invention, step 2 above may include:

[0046] Step 2.1: Obtain the initial control state vector and simultaneously extract the target state vector corresponding to the current operating batch from the preset process formula library. Both the initial control state vector and the target state vector contain parameter values ​​in the dimensions of guide hood height, wort temperature, steam flow rate, and pressure inside the pot. Specifically, this includes: retrieving the finally generated initial control state vector, which, after normalization, contains standardized parameter values ​​for four dimensions: guide hood height, wort temperature, steam flow rate, and pressure inside the pot, fully carrying the real-time operating status information of the current saccharification and boiling process; at the same time, based on the labeled process formula number, accurately retrieve and extract the target state vector that perfectly matches the current operating batch from the preset process formula library. The dimension order and parameter type of the target state vector are completely consistent with the initial control state vector, and the parameter values ​​of each dimension are the ideal standard values ​​specified by the corresponding process formula. This serves as the core benchmark for deviation comparison in subsequent steps, ensuring the standardization and accuracy of the comparison process.

[0047] Step 2.2 involves comparing the initial control state vector and the target state vector element by element according to a preset vector dimension order. The differences between the initial control state vector and the target state vector are calculated for the dimensions of guide hood height, wort temperature, steam flow rate, and pot pressure, respectively. These differences serve as the initial deviation values ​​for each dimension. Specifically, this includes strictly adhering to the preset vector dimension order: guide hood height, wort temperature, steam flow rate, and pot pressure. The retrieved initial control state vector and the extracted target state vector are precisely compared dimension by dimension and element by element. First, it is confirmed that the number of dimensions and parameter types of the two vectors are completely consistent, ensuring that the parameters of each dimension can achieve a one-to-one precise correspondence. This eliminates comparison errors caused by disordered dimension order or parameter mismatch, ensuring the rigor of the comparison process.

[0048] During the comparison process, detailed difference calculations were performed for each dimension. Specifically: First, the difference calculation for the guide shield height dimension was performed. Using the standardized real-time parameter value of the guide shield height in the initial state vector as a benchmark, the standardized standard parameter value of the guide shield height in the target state vector was subtracted. The result is the initial deviation value for the guide shield height dimension. A positive result indicates that the current guide shield height is higher than the target height; a negative result indicates that the current height is lower than the target height. Next, the difference calculation for the wort temperature dimension was performed, using the same logic as for the height dimension. The standardized real-time parameter value of the wort temperature in the initial state vector was subtracted from the standardized standard parameter value of the wort temperature in the target state vector to obtain the initial deviation value for the wort temperature dimension, directly reflecting the deviation between the current wort temperature and the target temperature. Finally, the difference calculation for the steam flow rate dimension was performed. The standardized real-time parameter value of the steam flow rate in the initial state vector was subtracted from the standardized standard parameter value of the steam flow rate in the target state vector to generate the initial deviation value for the steam flow rate dimension, confirming the difference between the current steam flow rate and the standard flow rate.

[0049] Finally, the difference in the pressure dimension inside the pot is calculated by subtracting the standardized standard parameter value of the pressure inside the pot in the target state vector from the standardized real-time parameter value of the pressure inside the pot in the initial state vector. This yields the initial deviation value of the pressure dimension inside the pot, quantifying the degree of deviation between the current pressure inside the pot and the target pressure. Through the above detailed calculations in each dimension, the initial deviation values ​​of the guide shroud height, wort temperature, steam flow rate, and pressure inside the pot are obtained respectively. Each deviation value clearly and accurately reflects the degree of deviation between the current process state and the ideal standard in the corresponding dimension, providing accurate and reliable deviation data for the positioning of key sections and the implementation of dynamic control actions.

[0050] Step 2.3: Combine the initial deviation values ​​corresponding to each dimension according to the preset vector dimension order to generate a complete control deviation vector containing deviation values ​​of four dimensions. Specifically, this includes: uniformly verifying the calculated initial deviation values ​​of the four dimensions to confirm that each deviation value is a valid calculation result for the corresponding dimension, without missing or errors, ensuring that the deviation data used for combination is accurate and reliable; after verification, strictly follow the preset vector dimension order to orderly combine the initial deviation values ​​of the four dimensions. The combination process is as follows: take the initial deviation value of the guide cover height dimension as the first dimension element of the control deviation vector. This element is directly associated with subsequent guide covers. The positioning calculation of the key section of the cover is the core foundation of the assembly process. The initial deviation value of the wort temperature dimension is used as the second dimension element of the control deviation vector, which is connected to the first dimension element in sequence to ensure that the dimension order corresponds completely with the control initial state vector and the target state vector without any disorder. The initial deviation value of the steam flow rate dimension is used as the third dimension element, which follows the wort temperature deviation value and maintains the orderly connection with the first two dimension elements to ensure the standardization of the vector structure. The initial deviation value of the pressure inside the pot is used as the fourth dimension element, which is the last dimension of the control deviation vector to complete the orderly arrangement of the four dimension deviation values.

[0051] During the combination process, the independence and integrity of each deviation value are strictly maintained. The original calculation result of any deviation value is not changed. The values ​​are arranged and integrated only according to a fixed dimensional order. After the combination is completed, a complete control deviation vector containing four-dimensional deviation values ​​is generated. This control deviation vector systematically integrates the deviation information of each key parameter of the current process. It can effectively and accurately characterize the overall deviation between the current saccharification and boiling process and the target process state, providing core input basis for determining the location of key sections.

[0052] In a preferred embodiment of the present invention, step 3 above may include:

[0053] Step 3.1: Analyze the control deviation vector to extract the deviation value of the guide shield height dimension and the rate of change of the guide shield height dimension deviation value relative to the previous sampling time, as the input basis for determining the cross-sectional position. Specifically, this includes: analyzing the generated complete control deviation vector, accurately identifying and extracting the guide shield height dimension deviation value of the first dimension in the vector, denoted as... Simultaneously, the control deviation vector generated at the previous sampling time is retrieved, and the corresponding guide shield height dimension deviation value in the vector is extracted and denoted as... The time interval between the previous sampling time and the current sampling time is denoted as That is, the sampling period, which corresponds to the sampling frequency preset in step 1.1. When the sampling frequency is 10Hz... ;

[0054] The rate of change of the height deviation relative to the previous sampling time is calculated using the height deviation values ​​at two consecutive sampling times, and denoted as . The calculation process can be expressed as a formula: ; Extract the height deviation value and the calculated rate of change Together, they serve as the core input for determining the location of key sections of the guide shield, ensuring the accuracy and relevance of section positioning.

[0055] Step 3.2: Based on the deviation values ​​and rate of change of the height dimension, and combined with the preset structural parameters of the guide shroud, calculate the specific spatial height positions of the upper edge section, the lower edge section, and the middle guide ring section of the guide shroud, respectively, as the position coordinates of multiple key sections. Specifically, this includes: obtaining the obtained height dimension deviation values. and its rate of change Simultaneously, it retrieves pre-stored pre-defined structural parameters of the guide shield cylinder, including the total height of the guide shield. Cylinder wall thickness Key parameters such as the installation position ratio k of the central guide ring (i.e., the ratio of the height of the guide ring from the lower edge of the guide shield to the total height of the guide shield, with a preset range of 0.4 to 0.6) are all inherent structural parameters of the guide shield and are denoted as fixed constants.

[0056] Based on these parameters, the specific spatial height positions of the upper edge section, lower edge section, and middle guide ring section of the guide shield are calculated. The calculation process can be expressed by the following formula:

[0057] ;

[0058] in This is the reference height (preset constant) of the lower edge of the guide hood relative to the bottom of the boiling pot when it is at the target height. The coordinates of the actual spatial height of the lower edge section of the guide shroud relative to the bottom of the boiling pot are shown. The coordinates of the actual spatial height of the upper section of the guide shroud relative to the bottom of the boiling pot are: The coordinates represent the actual spatial height of the guide ring section in the middle of the guide shroud relative to the bottom of the boiling pot. The total height of the guide shield; finally, the specific spatial height coordinates of the three key sections were determined. , , This serves as the precise positioning basis for subsequent parameter acquisition steps.

[0059] Step 3.3: Based on the location coordinates of multiple key sections, the average wort flow rate and local temperature parameters at each key section are collected in real time. Specifically, this includes determining the specific spatial height coordinates of the three calculated key sections. , , Afterwards, following the preset acquisition sequence—first the lower section, then the middle guide ring section, and finally the upper section—parameter acquisition is carried out for each key section in sequence. At each key section, 3 to 5 acquisition points are evenly selected along the circumference of the section, and the wort flow rate data at each point is collected in real time. The acquisition is performed at least 3 times, and the average wort flow rate parameter of that section is obtained by averaging the multiple acquisitions. This average flow rate parameter is denoted as v. The calculation process can be expressed as the formula: ;in The data includes wort flow rate data at different points and with different collection times for this cross section, where n is the total number of collections (n≥3). At the same time, the wort temperature data at the center point of this cross section is collected as a local temperature parameter, denoted as T, to ensure that the collected flow rate v and temperature T parameters can truly reflect the actual process state of this cross section and avoid the randomness of single collection data affecting the accuracy of the data.

[0060] Step 3.4: Associate and bind the position coordinates of each key section with the flow velocity and temperature parameters collected at that section to form control base point data composed of spatial coordinates and the flow velocity and temperature values ​​corresponding to the vertical axial height coordinates of the key annular section of the guide shroud. Specifically, this includes: after completing the acquisition of the position coordinates of each key section and the corresponding flow velocity and temperature parameters, as well as the spatial height coordinates of each key section (… or or The data is associated with the average flow velocity parameter v and the local temperature parameter T collected at the cross-section, ensuring that each spatial coordinate corresponds to a unique flow velocity value v and temperature value T, and preventing misalignment of parameters and coordinates. Through this association and binding method, a complete data unit (H, v, T) is formed, in which each set of data contains a spatial height coordinate H, a flow velocity value v, and a temperature value T. Multiple sets of data units together constitute a set of control base point data with spatial coordinates, providing complete and accurate data support for three-dimensional feature space mapping and dynamic control prism construction.

[0061] In a preferred embodiment of the present invention, step 4 above may include:

[0062] Step 4.1: Obtain the control base point data. Map the spatial height coordinates, flow velocity values, and temperature values ​​contained in each control base point data to the corresponding height axis, flow velocity axis, and temperature axis in the three-dimensional feature space, respectively, to form the spatial coordinate point of each control base point in the three-dimensional feature space. Specifically, this includes: first, obtaining the generated set of all control base point data. Each set of data consists of three associated parameters: spatial height coordinates, average wort flow velocity value, and local temperature value, which correspond to three different key sections: the lower edge, the middle, and the upper edge, respectively; then, establishing a standardized three-dimensional feature space. The three-dimensional feature space is a complete three-dimensional coordinate system with the orthogonal height axis, flow velocity axis, and temperature axis as the core.

[0063] In the three-dimensional feature space, the value range of the height axis corresponds to the actual spatial height range collected by the control base point, the value range of the flow velocity axis corresponds to the average flow velocity range of the wort collected at each cross-section, and the value range of the temperature axis corresponds to the local temperature range of each cross-section. Each set of data in the control base point dataset is independently mapped, that is, the spatial height coordinates are mapped to specific numerical positions on the height axis, the average wort flow velocity value is mapped to specific numerical positions on the flow velocity axis, and the local temperature value is mapped to specific numerical positions on the temperature axis. By positioning the parameter values ​​of the three dimensions one-to-one in the three-dimensional feature space, a unique and definite three-dimensional spatial coordinate point is generated for each control base point. This spatial coordinate point completely and accurately represents the spatial location of the control base point and the corresponding process parameter state, providing a solid data foundation for spatial relationship analysis.

[0064] Step 4.2 involves performing spatial distribution feature analysis on the spatial coordinates of multiple control base points in the three-dimensional feature space. This analysis identifies the spatial coordinates of the three control base points located at adjacent key sections with the shortest straight-line distance between them. These coordinates are then connected pairwise to form the bases of multiple triangles. Specifically, this includes: after obtaining the spatial coordinates of all control base points in the three-dimensional feature space, conducting a comprehensive and detailed spatial distribution feature analysis and clustering screening of these coordinates. All coordinate points are preprocessed to remove discrete outliers caused by acquisition errors, ensuring the accuracy of the data for analysis. The punctuation marks accurately reflect the distribution of process parameters at each key section. During the analysis, based on the spatial height coordinates of each control base point and the determined height ranges of the three key sections (lower section, middle guide ring section, and upper section), all spatial coordinate points are precisely divided into three independent subsets. Each subset contains only the coordinates of the control base point on the corresponding key section, corresponding to the coordinate point set of the lower section, the coordinate point set of the middle guide ring section, and the coordinate point set of the upper section, respectively. This ensures that there is no overlap or omission between the subsets, laying the foundation for the spatial relationship analysis of adjacent sections.

[0065] The process focuses on the spatial distribution relationship of coordinate points between adjacent sections, confirming that the analysis object is the combination of coordinate points of two sets of adjacent sections, namely the spatial positional relationship between the coordinate point set of the lower edge section and the coordinate point set of the middle guide ring section, and between the coordinate point set of the middle guide ring section and the coordinate point set of the upper edge section. It does not involve the analysis of coordinate points of non-adjacent sections, ensuring the focus of the analysis. In each pair of coordinate point combinations of adjacent sections, the cross-section coordinate point pairs are first identified, that is, each coordinate point pair consists of two coordinate points respectively taken from two adjacent key sections. Then, from all cross-section coordinate point pairs, a candidate set of three points that meets the quantity requirements is combined. Each candidate set of three points contains three coordinate points spanning adjacent sections, and each of the three coordinate points must correspond to a different control base point to avoid the same control base point participating in the combination repeatedly. For each candidate set of three points, the spatial straight-line distance between any two coordinate points in this set is calculated sequentially. After each pair of two points is calculated, the distance value is recorded in time. Then, the three pairwise distance values ​​are squared, and the three squared values ​​are added together to obtain the sum. The square root of the sum is then taken to obtain the overall spatial distance metric of the three-point combination. This metric can accurately reflect the overall spatial compactness between the three coordinate points.

[0066] All possible candidate sets of three points are iterated one by one to ensure that no combination that meets the conditions is missed. At the same time, all calculated overall spatial distance metrics are compared and sorted one by one, arranged in ascending order of metric value, and the three-point combination with the smallest metric value is selected. That is, the three control base point coordinates with the shortest total straight-line distance between adjacent key sections are accurately identified. For each set of three coordinate points that meet the shortest distance condition, a closed planar geometric structure is constructed by connecting them in pairs. During the connection process, it is ensured that each line segment accurately connects two coordinate points, and the three line segments are connected to form a complete triangle. This triangle is the base of the dynamic control triangular prism. Each triangle base can accurately correspond to the local area of ​​process parameter distribution between two adjacent key sections, representing the local optimal topological relationship of process parameter distribution between the two key sections. This provides a reliable and stable basic geometric unit for the accurate construction of the dynamic control triangular prism.

[0067] Step 4.3: Determine the key section where the base of each triangle is located, identify the preset key sections adjacent to this key section on the height coordinate axis, extend the three vertices of the base of the triangle along the height coordinate axis and connect them to the adjacent preset key sections, and form a dynamic control triangular prism formed by the six vertices by matching the spatial coordinates of the control base points corresponding to the projection positions of each vertex on the flow velocity axis and temperature axis. Specifically, for each constructed base of the triangle, accurately calibrate the section to which it belongs based on the spatial height coordinate attributes of each vertex. By matching the height values ​​of the three vertices with the height ranges of the three key sections of the lower edge, middle guide ring and upper edge one by one, confirm the reference belonging of the base of the triangle, and at the same time determine the combination of its corresponding adjacent sections, that is, determine whether it is a combination of the lower edge and middle guide ring sections or a combination of the middle guide ring and upper edge sections.

[0068] The preset key sections here refer to the core sections pre-set on the guide shroud structure for accurately collecting process parameters and constructing a three-dimensional feature space. Their positions are strictly matched with the structural design of the guide shroud and the requirements of the saccharification and boiling process. Specifically, they include three fixed sections: the lower edge section, the middle guide ring section, and the upper edge section. All three sections are perpendicular to the central axis of the guide shroud and are evenly and orderly distributed along the height direction of the guide shroud. The lower edge section is close to the wort surface in the boiling pot, the upper edge section is located at the top of the guide shroud cylinder, and the middle guide ring section is located in the middle area between the lower and upper edge sections and completely coincides with the middle guide ring of the guide shroud. Each preset key section corresponds to a fixed height range, and its height coordinates are stored in advance in the preset process parameter library as a reference for section positioning, parameter collection, and spatial extension. Next, using the height axis of the three-dimensional feature space as a reference, the preset key sections directly adjacent to the reference section in the height dimension are precisely located. The height difference between the two sections is checked to see if it conforms to the inherent structural parameters of the guide shroud. Non-directly adjacent cases caused by section division errors are eliminated to ensure that the two form a continuous and unique connection relationship in space: that is, the preset key section adjacent to the lower edge section is the middle guide ring section, the preset key sections adjacent to the middle guide ring section are the lower edge section and the upper edge section, the preset key section adjacent to the upper edge section is the middle guide ring section, and there are no other additional sections between the adjacent preset key sections, ensuring the continuity and correctness of the spatial relationship.

[0069] Using the three vertices of the triangle's base as starting points, a vertical extension operation is performed along the height axis. The extension displacement is strictly equal to the actual height difference between the reference section and the adjacent preset key section, ensuring that the coordinates of the extended vertex on the height axis completely coincide with the height coordinates of the adjacent key section. During the extension process, a precise matching search is simultaneously performed on the coordinate points subset of adjacent key sections. The search is based on the projected coordinates of the starting vertex on the velocity and temperature axes. By calculating the deviation values ​​between the starting vertex and each coordinate point of the adjacent section in the two dimensions of velocity and temperature, the coordinate point with the smallest deviation value that meets the preset accuracy threshold is selected and determined as the target vertex for the extension, ensuring that the two dimensions are accurately matched. The correspondence between the original and target vertices in the process parameters is as follows: After matching all the starting vertices with the target vertices, the three original vertices of the triangle base are connected in pairs to form the lower base, and the three target vertices are connected in pairs to form the upper base. Finally, each original vertex is perpendicularly connected to its corresponding target vertex to form six side edges, which ultimately enclose a closed three-dimensional structure consisting of six vertices, two triangle bases, and three rectangular sides, i.e., a dynamically controlled triangular prism. The spatial range of this dynamically controlled triangular prism strictly corresponds to the process area between two adjacent preset key sections, and its geometric shape can accurately map the coupling distribution and spatial correlation characteristics of the three-dimensional parameters of wort height, flow rate, and temperature in this area.

[0070] Step 4.4: Calculate the direction vector of each edge of the dynamically controlled triangular prism. The direction vector of each edge represents the trend of change between the two connected control base points in the velocity or temperature dimension. Specifically, after constructing all dynamically controlled triangular prisms, perform fine-grained identification and classification of the edges of each prism. Based on the geometric structural characteristics of the triangular prism, all its edges are divided into two categories: the first category is the base edges, including the three edges of the bottom face (the base of the original triangle) and the three edges of the top face (the triangle formed by the target vertices); the second category is the lateral edges, i.e., the edges connecting the original vertices. The three edges of the target vertex are labeled with the starting and ending vertex identifiers of each edge to ensure that there are no omissions or repetitions among the six edges, and that the vertex relationships of each edge are clear and traceable. For each edge, the direction vector is accurately calculated. The starting vertex of the edge is used as the reference point to extract its three-dimensional coordinates of height, flow velocity, and temperature in the three-dimensional feature space. Then, the corresponding three-dimensional coordinates of the ending vertex are extracted. By subtracting the starting vertex coordinates from the ending vertex coordinates, the coordinate differences in the three directions of height axis, flow velocity axis, and temperature axis are obtained. These three differences together constitute the direction vector of the edge.

[0071] After calculation, the three components of the direction vector are analyzed in detail, focusing on the physical meaning of the velocity axis component and the temperature axis component. The height axis component is only used to characterize the spatial extension direction of the edges, while the sign of the velocity axis component directly reflects the trend of the wort velocity change from the initial vertex to the final vertex. A positive value represents an increase in velocity, and a negative value represents a decrease in velocity. The absolute value of the component quantifies the magnitude of the velocity change. Similarly, the sign and absolute value of the temperature axis component characterize the trend and magnitude of the wort temperature increase and decrease. For the bottom edges, the direction vector mainly reflects the gradient change of process parameters between different control points on the same key section. For the side edges, the direction vector mainly reflects the evolution law of process parameters between corresponding control points of two adjacent key sections. The calculation results of the direction vectors of all edges will serve as the core basis for subsequent calculations of characteristic parameters such as the volume and axial projection of the triangular prism under dynamic control. At the same time, it provides accurate vector basis for quantitative analysis of the spatial distortion of the wort flow field and temperature field during saccharification and boiling.

[0072] In a preferred embodiment of the present invention, step 5 above may include:

[0073] Step 5.1: Obtain the generated dynamic control triangular prisms and extract the spatial coordinates of the six vertices from each dynamic control triangular prism. Specifically, this includes: obtaining a set of all constructed dynamic control triangular prisms; for each dynamic control triangular prism in the set, based on its internal preset geometric topological connection relationship: the connection relationship is a pre-defined rule used to regulate the vertex connection logic and structural form of the dynamic control triangular prism, precisely defining the connection method, belonging relationship, and spatial position association of the six vertices; specifically, the six vertices are divided into two groups, with three vertices in each group forming a triangular base; the two bases correspond to two adjacent preset key sections, and each vertex of the lower base is connected to the corresponding vertex of the upper base through a side edge perpendicular to the base, forming a closed triangular prism solid structure; based on this geometric topological connection relationship, accurately extract and identify the unique spatial coordinates of the six vertices contained in the solid structure; to ensure the accuracy of subsequent calculations, the six vertices are numbered in order and recorded as vertices. The first three vertices The three vertices form the base of the triangle located at the lower key section. Forming the base of a triangle located at the adjacent critical section above, and the vertex and , and , and Each vertex is vertically connected by three side edges; the spatial coordinates (H, v, T) of each extracted vertex are independently stored and associated, forming a vertex coordinate dataset specific to this triangular prism, providing the most basic raw data input for the calculation of volume, projection and feature parameters.

[0074] Step 5.2: Based on the spatial coordinates of the six vertices of each dynamically adjustable triangular prism, calculate the volume value of the dynamically adjustable triangular prism as the volume characteristic parameter of the prism. Specifically, this includes: calculating the volume value of the dynamically adjustable triangular prism based on the extracted spatial coordinates of the six vertices, which can be obtained by decomposing it into the product of the area of ​​the base triangle and the height difference between the two cross sections; selecting the three vertices of the base of the lower triangle. Calculate the area of ​​the base triangle using the cross product method. The mathematical expression is ;in From the vertex point to spatial vectors, From the vertex point to spatial vectors, This indicates the magnitude of the vector being calculated; the absolute distance between the lower and upper base surfaces along the height axis H, i.e., the height h, is calculated using the following expression: ;in For any vertex of the upper bottom surface (e.g.) The height coordinates of the triangular prism are used to dynamically adjust its volume. Calculated using the following formula ; in the formula This refers to the volumetric characteristic parameters of the triangular prism, which characterize the three-dimensional distribution scale of process parameters between two adjacent key sections.

[0075] Step 5.3: Obtain the calculated direction vectors of each edge. Based on the direction vectors, calculate the projected length of each edge of the dynamically controlled triangular prism in the direction of the flow velocity axis, which serves as the characteristic parameter of the flow velocity axial projection of this prism. Specifically, this includes: retrieving the calculated direction vectors of the six edges belonging to the dynamically controlled triangular prism. Each direction vector is represented in the form of a three-dimensional coordinate difference, denoted as... In this step, the focus is on extracting and calculating the projected length of the direction vector of each edge in the direction of the flow velocity axis, which serves as the characteristic parameter of the axial projection of the flow velocity. For any edge direction vector... Its projected length along the velocity axis It can be obtained by calculating the dot product of the direction vector and the unit vector of the velocity axis, and its mathematical expression is: ;in It is the unit vector along the velocity axis. This represents the coordinate difference of the direction vector of the edge in the velocity dimension, i.e., the difference in the velocity axis values ​​of the two vertices connected by the edge; this is calculated by traversing all six edges of the triangular prism. The values ​​are integrated into a set of characteristic parameters for the axial projection of the flow velocity. This set of parameters intuitively reflects the extent and trend of the extension of the triangular prism structure in the dimension of wort flow velocity variation.

[0076] Step 5.4: Calculate the projected length of each edge of the dynamically controlled triangular prism along the temperature axis based on the direction vector, using this as the temperature axial projection characteristic parameter of the prism. Specifically, this includes: continuing the logic of step 5.3, further calculating the projected length of the direction vector of each edge along the temperature axis, using this as the temperature axial projection characteristic parameter; for any edge direction vector... Its projected length along the temperature axis It is obtained by calculating the dot product of the direction vector and the unit vector of the temperature axis, and the mathematical expression is: ;in This is the unit vector along the temperature axis. This represents the coordinate difference of the direction vector of the edge along the temperature dimension, i.e., the difference in the temperature axis values ​​of the two vertices connected by the edge; complete the calculation for all six edges of the triangular prism. After the value is calculated, a set of temperature axial projection characteristic parameters is generated. This set of parameters is used to quantitatively characterize the spatial variation of wort temperature from one control point to another, providing core vector data for analyzing the temperature field distribution.

[0077] Step 5.5: Identify the shared face between two adjacent dynamically regulated triangular prisms. The shared face is formed by the two vertices of each of the two adjacent prisms, for a total of four vertices. Calculate the area value of the shared face based on the spatial coordinates of the four vertices forming the shared face, which serves as a characteristic parameter of the shared face area between adjacent prisms. Specifically, this includes: first, identifying all pairs of dynamically regulated triangular prisms that are adjacent to each other in the saccharification and boiling process space, i.e., two triangular prisms that share a geometric face; for each pair of adjacent triangular prisms, accurately locate the shared face they jointly form. This shared face is a quadrilateral plane, formed by the two vertices of each of the two triangular prisms, for a total of four independent spatial vertices, denoted as . ; to calculate the area value of the shared surface The quadrilateral is decomposed into two non-overlapping triangles. The areas of each triangle are calculated separately and then summed. The specific calculation process uses the vector cross product method, first calculating the area from the vertex... Area of ​​the triangle formed The formula is Then calculate the value from the vertex. Area of ​​the triangle formed The formula is Ultimately, the total area of ​​the shared surface It can be obtained using the following formula: ; This refers to the shared surface area characteristic parameter between adjacent prisms, which effectively characterizes the spatial contact range and connection tightness of two adjacent process parameter regions.

[0078] Step 5.6: Calculate the normal vector of the shared surface based on the spatial coordinates of the four vertices constituting the shared surface. This vector serves as the normal characteristic parameter of the shared surface in relation to adjacent prisms. Specifically, this includes: after completing the shared surface identification and area calculation, further calculating the normal vector of the shared surface as the normal characteristic parameter representing the spatial orientation of the shared surface. The calculation of the shared surface normal vector is based on the plane formed by any three (non-collinear) vertices of the four vertices constituting the shared surface; using the vertices... For example, first construct two direction vectors located on a shared plane. and The coordinates are subtracted to obtain the vector; then, the cross product of these two vectors is performed, and the resulting vector is the normal vector of the plane, whose mathematical expression is: ; where vector ,vector The cross product operation follows the rules of the three-dimensional vector cross product; the calculated result is... This refers to the normal characteristic parameter of the shared surface. This vector is perpendicular to the plane where the shared surface is located. Its direction represents the geometric normal at the spatial connection of two adjacent dynamically controlled triangular prisms, providing a key directional reference for the spatial splicing of polyhedra, collision detection, and analysis of the overall process feature field.

[0079] In a preferred embodiment of the present invention, step 5 above may include:

[0080] Step 5.7: Extract the target state vector corresponding to the current operating batch. Generate a target prism using the same mapping and construction method as the dynamically controlled triangular prism. Simultaneously, obtain the target volume feature parameters, target flow velocity axial projection feature parameters, target temperature axial projection feature parameters, target shared surface area feature parameters, and target shared surface normal feature parameters corresponding to the target prism using the same calculation method as the feature parameters of the dynamically controlled triangular prism. Specifically, this includes: extracting the target state vector completely corresponding to the current operating batch from the preset process formula library based on the determined process formula number matched by the current production batch. This target state vector contains standard process parameters in three dimensions: height, flow velocity, and temperature. Then, the target state vector is mapped and constructed using the same method as the dynamic control triangular prism. A three-dimensional feature space mapping rule, cross-section division method, vertex generation method, and spatial topology construction method that are completely consistent with the dynamic control triangular prism are constructed. In the same three-dimensional feature space, an ideal geometric structure corresponding one-to-one with the current position of the dynamic control triangular prism is generated, namely the target prism. Then, the same calculation logic and mathematical formulas as those used to calculate the feature parameters of the dynamic control triangular prism are used to solve the target volume feature parameters, target flow velocity axial projection feature parameters, target temperature axial projection feature parameters, target shared surface area feature parameters, and target shared surface normal feature parameters of the target prism, forming a complete target parameter system that corresponds one-to-one with the current actual parameters. All target parameters are identified by the subscript obj to clearly distinguish them from the actual parameters.

[0081] Step 5.8: Obtain the calculated characteristic parameters of the current dynamically controlled triangular prism, including volume characteristic parameters, flow velocity axial projection characteristic parameters, temperature axial projection characteristic parameters, shared surface area characteristic parameters, and shared surface normal characteristic parameters. Specifically, this involves sequentially retrieving the actual characteristic parameters calculated in steps 5.2 to 5.6 that correspond to the current process state. These parameters include the volume characteristic parameters of the dynamically controlled triangular prism, the flow velocity axial projection characteristic parameters corresponding to each edge, the temperature axial projection characteristic parameters corresponding to each edge, the shared surface area characteristic parameters between adjacent prisms, and the shared surface normal characteristic parameters. All actual characteristic parameters are uniformly identified using the subscript "real". During the retrieval process, the validity of the numerical values ​​and the dimensional matching of each parameter are verified to ensure that there are no missing, misaligned, or calculation anomalies, thus ensuring that the parameters participating in subsequent comparisons are all valid and usable actual process characteristic data.

[0082] Step 5.9 involves comparing each characteristic parameter of the currently dynamically controlled triangular prism with the target characteristic parameters of the corresponding target prism, calculating the volume deviation rate, flow velocity axial projection deviation vector, temperature axial projection deviation vector, shared surface area deviation rate, and shared surface normal deviation angle as five deviation components. Specifically, this includes accurately comparing each actual characteristic parameter of the currently dynamically controlled triangular prism with the target characteristic parameters of the corresponding target prism according to parameter type, and calculating five independent deviation components, including the volume deviation rate. The calculation formula is: ;in These are the actual volume characteristic parameters. For target volume characteristic parameters, the axial projection deviation vector of the flow velocity. The vector difference between the axial projection characteristic parameters of the actual flow velocity and the axial projection characteristic parameters of the target flow velocity is calculated using the following formula: in These are the axial projection characteristic parameters of the actual flow velocity. The characteristic parameters are the axial projections of the target flow velocity.

[0083] Temperature axial projection deviation vector The vector difference between the axial projection characteristic parameters of the actual temperature and the axial projection characteristic parameters of the target temperature is calculated using the following formula: ;in These are the axial projection characteristic parameters of the actual temperature. The target temperature axial projection characteristic parameter, shared surface area deviation rate The calculation formula is: ;in These are the actual shared surface area characteristic parameters. For the target shared surface area characteristic parameters, the shared surface normal deviation angle is... The spatial angle between the actual shared surface normal vector and the target shared surface normal vector is calculated using the following formula:

[0084] ;in For the actual shared surface normal characteristic parameters, As the target shared surface normal characteristic parameter, the above five calculation results together constitute a set of independent and dimensionally clear deviation components, which characterize the degree of deviation between the current process state and the target process state from different perspectives.

[0085] Step 5.10: The five deviation components are weighted and summed according to preset weighting coefficients to generate a comprehensive quantitative index, which serves as the control distortion index characterizing the degree of agreement between the current control state and the target state. Specifically, this includes: pre-setting normalized weighting coefficients for each of the five deviation components: volume deviation rate, velocity axial projection deviation vector magnitude, temperature axial projection deviation vector magnitude, shared surface area deviation rate, and shared surface normal deviation angle. Each weighting coefficient is a standardized value greater than zero and summed to 1. The weighting coefficients are denoted as follows: After dimensionless normalization of each deviation component, a weighted summation operation is performed according to preset weights to generate a unique comprehensive quantitative value. This comprehensive quantitative value is the regulation distortion index, which characterizes the overall degree of conformity between the current actual regulation state and the ideal target regulation state. Its calculation formula is ;in Let the magnitude of the axial projection deviation vector of the flow velocity be denoted as . The magnitude of the temperature axial projection deviation vector is denoted by . The smaller the value of the distortion index, the closer the current process flow field, temperature field and structural spatial distribution are to the target state, and the lower the degree of distortion. Conversely, the larger the value, the higher the overall deviation, and the more powerful the closed-loop control needs to be implemented.

[0086] In a preferred embodiment of the present invention, step 6 above may include:

[0087] Step 6.1: Obtain the calculated control distortion index and compare it with the preset distortion threshold. This includes: accurately retrieving the calculated control distortion index and verifying its validity to ensure there are no calculation errors or missing values, thus ensuring the accuracy and reliability of the basic data used for judgment; simultaneously, extracting the preset distortion threshold corresponding to the process formula of the current production batch from the set process control parameter library. The preset distortion threshold is a critical judgment value calibrated after multiple process experiments, simulations, and actual production verifications, taking into account the core precision requirements of the saccharification and boiling process, wort quality standards, and the stability of the guide hood control. It is not a universal fixed value, but rather a precise match with the target state vector and the structural parameters of the dynamic control prism of the current process formula. Different process formulas and different wort production specifications correspond to different distortion thresholds, avoiding judgment deviations caused by universal thresholds and ensuring the specificity of the judgment standard.

[0088] Furthermore, the preset distortion threshold is divided into different deviation levels and corresponding intervals. It not only sets the critical value of acceptable deviation, but also provides a quantitative basis for the gradient derivation of dynamic compensation. Its value is strictly adapted to the allowable fluctuation range of wort flow rate and temperature in the saccharification and boiling process, as well as the precision limit of the guide hood height adjustment. This ensures that the judgment result meets the production quality requirements, and will not lead to over-control or process fluctuation due to an overly strict threshold setting, or to excessive process deviation due to an overly lenient threshold setting. The qualified control distortion index is compared with the preset distortion threshold one-to-one with a precise value. This not only clarifies the relationship between the two, but also calculates the specific deviation range between the control distortion index and the distortion threshold in detail. The comparison results and deviation data are recorded simultaneously, providing a direct and accurate basis for the control status judgment and compensation amount derivation.

[0089] Step 6.2: If the control distortion index is less than or equal to the preset distortion threshold, the current control state is determined to meet the accuracy requirements, and no dynamic compensation is needed. Specifically, if the above numerical comparison results show that the control distortion index is less than or equal to the preset distortion threshold, then the five calculated deviation components are further combined to comprehensively determine that the current control state fully meets the preset accuracy requirements of the process. Among them, the actual height of the guide hood, the spatial distribution of wort flow rate and temperature are all within an acceptable range of deviation from the target process state, the geometric shape of the dynamically controlled triangular prism has a high degree of consistency with the target prism model, and there is no obvious distortion in the process flow field and temperature field. At this time, an accurate judgment result without compensation is generated, no adjustment is made to the existing control parameters, no dynamic compensation is generated for the height of the guide hood, the current guide hood height and various process parameters are kept stable, and the continuity of the saccharification and boiling process is ensured.

[0090] Step 6.3: If the controlled distortion index is greater than the preset distortion threshold, then based on the magnitude of the controlled distortion index and its deviation from the preset distortion threshold, the dynamic compensation amount for the guide shield height is derived according to the preset compensation amount mapping relationship. Specifically, if the above numerical comparison results show that the controlled distortion index is greater than the preset distortion threshold, then a precise determination of the degree of deviation is required. Combining the calculated five deviation components, a comprehensive analysis of the core sources of deviation is conducted, focusing on the volume deviation rate and the shared surface normal deviation angle, which are directly related to the guide shield height. Simultaneously, the axial projection deviation vector of the flow velocity and temperature are also considered. The specific values ​​of the axial projection deviation vector and the shared surface area deviation rate are used to confirm whether the current deviation is caused by the excessively high or low height of the guide shroud, or indirectly by the distortion of the wort flow rate and temperature spatial distribution caused by the height deviation. This ensures accurate location of the root cause of the deviation and provides a targeted basis for the subsequent compensation amount derivation. The specific value of the control distortion index and its deviation range from the preset distortion threshold are retrieved. At the same time, combined with the preset deviation level corresponding range, the current deviation level is determined. Different deviation levels correspond to different compensation gradients. The larger the deviation amplitude and the higher the deviation level, the larger the required dynamic compensation amount gradient is, to avoid insufficient or excessive compensation.

[0091] The pre-established compensation mapping relationship is invoked. This mapping relationship is constructed based on a large amount of experimental data, simulation results, and actual production verification data of the saccharification and boiling process. It clarifies the one-to-one correspondence between different deviation levels, different deviation amplitudes, and dynamic compensation amounts, and fully adapts to the target state vector of the current process formula, the structural parameters of the guide shroud, and the adjustment accuracy of the drive mechanism. The mapping relationship also presets upper and lower limit thresholds for the compensation amount. The upper limit does not exceed the maximum adjustment range of the guide shroud drive mechanism, and the lower limit is not lower than the minimum adjustment accuracy allowed by the process, ensuring that the compensation amount meets the actual control requirements. In the derivation process, combined with the current... The specific values ​​of the previous out-of-tolerance level, deviation magnitude, and five deviation components are used to fine-tune the basic compensation amount in the mapping relationship. For example, when the normal deviation angle of the shared surface is too large, the compensation amount is appropriately increased to quickly correct the guide cover height deviation. When the flow rate and temperature projection deviations are small, the compensation amount is appropriately reduced to avoid process fluctuations. Finally, a dynamic compensation amount that accurately matches the current out-of-tolerance level, meets process requirements, and is compatible with the drive mechanism is derived. This compensation amount clarifies the specific value that the guide cover height needs to be adjusted, and also marks the reference basis for the adjustment direction, providing accurate and reliable compensation data support for deviation vector correction.

[0092] Step 6.4: Obtain the control deviation vector, and superimpose the dynamic compensation amount onto the deviation value corresponding to the guide hood height dimension in the control deviation vector to form a corrected control deviation vector. Specifically, this includes: first, retrieving the generated original control deviation vector, verifying the completeness and validity of the vector, confirming that the deviation values ​​of the four dimensions of guide hood height, wort temperature, steam flow rate, and pressure inside the pot are all present and without any abnormalities, focusing on locating the deviation value corresponding to the guide hood height dimension in the vector, and confirming the meaning of the positive and negative values: a positive value indicates that the current guide hood height is higher than the target height, and a negative value indicates that the current guide hood height is higher than the target height. The height is lower than the target height. Then, the obtained dynamic compensation amount is precisely algebraically superimposed with the height dimension deviation value. The superposition process strictly follows the physical direction rules of height adjustment. That is, if the dynamic compensation amount is positive, the height deviation value is increased (the corresponding guide hood needs to be adjusted downwards). If the dynamic compensation amount is negative, the height deviation value is decreased (the corresponding guide hood needs to be adjusted upwards). At the same time, the deviation values ​​of the three dimensions of wort temperature, steam flow rate, and pressure inside the pot in the control deviation vector remain unchanged, without affecting the control of other process parameters. Finally, a corrected and optimized control deviation vector is formed, which provides accurate input data for the generation of adjustment commands.

[0093] Step 6.5: Using the corrected control deviation vector as input, convert it into an executable guide cover height adjustment command according to the preset command generation rules. This serves as the optimized control command. Specifically, this includes: using the corrected control deviation vector as core input data, first parsing the vector to extract the corrected deviation value of the guide cover height dimension, confirming the direction and specific amplitude of the guide cover adjustment; based on the preset command generation rules, which are formulated by combining the operating characteristics of the guide cover drive mechanism, the adjustment accuracy requirements, and the dynamic response requirements of the saccharification and boiling process, it covers four core aspects: parameter mapping, adjustment gradient matching, safety limit, and response speed adaptation. This enables precise conversion of the corrected height deviation value into an identifiable and executable adjustment signal for the drive mechanism. Specifically, the parameter mapping rules... The correspondence between the corrected height deviation value and the adjustment range was determined to ensure that the adjustment range and the deviation value are accurately matched. The larger the deviation, the larger the adjustment range, but it does not exceed the preset gradient upper limit. The adjustment gradient matching rule divides different adjustment levels according to the deviation range. Small deviations correspond to fine adjustment levels, and large deviations correspond to fast adjustment levels, taking into account both adjustment accuracy and efficiency. The safety limit rule presets the upper and lower limit values ​​of the guide hood height adjustment. If the converted adjustment signal exceeds the limit, it is automatically corrected to the limit value to avoid excessive adjustment of the guide hood, which may cause mechanical damage or process abnormalities. The response speed adaptation rule combines the dynamic response requirements of the saccharification and boiling process and adjusts the adjustment speed according to the current process stage (such as the initial, middle, and late stages of boiling) to avoid excessive adjustment causing wort flow field disturbances and temperature fluctuations, and excessive adjustment making it difficult for the deviation to converge quickly.

[0094] Through this preset rule, the corrected height deviation value can be converted into an adjustment signal containing key information such as adjustment direction (determined by the sign of the deviation, positive values ​​are adjusted downwards and negative values ​​are adjusted upwards), adjustment range (matching the deviation value with the gradient level), adjustment speed (adapting to the process stage), and safety verification parameters. This ensures that the adjustment signal is effectively matched with the operating parameters of the guide shroud drive mechanism (such as drive power, adjustment accuracy, and response speed), avoiding adjustment errors caused by incompatible commands. Finally, the converted adjustment signal is integrated into a complete guide shroud height adjustment command. This command, as the optimized final control command, can be directly issued to the guide shroud drive mechanism after validity verification. It drives the guide shroud to perform precise height adjustment actions, gradually reducing the deviation between the current process state and the target state, and causing the process flow field and temperature field to gradually converge towards the target state, ensuring the stability of the saccharification and boiling process and the quality of the wort.

[0095] In a preferred embodiment of the present invention, step 7 above may include:

[0096] Step 7.1: Drive the guide cover to perform a height adjustment action according to the optimized control command. After the adjustment action is completed, obtain the actual height position of the guide cover after adjustment. Specifically, this includes: sending the generated optimized guide cover height adjustment command to the guide cover drive mechanism after final validity verification; the drive mechanism smoothly executes the guide cover height adjustment action according to the adjustment direction, adjustment range, and adjustment speed in the command; during the adjustment process, the operating status of the drive mechanism is monitored in real time to avoid abnormal situations such as jamming or over-adjustment; after the guide cover drive mechanism stops running and the adjustment action is completely completed, accurately collect the actual height position data of the guide cover after adjustment; perform noise reduction and calibration processing on the collected height data to confirm that the data is free of abnormalities and deviations; and simultaneously record the adjustment completion time and the actual height value to provide basic height data for the generation of the feedback state vector.

[0097] Step 7.2: Following the same data acquisition method as the initial state vector acquisition and control, after the guide hood height adjustment is completed, acquire the real-time height position data and synchronized process parameters again to generate a feedback state vector. Specifically, this includes: strictly adhering to the same data acquisition method, frequency, location, and processing standards as the initial state vector acquisition and control in Step 1.2; after the guide hood height adjustment is completed, allow it to stand for a preset time to ensure the saccharification and boiling process stabilizes, avoiding data distortion caused by process fluctuations after adjustment; then conduct comprehensive data acquisition again, including real-time height position data after guide hood adjustment, and various process parameters synchronized with the height data, specifically covering wort flow rate, wort temperature, and steam flow rate. Core process parameters such as flow rate and pressure inside the pot are collected. Among them, the wort flow rate and temperature are still collected at the three key sections determined in step 3, and the collection points for steam flow rate and pressure inside the pot are completely consistent with those in step 1.2. During the collection process, a synchronous collection trigger mechanism is activated to ensure that all parameters are collected at the same time point, eliminating problems such as data asynchrony and missing data. All collected data undergoes multiple rounds of screening and calibration to remove abnormal data caused by sensor fluctuations and instantaneous process disturbances. Then, according to the preset vector dimension order, the calibrated parameters are integrated in an orderly manner to form a feedback state vector. This vector is completely consistent with the number of dimensions, parameter types, and arrangement order of the initial control state vector, providing an accurate and suitable data foundation for dimension-by-dimensional comparison and ensuring the reliability of the comparison results.

[0098] Step 7.3: The feedback state vector and the initial control state vector are compared dimension by dimension according to the preset vector dimension order to calculate the deviation value of each dimension. By comprehensively processing the deviation values ​​of all dimensions, the accuracy deviation value of this adjustment is obtained. Specifically, this includes: retrieving the generated feedback state vector and the generated initial control state vector, and re-verifying the validity and completeness of the two vectors to confirm that there is no missing data or parameter misalignment. The common dimension order of the two vectors is determined, that is, they are strictly arranged in the fixed dimension order of guide hood height, wort flow rate, wort temperature, steam flow rate, and pot pressure to ensure that each dimension corresponds one-to-one, without misalignment or omission. The wort flow rate is a dedicated parameter of the control base point data and is only used for three-dimensional feature space modeling and dynamic control prism construction, and is not included in the state vector dimension composition.

[0099] The two vectors are precisely compared dimension by dimension according to the preset vector dimension order. For each dimension, the corresponding parameter value in the feedback state vector is subtracted from the corresponding parameter value in the control initial state vector to calculate the independent deviation value of each dimension. The sign (indicating the direction of deviation) and magnitude (indicating the degree of deviation) of the deviation value of each dimension are recorded simultaneously to confirm the change of each dimension parameter compared with the initial state. Then, the deviation values ​​of all dimensions are comprehensively processed. Weight coefficients adapted to process requirements are set for each dimension in advance. The three core dimensions directly related to the control effect are given the weight of the guide hood height, wort flow rate and wort temperature. The two auxiliary dimensions of steam flow rate and pot pressure are also taken into account. All deviation values ​​are standardized to eliminate the influence of the unit difference of the parameters of different dimensions. Finally, the accuracy deviation value of the guide hood height adjustment is obtained. This value directly reflects the deviation between the actual effect and the expected effect of the adjustment action, and provides a quantitative basis for calibration and correction.

[0100] Step 7.4: The accuracy deviation value is used as a calibration factor and input into the construction method of the dynamic control prism to correct at least one of the control base point screening rules, adjacent base point determination rules, and prism connection rules used when constructing the dynamic control prism, generating an updated construction method. Specifically, this includes: using the calculated accuracy deviation value as a core calibration factor. This core calibration factor is a quantitative characterization parameter of the height adjustment effect of this guide cover. It is a comprehensive quantitative value obtained after weight allocation and standardization based on the dimensional deviation values ​​between the feedback state vector and the initial control state vector. Its value directly reflects the degree of deviation between the actual effect and the expected effect of this adjustment action. The larger the value, the larger the adjustment deviation and the worse the control effect. The smaller the value, the closer the adjustment effect is to the expectation. At the same time, its positive or negative sign can help determine the direction of deviation and provide directional guidance for subsequent rule correction.

[0101] The core calibration factor is directly input into the construction method of the dynamically controlled triangular prism in step 4. First, a comprehensive analysis of the magnitude and sign of the core calibration factor is conducted to identify the core source of the adjustment deviation. Then, the key rules used in constructing the dynamically controlled triangular prism are specifically modified. The modification scope should at least cover one of the following: the control base point screening rule, the adjacent base point determination rule, and the prism connection rule. This ensures that the modified construction method can accurately adapt to the adjusted process state and improve the accuracy of the next round of control. If the core calibration factor value is large, it indicates a deviation in the control base point acquisition. The control base point screening rule should be modified, the accuracy threshold of base point acquisition optimized, control base points with large deviations removed, and supplemented. Collect more accurate base point data; if the analysis finds that there is a deviation in the determination of adjacent base points, resulting in deviation in the construction of the triangular prism, then correct the adjacent base point determination rules, adjust the determination threshold of the spatial straight-line distance between base points, and optimize the determination logic to ensure the accuracy of adjacent base point determination; if there is misalignment in the connection of the prism, affecting the accuracy of the feature parameter calculation, then correct the prism connection rules, optimize the vertex connection logic and geometric topology relationship to ensure that the connection of the six vertices is accurate and without misalignment; after the correction is completed, integrate all the optimized rules to generate an updated dynamic control triangular prism construction method, providing calibrated rule support for the accurate construction of the dynamically controlled triangular prism in the next round of process control, gradually reducing control deviation and improving the overall process control accuracy.

[0102] like Figure 2 As shown, embodiments of the present invention also provide a guide cover adjustment data acquisition and accuracy calibration system, comprising:

[0103] The status acquisition and fusion module is used to acquire real-time height position data of the guide hood during the saccharification and boiling process, and combine it with synchronously acquired process parameters to generate an initial control state vector with process identifier.

[0104] The deviation vector calculation module is used to compare the initial state vector of regulation with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector.

[0105] The critical section and base point acquisition module is used to determine the positions of multiple critical sections of the guide shroud based on the control deviation vector, and to acquire flow velocity and temperature parameters at each critical section to form a set of control base point data with spatial coordinates.

[0106] The 3D spatial mapping and prism construction module is used to map the control base point data to a 3D feature space consisting of height, flow rate and temperature, and to construct multiple dynamic control prisms, with the vertex of each dynamic control prism corresponding to the coordinates of a control base point.

[0107] The distortion index calculation module is used to calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space, so as to obtain the control distortion index that characterizes the degree of fit between the current control state and the target state.

[0108] The dynamic compensation and command generation module is used to compare the control distortion index with the preset threshold. If it exceeds the threshold, the dynamic compensation amount for the height of the guide cover is derived based on the control distortion index to correct the control deviation vector and generate the optimized control command.

[0109] The adjustment execution and accuracy calibration module is used to drive the guide cover to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

[0110] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0111] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for data acquisition and accuracy calibration of guide cover adjustment, characterized in that, The method includes: Real-time height position data of the guide hood during the saccharification and boiling process is collected, and combined with the synchronously acquired process parameters, an initial state vector with process identifier is generated for regulation. The initial state vector of regulation is compared with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector. The positions of multiple key sections of the guide shroud are determined based on the control deviation vector, and flow velocity and temperature parameters are collected at each key section to form a set of control base point data with spatial coordinates; The control base point data is mapped to a three-dimensional feature space consisting of height, flow rate and temperature to construct multiple dynamic control prisms, with each vertex of the dynamic control prism corresponding to the coordinates of a control base point. Calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space to obtain the control distortion index, which characterizes the degree of fit between the current control state and the target state. The control distortion index is compared with a preset threshold. If it exceeds the threshold, a dynamic compensation amount for the height of the guide cover is derived based on the control distortion index to correct the control deviation vector and generate an optimized control command. The guide cover is driven to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

2. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 1, characterized in that, Real-time height position data of the guide hood during the saccharification and boiling process is collected, and combined with synchronously acquired process parameters, a control initial state vector with process identifiers is generated, including: The height position signal of the guide cover is acquired in real time according to the preset sampling frequency, which is used as the original height time series data; Simultaneously collect wort temperature, steam flow rate, and pressure parameters inside the pot during the saccharification and boiling process to create a synchronous process time series dataset. The original height time series data and the synchronous process time series dataset are input into a pre-built multi-source information fusion model. The pre-built multi-source information fusion model performs spatiotemporal feature alignment on the original height time series data and the synchronous process time series dataset based on the attention mechanism to identify and remove abnormal fluctuation points. At the same time, interpolation compensation is used to repair missing data and generate enhanced aligned multi-source monitoring data. The aligned multi-source monitoring data is matched with the features of each formula in the preset process formula library to identify the process formula type of the current batch. The aligned multi-source monitoring data is then labeled with the corresponding process formula identifier and collection time tag to generate labeled data with complete process identifier. Following a preset vector dimension order, the parameter values ​​contained in the labeled data are normalized and then combined to generate an initial state vector for regulation.

3. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 2, characterized in that, The initial state vector of the control is compared with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension, forming the control deviation vector, including: The initial state vector of regulation is obtained, and the target state vector corresponding to the current batch is extracted from the preset process formula library. Both the initial state vector and the target state vector contain parameter values ​​in the dimensions of guide hood height, wort temperature, steam flow rate, and pressure inside the pot. The initial state vector and the target state vector are compared element by element according to the preset vector dimension order. The differences between the initial state vector and the target state vector in the dimensions of guide hood height, wort temperature, steam flow rate and pressure inside the pot are calculated respectively, and used as the initial deviation values ​​for each dimension. The initial deviation values ​​corresponding to each dimension are combined according to the preset vector dimension order to generate a complete control deviation vector containing deviation values ​​of four dimensions.

4. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 3, characterized in that, The positions of multiple key sections of the guide shield are determined based on the control deviation vector, and flow velocity and temperature parameters are collected at each key section to form a set of control base point data with spatial coordinates, including: The control deviation vector is analyzed to extract the deviation value of the guide shield height dimension and the rate of change of the deviation value of the guide shield height dimension relative to the previous sampling time, which are used as the input basis for determining the cross-sectional position. Based on the deviation value and rate of change of the height dimension, and combined with the preset structural parameters of the guide shroud, the specific spatial height positions of the upper edge section, the lower edge section, and the middle guide ring section of the guide shroud are calculated respectively, which serve as the position coordinates of multiple key sections; Based on the location coordinates of multiple key sections, the average flow rate parameters and local temperature parameters of the wort at each key section are collected in real time. The location coordinates of each key section are associated and bound with the flow velocity and temperature parameters collected at that section, forming control baseline data composed of spatial coordinates and the flow velocity and temperature values ​​corresponding to the vertical axial height coordinates of the key annular section of the guide shroud.

5. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 4, characterized in that, The control baseline data is mapped to a three-dimensional feature space consisting of height, flow velocity, and temperature to construct multiple dynamic control prisms. Each prism's vertex corresponds to the coordinates of a control baseline, and its edges reflect the gradient changes between various parameters, including: Acquire control baseline data, and map the spatial height coordinates, flow velocity values, and temperature values ​​contained in each control baseline data to the corresponding height axis, flow velocity axis, and temperature axis in the three-dimensional feature space, respectively, to form the spatial coordinate points of each control baseline in the three-dimensional feature space; Spatial distribution feature analysis of the spatial coordinates of multiple control base points in the three-dimensional feature space is performed to identify the spatial coordinates of the three control base points located at adjacent key sections with the shortest straight-line distance between them. Multiple triangle bases are formed by connecting them in pairs. Determine the key section where the base of each triangle is located, identify the preset key section adjacent to this key section on the height coordinate axis, extend the three vertices of the base of the triangle along the height coordinate axis and connect them to the adjacent preset key sections, and form a dynamic control triangular prism formed by six vertices by the spatial coordinate points corresponding to the control base points corresponding to the projection positions of each vertex on the flow velocity axis and temperature axis. The direction vectors of each edge of the dynamically controlled triangular prism are calculated. The direction vector of each edge represents the trend of change between the two connected control base points in the velocity or temperature dimension.

6. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 5, characterized in that, Calculate the characteristic parameters of each dynamically controlled prism, including: Obtain the generated dynamic control triangular prisms and extract the spatial coordinates of the six vertices from each dynamic control triangular prism. The volume of each dynamically adjustable triangular prism is calculated based on the spatial coordinates of its six vertices, and is used as the volume characteristic parameter of the prism. Obtain the calculated direction vector of each edge, and calculate the projection length of each edge of each dynamically controlled triangular prism in the direction of the flow velocity axis based on the direction vector, which is used as the flow velocity axial projection characteristic parameter of this prism; The projection length of each edge of each dynamically controlled triangular prism in the temperature axis direction is calculated based on the direction vector, and is used as the temperature axial projection characteristic parameter of this prism. Identify the shared surface between two adjacent dynamically adjustable triangular prisms. The shared surface is formed by two vertices of each of the two adjacent prisms, for a total of four vertices. Calculate the area value of the shared surface based on the spatial coordinates of the four vertices that constitute the shared surface, and use it as a characteristic parameter of the shared surface area of ​​the relationship between the adjacent prisms. The normal vector of the shared surface is calculated based on the spatial coordinates of the four vertices that constitute the shared surface, and is used as the shared surface normal characteristic parameter for the relationship between adjacent prisms.

7. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 6, characterized in that, Calculate the relative positions of each dynamically controlled prism in the feature space, and use it together with the feature parameters to generate a control distortion index characterizing the degree of fit between the current control state and the target state, including: Extract the target state vector corresponding to the current running batch, generate the target prism according to the same mapping and construction method as the dynamic control prism, and obtain the target volume feature parameters, target flow velocity axial projection feature parameters, target temperature axial projection feature parameters, target shared surface area feature parameters, and target shared surface normal feature parameters of the target prism according to the same calculation method as the feature parameters of the dynamic control prism. Obtain the calculated characteristic parameters of the current dynamically controlled triangular prism, including volume characteristic parameters, flow velocity axial projection characteristic parameters, temperature axial projection characteristic parameters, shared surface area characteristic parameters, and shared surface normal characteristic parameters; The current dynamic control triangular prism's various characteristic parameters are compared with the target prism's various target characteristic parameters at the corresponding positions, and the volume deviation rate, flow velocity axial projection deviation vector, temperature axial projection deviation vector, shared surface area deviation rate, and shared surface normal deviation angle are calculated as five deviation components. The five deviation components are weighted and summed according to preset weighting coefficients to generate a comprehensive quantitative index, which serves as the regulation distortion index characterizing the degree of conformity between the current regulation state and the target state.

8. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 7, characterized in that, The control distortion index is compared with a preset threshold. If it exceeds the threshold, a dynamic compensation amount for the guide shield height is derived based on the control distortion index to correct the control deviation vector and generate optimized control commands, including: Obtain the calculated control distortion index and compare the control distortion index with the preset distortion threshold. If the distortion index is less than or equal to the preset distortion threshold, the current control state is determined to meet the accuracy requirements, and there is no need to generate dynamic compensation. If the distortion index is greater than the preset distortion threshold, the dynamic compensation amount for the height of the guide shield is derived according to the preset compensation amount mapping relationship based on the magnitude of the distortion index and the deviation between it and the preset distortion threshold. Obtain the control deviation vector, and superimpose the dynamic compensation amount onto the deviation value of the corresponding guide cover height dimension in the control deviation vector to form a corrected control deviation vector; The corrected control deviation vector is used as input and converted into an executable guide cover height adjustment command according to the preset command generation rules, which serves as the optimized control command.

9. The method for data acquisition and accuracy calibration of guide cover adjustment according to claim 8, characterized in that, The guide cover is driven to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor. This is used to update the construction method of the dynamic control prism in real time, including: The guide cover is driven to perform a height adjustment action according to the optimized control command. After the adjustment action is completed, the actual height position of the guide cover after adjustment is obtained. Following the same data acquisition method as the initial state vector acquisition and control, after the guide cover height adjustment is completed, real-time height position data and synchronized process parameters are acquired again to generate a feedback state vector; The feedback state vector and the initial control state vector are compared dimension by dimension according to the preset vector dimension order to calculate the deviation value of each dimension. By comprehensively processing the deviation values ​​of all dimensions, the accuracy deviation value of this adjustment is obtained. The accuracy deviation value is used as a calibration factor and input into the construction method of the dynamically adjustable prism to correct at least one of the control base point screening rules, adjacent base point determination rules and prism connection rules used when constructing the dynamically adjustable triangular prism, thereby generating an updated construction method.

10. A guide cover adjustment data acquisition and accuracy calibration system, wherein the system implements the method as described in any one of claims 1 to 9, characterized in that, include: The status acquisition and fusion module is used to acquire real-time height position data of the guide hood during the saccharification and boiling process, and combine it with synchronously acquired process parameters to generate an initial control state vector with process identifier. The deviation vector calculation module is used to compare the initial state vector of regulation with the target state vector in the preset process formula to extract the initial deviation values ​​of each dimension and form the regulation deviation vector. The critical section and base point acquisition module is used to determine the position of multiple critical sections of the guide shroud based on the control deviation vector, and to acquire flow velocity and temperature parameters at each critical section to form a set of control base point data with spatial coordinates. The 3D spatial mapping and prism construction module is used to map the control base point data to a 3D feature space consisting of height, flow rate and temperature, and to construct multiple dynamic control prisms, with the vertex of each dynamic control prism corresponding to the coordinates of a control base point. The distortion index calculation module is used to calculate the characteristic parameters of each dynamic control prism and their relative positions in the characteristic space, so as to obtain the control distortion index that characterizes the degree of fit between the current control state and the target state. The dynamic compensation and command generation module is used to compare the control distortion index with the preset threshold. If it exceeds the threshold, the dynamic compensation amount for the height of the guide cover is derived based on the control distortion index to correct the control deviation vector and generate the optimized control command. The adjustment execution and accuracy calibration module is used to drive the guide cover to perform height adjustment according to the optimized control command. After the adjustment is completed, feedback data is collected again to generate a feedback state vector. The feedback state vector is compared with the initial control state vector to calculate the accuracy deviation value of this adjustment as a calibration factor, which is used to update the construction method of the dynamic control prism in real time.

Images