Sterilized milk dynamic sterilization optimization control method and device based on efficient production

Abstract

The invention discloses a sterilized milk dynamic sterilization optimization control method and device based on efficient production, and relates to the related field of dairy product dynamic sterilization optimization control, and the method comprises the steps: respectively connecting temperature, flow, valve feedback and steam control signal links of a target sterilized milk production line in series with a digital potentiometer through an SPI control interface; obtaining a digital potentiometer array; traversing temperature, flow, valve feedback and steam control signal links to perform dynamic sterilization working condition signal time sequence monitoring; performing time sequence memory integration on the multi-source dynamic sterilization condition signal set sequence; switching action mapping analysis of the digital potentiometer array is carried out based on the state characteristics of the multi-source dynamic sterilization working condition signals, and the switching action mapping analysis is distributed to the digital potentiometer array for dynamic sterilization optimization control. The technical problem that existing dynamic sterilization control of sterilized milk is insufficient in sterilization control accuracy and dynamic response capability is solved, and the technical effect of improving the sterilization control accuracy and dynamic response capability is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of dynamic sterilization optimization control of dairy products, and particularly relates to a dynamic sterilization optimization control method and device for sterilized milk based on efficient production. BACKGROUND

[0002] In the industrial production of sterilized milk, the sterilization process is a core link, and its control precision and dynamic response capability directly determine the safety quality, nutrient retention rate and production energy efficiency of the product. At present, the industry generally adopts a single-loop constant value control strategy with a PLC / PID controller as the core, and independently adjusts each actuator by presetting fixed set points of temperature, flow and other parameters. However, due to the strong coupling relationship between temperature, flow, pressure and other variables in the production line, and the slow time-varying of sterilization conditions caused by equipment fouling, changes in raw material properties and other factors, this control method based on fixed parameters and isolated loops is difficult to perceive and adapt to the dynamic correlation characteristics and overall energy efficiency of the system, often showing control lag, overshoot or continuous oscillation, resulting in insufficient or excessive sterilization, high steam energy consumption, and even system instability due to parameter mismatch.

[0003] At present, the dynamic sterilization control of sterilized milk in the related art has the technical problems of insufficient sterilization control precision and dynamic response capability. SUMMARY

[0004] The present application provides a dynamic sterilization optimization control method and device for sterilized milk based on efficient production, which adopts a digital potentiometer array formed by connecting four types of control signal links of the target sterilized milk production line through an SPI control interface and a digital potentiometer in series, performs dynamic sterilization process signal timing monitoring on the four types of links, determines a multi-source dynamic sterilization process signal set sequence, integrates the sequence timing memory, determines the signal state characteristics, analyzes the digital potentiometer array switching action based on the characteristics and distributes it, and performs dynamic sterilization optimization control. The technical means solve the technical problems of insufficient sterilization control precision and dynamic response capability of the existing dynamic sterilization control of sterilized milk, and achieve the technical effect of improving the precision and dynamic response capability of sterilization control.

[0005] The application provides a dynamic sterilization optimization control method for high-efficiency production-based sterilized milk, comprising the following steps: connecting a temperature control signal link, a flow control signal link, a valve feedback control signal link and a steam control signal link of a target sterilized milk production line through SPI control interfaces and digital potentiometers in series to obtain a digital potentiometer array; performing dynamic sterilization working condition signal timing monitoring on the temperature control signal link, the flow control signal link, the valve feedback control signal link and the steam control signal link to determine a multi-source dynamic sterilization working condition signal set sequence; performing timing memory integration on the multi-source dynamic sterilization working condition signal set sequence to determine a multi-source dynamic sterilization working condition signal state feature; and performing switching action mapping analysis of the digital potentiometer array based on the multi-source dynamic sterilization working condition signal state feature and distributing the digital potentiometer array to perform dynamic sterilization optimization control.

[0006] In a possible implementation, the temperature control signal link, the flow control signal link, the valve feedback control signal link and the steam control signal link of the target sterilized milk production line are connected through SPI control interfaces and digital potentiometers in series to obtain a digital potentiometer array, and the following processing is performed: a first digital potentiometer is connected through a first SPI control interface to a thermocouple signal and an amplifier feedback loop in the temperature control signal link; a second digital potentiometer is connected through a second SPI control interface to an RC filter node and a PLC analog input in the flow control signal link; a third digital potentiometer is connected through a third SPI control interface to an output of a valve position transmitter and a PLC analog input in the valve feedback control signal link; a fourth digital potentiometer is connected through a fourth SPI control interface to an AO output of a PLC and an I / P control input in the steam control signal link; and the first digital potentiometer, the second digital potentiometer, the third digital potentiometer and the fourth digital potentiometer are summarized to obtain a digital potentiometer array.

[0007] In a possible implementation, the timing memory integration is performed on the multi-source dynamic sterilization working condition signal set sequence to determine a multi-source dynamic sterilization working condition signal state feature, and the following processing is performed: the multi-source dynamic sterilization working condition signal set sequence is subjected to adjacent signal pre-memory association verification to obtain a memory association verification result sequence; a multi-source dynamic sterilization working condition signal set corresponding to a memory association verification result that fails in verification in the memory association verification result sequence is taken as a segmentation node, and the multi-source dynamic sterilization working condition signal set sequence is segmented to obtain Q multi-source dynamic sterilization working condition signal set subsequences, where Q is a positive integer; the Q multi-source dynamic sterilization working condition signal set subsequences are iterated to perform timing memory integration to determine Q integrated multi-source dynamic sterilization working conditions; and the Q integrated multi-source dynamic sterilization working conditions are subjected to signal state analysis to determine the multi-source dynamic sterilization working condition signal state feature.

[0008] In a possible implementation, the set sequence of the multi-source dynamic sterilization condition signals is subjected to adjacent signal pre-sequence memory correlation verification to obtain a sequence of memory correlation verification results, and the following processing is performed: a first multi-source dynamic sterilization condition signal set and a second multi-source dynamic sterilization condition signal set are extracted from the set sequence of the multi-source dynamic sterilization condition signals; an inner product mapping of adjacent signals of the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set is performed to obtain a first adjacent signal inner product mapping result; when the first adjacent signal inner product mapping result is less than a preset threshold, a first memory correlation verification result corresponding to the second multi-source dynamic sterilization condition signal set is verification failure; when the first adjacent signal inner product mapping result is greater than or equal to the preset threshold, the first memory correlation verification result corresponding to the second multi-source dynamic sterilization condition signal set is verification success; inner product mappings of adjacent multi-source dynamic sterilization condition signal sets in the set sequence of the multi-source dynamic sterilization condition signals are performed, and the results are compared with the preset threshold to obtain the sequence of the memory correlation verification results.

[0009] In a possible implementation, the Q multi-source dynamic sterilization condition signal set subsequences are traversed for time sequence memory integration to determine Q integrated multi-source dynamic sterilization condition signals, and the following processing is performed: a first multi-source dynamic sterilization condition signal set subsequence is extracted from the Q multi-source dynamic sterilization condition signal set subsequences; a first time sequence memory is obtained by performing time sequence memory integration on a first multi-source dynamic sterilization condition signal set and a second multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence; the first time sequence memory is used to perform time sequence memory integration on a third multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence, and the same is repeated, and according to a time sequence memory integration result each time, a next multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence is subjected to time sequence memory integration until a last multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence is reached to obtain a first integrated multi-source dynamic sterilization condition signal; and the first integrated multi-source dynamic sterilization condition signal is added to the Q integrated multi-source dynamic sterilization condition signals.

[0010] In a possible implementation, the first multi-source dynamic sterilization process signal set sub-sequence and the second multi-source dynamic sterilization process signal set are subjected to time sequence memory integration to obtain a first time sequence memory, and the following processing is performed: from the four dimensions of the temperature control signal, the flow control signal, the valve feedback control signal, and the steam control signal, the fine-grained correlation similarity of the first multi-source dynamic sterilization process signal set and the second multi-source dynamic sterilization process signal set is calculated to obtain a first fine-grained correlation similarity set; the first fine-grained correlation similarity set is subjected to adjacency matrix processing to obtain a first adjacency matrix; the second multi-source dynamic sterilization process signal set is subjected to convolution integration by using the first adjacency matrix to obtain the first time sequence memory.

[0011] In a possible implementation, the first fine-grained correlation similarity set is subjected to adjacency matrix processing to obtain a first adjacency matrix, and the following processing is performed: the first fine-grained correlation similarity set is subjected to standardization processing to obtain a first fine-grained correlation similarity standard value set; the first fine-grained correlation similarity standard value set is added to an initially empty matrix to construct the first adjacency matrix.

[0012] In a possible implementation, the switching action mapping analysis based on the multi-source dynamic sterilization process signal state feature is performed and distributed to the digital potentiometer array for dynamic sterilization optimization control, and the following processing is performed: a switching action mapping analyzer is pre-constructed, and the multi-source dynamic sterilization process signal state feature is transmitted to the switching action mapping analyzer for analysis by using the digital potentiometer array to obtain a digital potentiometer switching action array; the digital potentiometer switching action array is distributed to the digital potentiometer array for dynamic sterilization optimization control.

[0013] In a possible implementation, the following processing is performed: the resolution of the digital potentiometer is greater than or equal to 8 bits, and an end-to-end resistance of 10 kΩ-50 kΩ is adopted.

[0014] This application also provides a dynamic sterilization optimization control device for sterilized milk based on high-efficiency production, comprising: a digital potentiometer array acquisition module, used to connect the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line in series with digital potentiometers through an SPI control interface to obtain a digital potentiometer array; a dynamic sterilization condition signal timing monitoring module, used to traverse the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link to perform dynamic sterilization condition signal timing monitoring and determine a multi-source dynamic sterilization condition signal set sequence; a timing memory integration module, used to perform timing memory integration on the multi-source dynamic sterilization condition signal set sequence to determine the multi-source dynamic sterilization condition signal state characteristics; and a dynamic sterilization optimization control module, used to perform switching action mapping and parsing of the digital potentiometer array based on the multi-source dynamic sterilization condition signal state characteristics, and distribute the mapping to the digital potentiometer array for dynamic sterilization optimization control.

[0015] The proposed method and apparatus for dynamic sterilization optimization control of sterilized milk based on high-efficiency production, as described in this application, firstly connects the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line in series with digital potentiometers via an SPI control interface to obtain a digital potentiometer array. Next, it iterates through the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link to monitor the timing of dynamic sterilization operating conditions, determining a multi-source dynamic sterilization operating condition signal set sequence. Then, it integrates the timing memory of this multi-source dynamic sterilization operating condition signal set sequence to determine the state characteristics of the multi-source dynamic sterilization operating condition signals. Finally, based on the state characteristics of the multi-source dynamic sterilization operating condition signals, it performs switching action mapping and analysis of the digital potentiometer array and distributes the data to the digital potentiometer array for dynamic sterilization optimization control. Through the above process, the method and apparatus proposed in this application achieve the technical effect of improving the accuracy and dynamic response capability of sterilization control. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating the dynamic sterilization optimization control method for sterilized milk based on high-efficiency production, as provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of the dynamic sterilization optimization control device for sterilized milk based on high-efficiency production, provided in an embodiment of this application.

[0019] Figure labeling: Digital potentiometer array acquisition module 10, dynamic sterilization condition signal timing monitoring module 20, timing memory integration module 30, dynamic sterilization optimization control module 40. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides a method for optimizing and controlling the dynamic sterilization of sterilized milk based on efficient production, such as... Figure 1 As shown, the method includes: In step S100, the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line are connected in series with digital potentiometers through the SPI control interface to obtain a digital potentiometer array. The resolution of the digital potentiometer is greater than or equal to 8 bits, and an end-to-end resistor of 10kΩ–50kΩ is used.

[0022] Specifically, the system's hardware infrastructure is modified by connecting digital potentiometers in series to the existing critical control loops of the production line. These digital potentiometers are resistive devices precisely controlled by digital signals. They are connected to the central controller via the SPI (Serial Peripheral Interface) communication protocol, forming a digital potentiometer array. In this way, the central controller can dynamically change the resistance values ​​of these digital potentiometers by sending digital commands, thereby fine-tuning the current or voltage signals flowing through these loops and indirectly controlling temperature, flow rate, valves, and steam. Digital potentiometers with a resolution of 8 bits or greater are selected; a resolution of 8 bits or greater means that each potentiometer has at least 256 adjustable levels, providing sufficient adjustment accuracy. A resistance range of 10kΩ–50kΩ is used to match the signal levels of industrial sensors and controllers, ensuring effective signal adjustment without placing excessive load on the existing circuitry.

[0023] For example, in a temperature control loop, the weak voltage signal generated by the thermocouple needs to be amplified before it can be recognized by the PLC. By connecting a digital potentiometer here, the central controller can use SPI to control the digital potentiometer to slightly increase or decrease the resistance, thereby slightly raising or lowering the signal voltage sent to the amplifier, ultimately achieving fine-tuning of the set temperature and ensuring that the sterilization temperature remains stable at the optimal value.

[0024] In one possible implementation, the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line are respectively connected in series with digital potentiometers via an SPI control interface to obtain a digital potentiometer array. Step S100 further includes step S110, where a first digital potentiometer is connected in series between the thermocouple signal and the amplifier feedback loop in the temperature control signal link via a first SPI control interface. Specifically, a digital potentiometer is selected and connected in series between the temperature sensor, i.e., the thermocouple, and the subsequent signal amplifier. Since the thermocouple signal is a weak voltage in the millivolt range, connecting a resistor in series here allows for very precise biasing or attenuation of the original voltage signal by changing the resistance value. The resistance value and adjustment range of the digital potentiometer are set according to the temperature control requirements. For example, a 12-bit resolution, 20kΩ digital potentiometer is selected and connected in series in the temperature control signal link to adjust the amplification factor of the thermocouple signal.

[0025] Step S120: Connect the second digital potentiometer in series with the RC filter node and the PLC analog input in the flow control signal link via the second SPI control interface. Specifically, connect the digital potentiometer to the path of the flow transmitter output signal after RC filtering and before entering the PLC analog input interface. The RC filter is used to eliminate high-frequency noise. Set the resistance value and adjustment range of the digital potentiometer according to the flow control requirements. For example, select a 10-bit resolution, 15kΩ digital potentiometer and connect it in series with the flow control signal link to adjust the output signal of the RC filter node.

[0026] In step S130, the third digital potentiometer is connected in series between the output of the valve position transmitter and the analog input of the PLC in the valve feedback control signal link via the third SPI control interface. Specifically, similar to step S120, the digital potentiometer is connected in series between the valve position transmitter and the PLC, where the valve position transmitter is used to report the actual valve opening. The resistance value and adjustment range of the digital potentiometer are set according to the valve feedback control requirements. For example, an 8-bit resolution, 10kΩ digital potentiometer is selected and connected in series in the valve feedback control signal link to adjust the output signal of the valve position transmitter.

[0027] Step S140: Connect the fourth digital potentiometer in series with the PLC's AO output and I / P control input in the steam control signal link via the fourth SPI control interface. Specifically, connect the digital potentiometer in series between the PLC's analog output module and the current / pressure converter. The PLC outputs a control signal to the I / P converter through the AO module, and the I / P converter proportionally converts it into a pressure signal to drive the steam valve. Set the resistance value and adjustment range of the digital potentiometer according to the steam control requirements. For example, select a 12-bit resolution, 25kΩ digital potentiometer and connect it in series with the steam control signal link to adjust the PLC's AO output signal.

[0028] Step S150: Combine the first, second, third, and fourth digital potentiometers to obtain a digital potentiometer array. Specifically, the four independently installed digital potentiometers are logically and physically assembled into a unified array that can be centrally addressed and managed by the central controller. The controller can send control commands to any one of the digital potentiometers in the array simultaneously or individually. For example, using an SPI interface manager, the four digital potentiometers can be combined into a single digital potentiometer array and controlled uniformly via the SPI interface.

[0029] Step S200: Traverse the temperature control signal link, flow control signal link, valve feedback control signal link and steam control signal link to perform dynamic sterilization condition signal timing monitoring, and determine the multi-source dynamic sterilization condition signal set sequence.

[0030] Specifically, real-time data is continuously and cyclically collected from the four control links already embedded with digital potentiometers. This data is recorded in chronological order, forming a time-varying sequence of signals including four dimensions: temperature, flow rate, valve opening, and steam pressure. The data at each point in time is a set containing four signal values, and a series of time points constitutes a multi-source dynamic sterilization condition signal sequence.

[0031] For example, data is collected once per second. At time T1, the following data is collected: temperature = 135.2℃, flow rate = 1000L / h, valve opening = 65%, steam pressure = 0.55MPa; at time T2, the following data is collected: temperature = 135.5℃, flow rate = 998L / h, valve opening = 66%, steam pressure = 0.56MPa. Combining these time-ordered sets constitutes a multi-source signal sequence reflecting the dynamic changes throughout the sterilization process.

[0032] Step S300: Perform time-series memory integration on the sequence of multi-source dynamic sterilization operating condition signals to determine the state characteristics of the multi-source dynamic sterilization operating condition signals.

[0033] Specifically, the large amount of time-series data collected in step S200 is processed. The original data sequence is lengthy and contains noise, making direct use for control inefficient. Through time-series memory integration, useful information from the long-term series is extracted and fused into a state feature that represents the overall operating conditions within that time period, used to describe the current state of the signal. For example, an LSTM (Long Short-Term Memory) network is used to perform time-series memory integration on the collected temperature, flow rate, and other signals, and then a feature extraction algorithm is used to determine its state features.

[0034] In one possible implementation, the multi-source dynamic sterilization condition signal set sequence is integrated using temporal memory to determine the state characteristics of the multi-source dynamic sterilization condition signals. Step S300 further includes step S310, which involves performing adjacent signal pre-sequence memory association verification on the multi-source dynamic sterilization condition signal set sequence to obtain a memory association verification result sequence. Specifically, the similarity index between signal sets at two adjacent time points is calculated. If the similarity is high, it indicates stable operating conditions; if the similarity suddenly becomes very low, it indicates a significant change has occurred, such as equipment start-up / shutdown or raw material switching. Adjacent signal pre-sequence memory association verification is used to identify abrupt change points in the sequence.

[0035] Step S320: Using the set of multi-source dynamic sterilization condition signals corresponding to the failed memory association verification results in the memory association verification result sequence as the segmentation node, the multi-source dynamic sterilization condition signal set sequence is segmented to obtain Q multi-source dynamic sterilization condition signal set sub-sequences, where Q is a positive integer. Specifically, using the mutation point found in step S310, i.e., the point of verification failure, as the boundary, the original time series is cut into multiple shorter sub-sequences with relatively stable internal states. Each sub-sequence represents an independent condition stage. For example, if the entire sequence experiences verification failure at times T3 and T15, then the sequence is divided into three sub-sequences: Sub-sequence 1: T1-T2 heating stage, Sub-sequence 2: T4-T14 holding stage, Sub-sequence 3: T16-Tn cooling stage, where n is the last time.

[0036] Step S330: Traverse the Q sets of multi-source dynamic sterilization condition signal subsequences for temporal memory integration to determine Q integrated multi-source dynamic sterilization condition signals. Specifically, for each segmented, internally stationary subsequence, a recursive fusion method is used to fuse the data from all times within the subsequence into a single, information-rich integrated signal. This integrated signal contains all the temporal memories of that subsequence.

[0037] Step S340: Perform signal state analysis on the Q integrated multi-source dynamic sterilization condition signals to determine the state characteristics of the multi-source dynamic sterilization condition signals. Specifically, perform final feature extraction on each integrated signal generated in step S330, including calculating its mean, variance, and other statistical features, frequency domain features, or using a classifier / regressor to determine its state category, such as optimal sterilization state, slight oscillation state, deviation from the set state, etc. Finally, output the state characteristics of the multi-source dynamic sterilization condition signals to determine the current state of the signal. For example: For the integrated signal of the heat preservation stage subsequence, after analysis, it is found that its average temperature is stable at 135℃, but the flow rate has a small periodic fluctuation. Then the state characteristics of this stage can be marked as: State: steady-state heat preservation, health: good, note: slight fluctuation in flow rate.

[0038] In one possible implementation, the sequence of multi-source dynamic sterilization condition signals is subjected to adjacent signal pre-order memory association verification to obtain a memory association verification result sequence. Step S310 further includes step S311, extracting the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set from the sequence of multi-source dynamic sterilization condition signals. Specifically, from the complete signal sequence arranged in chronological order, data sets from two adjacent time points are sequentially extracted for analysis. The first multi-source dynamic sterilization condition signal set refers to the data from the previous time point, and the second multi-source dynamic sterilization signal set refers to the data from the immediately following time point. These two sets constitute a pair of adjacent data whose association needs to be verified.

[0039] Step S312: Map the adjacent signal inner product of the first and second multi-source dynamic sterilization operating condition signal sets to obtain the first adjacent signal inner product mapping result. Specifically, calculate the similarity between the two signal sets using vector inner product for quantitative comparison. First, treat each signal set containing four values—temperature, flow rate, valve opening, and steam pressure—as a vector in a multi-dimensional space. Then, calculate the inner product of these two vectors, namely the vector of the first multi-source dynamic sterilization operating condition signal set and the vector of the second multi-source dynamic sterilization operating condition signal set. The magnitude of the inner product value reflects the overall similarity of the two vectors in direction and amplitude. A larger inner product value indicates that the operating conditions at the two moments are more similar and the changes are more stable; a smaller inner product value indicates that significant changes have occurred.

[0040] Step S313: When the inner product mapping result of the first adjacent signal is less than a preset threshold, the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is deemed a verification failure. Specifically, the calculated inner product result is compared with a preset threshold. When the inner product result is less than the threshold, it indicates a significant difference between the signal sets at two adjacent times, and a sudden change in the operating condition has occurred. Therefore, the association verification between the second multi-source dynamic sterilization condition signal set and the first multi-source dynamic sterilization condition signal set is determined to have failed, and this point is marked as a split point.

[0041] Step S314: When the mapping result of the inner product of the first adjacent signals is greater than or equal to a preset threshold, the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is considered successful. Specifically, also based on the comparison of the inner product and the threshold, when the inner product result is greater than or equal to the preset threshold, it indicates that the signal sets at two adjacent times are highly similar, the operating conditions are stable and continuous, and no sudden changes have occurred. Therefore, the association verification between the second multi-source dynamic sterilization condition signal set and the first multi-source dynamic sterilization condition signal set is deemed successful.

[0042] Step S315 involves performing adjacency signal inner product mapping on all adjacent multi-source dynamic sterilization condition signal sets in the multi-source dynamic sterilization condition signal set sequence, and comparing the result with a preset threshold to obtain the memory association verification result sequence. Specifically, a loop program is used to process each pair of adjacent signal sets in the sequence sequentially through steps S312 to S314. Each comparison generates a success or failure label. Finally, all these labels are arranged in chronological order to form a memory association verification result sequence corresponding to the original data sequence.

[0043] For example, for a sequence containing 100 time points, the above process generates a sequence containing 99 validation results. This result sequence marks all the moments when state changes occur during the entire sterilization process, which is used to provide boundaries for data segmentation.

[0044] In one possible implementation, the Q sub-sequences of the multi-source dynamic sterilization condition signal set are traversed and integrated using time-series memory to determine the Q integrated multi-source dynamic sterilization condition signals. Step S330 further includes step S331, extracting a first multi-source dynamic sterilization condition signal set sub-sequence from the Q sub-sequences of the multi-source dynamic sterilization condition signal set. Specifically, from all the sub-sequences obtained in step S320, one sub-sequence is arbitrarily selected as the current processing object, i.e., the first multi-source dynamic sterilization condition signal set sub-sequence. The global data processing task is decomposed into independent processing tasks for multiple locally stable sub-sequences.

[0045] Step S332: Integrate the first and second sets of multi-source dynamic sterilization signal sets from the first multi-source dynamic sterilization signal set subsequence with temporal memory to obtain the first temporal memory. Specifically, from the currently processed subsequence, extract the data from the first two time points, namely the first and second sets of multi-source dynamic sterilization signal sets. Use fusion methods such as weighted averaging based on attention mechanisms or gated cyclic unit calculations to fuse these two discrete data points into a new comprehensive data representation that includes information from the first two time points. This comprehensive representation is the first temporal memory.

[0046] Step S333: The first temporal memory is used to integrate the third multi-source dynamic sterilization condition signal set of the first multi-source dynamic sterilization condition signal set subsequence. This process is repeated, integrating the next multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence based on the integration result of each iteration, until the last position of the first multi-source dynamic sterilization condition signal set subsequence is reached, thus obtaining the first integrated multi-source dynamic sterilization condition signal. Specifically, the temporal memory integration is recursively iterated, continuously fusing the temporal memory generated in the previous step with the data of the next moment in the subsequence, such as the third, fourth, and so on, until the last position. Each fusion combines new instantaneous data with the current memory, updating the memory to include information over a longer time span. The memory is continuously enriched and updated as each new data point is processed, until the last data point of the subsequence is processed. Finally, the first integrated multi-source dynamic sterilization condition signal is obtained, which is a complete representation of all information in the entire subsequence.

[0047] Step S334: Add the first integrated multi-source dynamic sterilization condition signal to the Q integrated multi-source dynamic sterilization condition signals. Specifically, create a list or array set to store the final result, and store the first integrated multi-source dynamic sterilization condition signal obtained in step S333, which represents all information of the sub-sequences of the first multi-source dynamic sterilization condition signal set, into this set. This set ultimately contains the integrated signals of all Q sub-sequences.

[0048] In one possible implementation, the first and second sets of multi-source dynamic sterilization signal sets in the first multi-source dynamic sterilization signal set subsequence are integrated using temporal memory to obtain a first temporal memory. Step S332 further includes step S3321, calculating the fine-grained correlation similarity between the first and second sets of multi-source dynamic sterilization signal sets from four dimensions: temperature control signal, flow control signal, valve feedback control signal, and steam control signal, to obtain a first fine-grained correlation similarity set. Specifically, the fine-grained similarity analysis no longer treats the entire signal set at two time points as a whole to calculate a single similarity, but instead calculates the numerical correlation between the first and second sets of multi-source dynamic sterilization signal sets on each control variable dimension independently. This correlation can be quantified by calculating the cosine similarity, Pearson correlation coefficient, or Gaussian kernel function based on their difference between the two values ​​on that dimension. Ultimately, we obtain a set containing four similarity values, namely the first fine-grained association similarity set, which describes whether the change patterns of each control variable are consistent between two adjacent time points.

[0049] Step S3322 involves performing adjacency matrix transformation on the first fine-grained association similarity set to obtain the first adjacency matrix. Specifically, the discrete similarity information is structured, that is, the four control variables are treated as four nodes in a fully connected graph, and the pairwise association similarities between the four variables calculated in step S3321 are filled into the diagonal of a 4×4 matrix. This matrix is ​​the first adjacency matrix, which quantitatively describes the strength of the correlation between the dynamic behaviors of each control variable during the transition from the first time step to the second time step.

[0050] Step S3323: The first adjacency matrix is ​​used to perform convolutional integration on the second set of multi-source dynamic sterilization operation signals to obtain the first temporal memory. Specifically, the first adjacency matrix obtained in step S3322 is used as a graph structure in a graph convolutional network, and the four signal values ​​in the second set of multi-source dynamic sterilization operation signals are regarded as the initial features of four nodes in the graph. A graph convolution operation is performed, which essentially allows the information of each node to interact and aggregate with the information of its associated nodes according to the association strength defined by the first adjacency matrix. Through this weighted aggregation, the information in the second set of multi-source dynamic sterilization operation signals is filtered or integrated according to its inherent association pattern, outputting a new feature representation that considers the mutual influence between variables, which is the first temporal memory. The first temporal memory is no longer just the raw data at the second time step, but an enhanced representation that includes information on the state transition pattern from the first to the second time step.

[0051] In one possible implementation, adjacency matrixing of the first fine-grained association similarity set is performed to obtain a first adjacency matrix. Step S3322 further includes step S33221, which standardizes the first fine-grained association similarity set to obtain a first set of fine-grained association similarity standard values. Specifically, the four similarity values ​​in the first fine-grained association similarity set are processed using a standardization algorithm. For example, Min-Max standardization or Z-Score standardization can be used. Through standardization, it is ensured that all similarity values ​​are within a uniform numerical range, eliminating the influence of units and improving numerical stability.

[0052] Step S33222: Add the first set of fine-grained association similarity standard values ​​to the initially empty matrix to construct the first adjacency matrix. Specifically, initialize a 4×4 matrix with all elements equal to zero, and place the standardized values ​​from the first set of fine-grained association similarity standard values ​​on the main diagonal of the matrix to represent the association between each variable and itself. This filled matrix is ​​the first adjacency matrix used for graph convolution operations.

[0053] Step S400: Based on the status characteristics of the multi-source dynamic sterilization working condition signal, perform the switching action mapping and analysis of the digital potentiometer array, and distribute it to the digital potentiometer array for dynamic sterilization optimization control.

[0054] Specifically, the system takes the analyzed multi-source dynamic sterilization condition signal characteristics, which accurately describe the current system operating status, as input. Through a preset control decision mechanism, it parses out the precise resistance adjustment commands (switching actions) required for the four digital potentiometers. These commands are then simultaneously sent to the corresponding digital potentiometers for execution, optimizing the control parameters of the sterilization process. This improves production efficiency, stabilizes product quality, and reduces energy consumption while ensuring sterilization effectiveness.

[0055] In one possible implementation, the switching action mapping of the digital potentiometer array is parsed based on the state characteristics of the multi-source dynamic sterilization operating condition signal, and distributed to the digital potentiometer array for dynamic sterilization optimization control. Step S400 further includes step S410, pre-constructing a switching action mapping parser, and transmitting the state characteristics of the multi-source dynamic sterilization operating condition signal of the digital potentiometer array to the switching action mapping parser for analysis to obtain the digital potentiometer switching action array. Specifically, a switching action mapping parser is pre-designed and constructed. The switching action mapping parser is essentially a control strategy function, which can be implemented based on various technologies. Specifically, a fuzzy rule base can be used to store a series of expert experience rules of "IF (state characteristic) THEN (execute action)". For example, "IF state is 'heating too slow' THEN increase the resistance value of the temperature loop digital potentiometer at level X, increase the resistance value of the steam loop digital potentiometer at level Y". Alternatively, a lookup table can be used, that is, a preset table that lists the optimal digital potentiometer resistance value configuration corresponding to various typical state characteristics. Machine learning models can also be used, such as a neural network model trained on a large amount of historical data. This model can learn the nonlinear mapping relationship between complex state features and optimal control actions, thereby achieving precise control.

[0056] During operation, the multi-source dynamic sterilization condition signal status characteristics obtained from real-time analysis are input into the switching action mapping resolver. After calculation or query, the resolver outputs a set of instructions containing four elements, namely the digital potentiometer switching action array. This array specifies the target resistance value or the range that the first to fourth digital potentiometers need to be adjusted to.

[0057] Step S420 involves distributing the digital potentiometer switching action array to the digital potentiometer array for dynamic sterilization optimization control. Specifically, the central controller, via the SPI communication bus, sends the instructions generated in step S410 in the digital potentiometer switching action array, in the form of digital signals, sequentially or in parallel to each corresponding digital potentiometer in the array. Upon receiving the corresponding instruction, each digital potentiometer drives its internal electronic switch, changing the position of the resistance tap and adjusting the resistance value to the target value required by the instruction. This change in resistance value directly fine-tunes the voltage or current of its corresponding signal link, ultimately transmitting the signal to the actuator, such as a heater or regulating valve, achieving dynamic and refined optimization control of the sterilization process. This perception-decision-execution cycle continues continuously, forming an intelligent control system capable of adapting to changes in operating conditions.

[0058] This application employs a technique that connects four types of control signal links in the target sterilized milk production line to a digital potentiometer array via an SPI control interface. It then traverses these four types of links to monitor the timing of dynamic sterilization signals, determines a sequence of multi-source dynamic sterilization signals, memorizes and integrates this sequence, identifies signal state characteristics, analyzes and distributes the switching actions of the digital potentiometer array based on these characteristics, and performs dynamic sterilization optimization control. This technique solves the technical problems of insufficient sterilization control accuracy and dynamic response capability in existing dynamic sterilization control of sterilized milk, achieving the technical effect of improving sterilization control accuracy and dynamic response capability.

[0059] In the above text, refer to Figure 1 This paper describes in detail a dynamic sterilization optimization control method for sterilized milk based on efficient production, according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a dynamic sterilization optimization control device for sterilized milk based on efficient production, according to an embodiment of the present invention.

[0060] The dynamic sterilization optimization control device for sterilized milk based on high-efficiency production, according to embodiments of the present invention, addresses the technical problems of insufficient sterilization control accuracy and dynamic response capability in existing dynamic sterilization control methods for sterilized milk, thereby improving the technical effect of enhancing sterilization control accuracy and dynamic response capability. The dynamic sterilization optimization control device for sterilized milk based on high-efficiency production includes: a digital potentiometer array acquisition module 10, a dynamic sterilization condition signal timing monitoring module 20, a timing memory integration module 30, and a dynamic sterilization optimization control module 40.

[0061] The digital potentiometer array acquisition module 10 is used to connect the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line in series with digital potentiometers through an SPI control interface to obtain a digital potentiometer array; the dynamic sterilization condition signal timing monitoring module 20 is used to traverse the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link to perform dynamic sterilization condition signal timing monitoring and determine the multi-source dynamic sterilization condition signal set sequence; the timing memory integration module 30 is used to perform timing memory integration on the multi-source dynamic sterilization condition signal set sequence to determine the multi-source dynamic sterilization condition signal state characteristics; the dynamic sterilization optimization control module 40 is used to perform switching action mapping and parsing of the digital potentiometer array based on the multi-source dynamic sterilization condition signal state characteristics and distribute it to the digital potentiometer array for dynamic sterilization optimization control.

[0062] The detailed configuration of the digital potentiometer array acquisition module 10 is explained below: As mentioned above, the temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line are respectively connected in series with digital potentiometers through an SPI control interface to obtain a digital potentiometer array. The digital potentiometer array acquisition module 10 may further include: a first digital potentiometer series connection unit for connecting the first digital potentiometer in series with the thermocouple signal and the amplifier feedback loop in the temperature control signal link through a first SPI control interface; and a second digital potentiometer series connection unit for connecting the second digital potentiometer in series with the second SPI control interface through a second SPI control interface. A PI control interface is connected in series between the RC filter node and the PLC analog input in the flow control signal link; a third digital potentiometer series unit is used to connect the third digital potentiometer in series with the output of the valve position transmitter and the PLC analog input in the valve feedback control signal link via the third SPI control interface; a fourth digital potentiometer series unit is used to connect the fourth digital potentiometer in series with the AO output and I / P control input of the PLC in the steam control signal link via the fourth SPI control interface; a digital potentiometer array generation unit is used to combine the first, second, third, and fourth digital potentiometers to obtain a digital potentiometer array.

[0063] The detailed description of the specific configuration of the time-series memory integration module 30 is as follows: As mentioned above, to perform time-series memory integration on the multi-source dynamic sterilization condition signal set sequence and determine the state characteristics of the multi-source dynamic sterilization condition signal, the time-series memory integration module 30 may further include: an adjacent signal pre-series memory association verification unit for performing adjacent signal pre-series memory association verification on the multi-source dynamic sterilization condition signal set sequence to obtain a memory association verification result sequence; and a sequence segmentation unit for segmenting the memory association verification result sequence into memory association verification result sequences that failed verification. The set of multi-source dynamic sterilization condition signals corresponding to the verification result is used as a segmentation node to segment the sequence of the multi-source dynamic sterilization condition signal set, obtaining Q sub-sequences of multi-source dynamic sterilization condition signals, where Q is a positive integer; the temporal memory integration unit is used to traverse the Q sub-sequences of multi-source dynamic sterilization condition signals for temporal memory integration, and determine Q integrated multi-source dynamic sterilization condition signals; the signal state parsing unit is used to perform signal state parsing on the Q integrated multi-source dynamic sterilization condition signals, and determine the state characteristics of the multi-source dynamic sterilization condition signals.

[0064] The process involves performing adjacency signal pre-order memory association verification on the multi-source dynamic sterilization condition signal set sequence to obtain a memory association verification result sequence. The adjacency signal pre-order memory association verification unit may further include: a multi-source dynamic sterilization condition signal set extraction subunit for extracting a first multi-source dynamic sterilization condition signal set and a second multi-source dynamic sterilization condition signal set from the multi-source dynamic sterilization condition signal set sequence; and an adjacency signal inner product mapping subunit for mapping the adjacency signal inner product of the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set to obtain a first adjacency signal inner product mapping result; verification. The subunit is used to determine that the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is a verification failure when the first adjacent signal inner product mapping result is less than a preset threshold, and the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is a verification success when the first adjacent signal inner product mapping result is greater than or equal to the preset threshold. The memory association verification result sequence acquisition subunit is used to perform adjacent signal inner product mapping on all adjacent multi-source dynamic sterilization condition signal sets in the multi-source dynamic sterilization condition signal set sequence, and compare the result with the preset threshold to obtain the memory association verification result sequence.

[0065] Specifically, the process involves traversing the Q sets of multi-source dynamic sterilization condition signal subsequences for time-series memory integration to determine Q integrated multi-source dynamic sterilization condition signals. The time-series memory integration unit may further include: a first multi-source dynamic sterilization condition signal subsequence extraction subunit for extracting a first multi-source dynamic sterilization condition signal subsequence from the Q sets of multi-source dynamic sterilization condition signal subsequences; a first time-series memory acquisition subunit for integrating the first and second sets of multi-source dynamic sterilization condition signal sets from the first multi-source dynamic sterilization condition signal subsequence with time-series memory to obtain a first time-series memory; and a first integrated multi-source dynamic sterilization condition... The signal acquisition subunit is used to integrate the third multi-source dynamic sterilization condition signal set of the first multi-source dynamic sterilization condition signal set subsequence using the first temporal memory, and so on, to integrate the next multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence according to the result of each temporal memory integration, until the last position of the first multi-source dynamic sterilization condition signal set subsequence is reached, thereby obtaining the first integrated multi-source dynamic sterilization condition signal; Q integrated multi-source dynamic sterilization condition signal acquisition subunits are used to add the first integrated multi-source dynamic sterilization condition signal into the Q integrated multi-source dynamic sterilization condition signals.

[0066] Specifically, the first and second sets of multi-source dynamic sterilization operating condition signal sets in the first multi-source dynamic sterilization operating condition signal set subsequence are integrated with temporal memory to obtain a first temporal memory. The first temporal memory acquisition subunit may further include: a fine-grained correlation similarity calculation component for calculating the fine-grained correlation similarity between the first and second sets of multi-source dynamic sterilization operating condition signal sets from four dimensions: temperature control signal, flow control signal, valve feedback control signal, and steam control signal, respectively, to obtain a first fine-grained correlation similarity set; an adjacency matrix processing component for performing adjacency matrix processing on the first fine-grained correlation similarity set to obtain a first adjacency matrix; and a convolution integration component for using the first adjacency matrix to perform convolution integration on the second set of multi-source dynamic sterilization operating condition signal sets to obtain the first temporal memory.

[0067] Specifically, the adjacency matrixization process of the first fine-grained association similarity set is performed to obtain a first adjacency matrix. The adjacency matrixization process component may further include: a standardization process sub-component for standardizing the first fine-grained association similarity set to obtain a first fine-grained association similarity standard value set; and a first adjacency matrix construction sub-component for adding the first fine-grained association similarity standard value set into an initially empty matrix to construct the first adjacency matrix.

[0068] The specific configuration of the dynamic sterilization optimization control module 40 is described in detail below: As mentioned above, the switching action mapping of the digital potentiometer array is parsed based on the state characteristics of the multi-source dynamic sterilization operating condition signal, and distributed to the digital potentiometer array for dynamic sterilization optimization control. The dynamic sterilization optimization control module 40 may further include: a switching action mapping unit for pre-constructing a switching action mapping parser, and transmitting the state characteristics of the multi-source dynamic sterilization operating condition signal to the switching action mapping parser for analysis to obtain a digital potentiometer switching action array; and a switching action array distribution unit for distributing the digital potentiometer switching action array to the digital potentiometer array for dynamic sterilization optimization control.

[0069] The digital potentiometer array acquisition module 10 may further include: the resolution of the digital potentiometer is greater than or equal to 8 bits, and it uses an end-to-end resistor of 10kΩ–50kΩ.

[0070] The dynamic sterilization optimization control device for sterilized milk based on high-efficiency production provided in this embodiment of the invention can execute the dynamic sterilization optimization control method for sterilized milk based on high-efficiency production provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0071] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0072] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing and controlling the dynamic sterilization of sterilized milk based on efficient production, characterized in that, The method includes: The temperature control signal link, flow control signal link, valve feedback control signal link and steam control signal link of the target sterilized milk production line are respectively connected in series with digital potentiometers through the SPI control interface to obtain a digital potentiometer array; The timing of dynamic sterilization operating condition signals is monitored by traversing the temperature control signal link, flow control signal link, valve feedback control signal link and steam control signal link to determine the set sequence of multi-source dynamic sterilization operating condition signals. The time-series memory integration of the multi-source dynamic sterilization condition signal set sequence is performed to determine the state characteristics of the multi-source dynamic sterilization condition signal; Based on the status characteristics of the multi-source dynamic sterilization operating condition signal, the switching action mapping of the digital potentiometer array is analyzed and distributed to the digital potentiometer array for dynamic sterilization optimization control.

2. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 1, characterized in that, The temperature control signal link, flow control signal link, valve feedback control signal link, and steam control signal link of the target sterilized milk production line are respectively connected in series with digital potentiometers via an SPI control interface to obtain a digital potentiometer array, including: The first digital potentiometer is connected in series with the thermocouple signal and the amplifier feedback loop in the temperature control signal link through the first SPI control interface; The second digital potentiometer is connected in series between the RC filter node and the PLC analog input in the flow control signal link via the second SPI control interface; The third digital potentiometer is connected in series with the output of the valve position transmitter and the analog input of the PLC in the valve feedback control signal link through the third SPI control interface; The fourth digital potentiometer is connected in series between the PLC's AO output and I / P control input in the steam control signal link via the fourth SPI control interface. The first, second, third, and fourth digital potentiometers are combined to obtain a digital potentiometer array.

3. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 1, characterized in that, The time-series memory integration of the multi-source dynamic sterilization condition signal set sequence is performed to determine the state characteristics of the multi-source dynamic sterilization condition signals, including: The sequence of multi-source dynamic sterilization condition signals is subjected to adjacent signal pre-order memory association verification to obtain a memory association verification result sequence. The set of multi-source dynamic sterilization condition signals corresponding to the memory association verification results that failed verification in the memory association verification result sequence is used as the segmentation node to segment the multi-source dynamic sterilization condition signal set sequence to obtain Q multi-source dynamic sterilization condition signal set sub-sequences, where Q is a positive integer; The Q sets of multi-source dynamic sterilization condition signals are traversed and time-series memory integration is performed to determine the Q integrated multi-source dynamic sterilization condition signals. The signal state characteristics of the Q integrated multi-source dynamic sterilization operating condition signals are determined by performing signal state analysis.

4. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 3, characterized in that, The sequence of multi-source dynamic sterilization condition signals is subjected to adjacent signal pre-order memory association verification to obtain a memory association verification result sequence, including: Extract the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set from the multi-source dynamic sterilization condition signal set sequence; Map the inner product of the adjacent signals of the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set to obtain the inner product mapping result of the first adjacent signal; When the mapping result of the inner product of the first adjacent signal is less than the preset threshold, the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is a verification failure. When the mapping result of the inner product of the first adjacent signal is greater than or equal to the preset threshold, the first memory association verification result corresponding to the second multi-source dynamic sterilization condition signal set is verified successfully. The adjacent multi-source dynamic sterilization condition signal sets in the multi-source dynamic sterilization condition signal set sequence are mapped by the adjacent signal inner product, and the result is compared with a preset threshold to obtain the memory association verification result sequence.

5. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 4, characterized in that, The Q sets of multi-source dynamic sterilization condition signals are traversed and integrated using time-series memory to determine the Q integrated multi-source dynamic sterilization condition signals, including: Extract the first multi-source dynamic sterilization condition signal set subsequence from the Q multi-source dynamic sterilization condition signal set subsequences; The first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set of the first multi-source dynamic sterilization condition signal set subsequence are integrated by time sequence memory to obtain the first time sequence memory; The first time-series memory is used to integrate the third multi-source dynamic sterilization condition signal set of the first multi-source dynamic sterilization condition signal set subsequence. Similarly, based on the result of each time-series memory integration, the next multi-source dynamic sterilization condition signal set in the first multi-source dynamic sterilization condition signal set subsequence is integrated with time-series memory until the last position of the first multi-source dynamic sterilization condition signal set subsequence is reached, thereby obtaining the first integrated multi-source dynamic sterilization condition signal. The first integrated multi-source dynamic sterilization condition signal is added to the Q integrated multi-source dynamic sterilization condition signals.

6. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 5, characterized in that, The first and second sets of multi-source dynamic sterilization condition signals from the first multi-source dynamic sterilization condition signal set subsequence are integrated into a time-series memory to obtain a first time-series memory, including: From the four dimensions of temperature control signal, flow control signal, valve feedback control signal and steam control signal, the fine-grained correlation similarity between the first multi-source dynamic sterilization condition signal set and the second multi-source dynamic sterilization condition signal set is calculated to obtain the first fine-grained correlation similarity set. Perform adjacency matrix transformation on the first fine-grained association similarity set to obtain the first adjacency matrix; The first temporal memory is obtained by convolving and integrating the second set of multi-source dynamic sterilization condition signals using the first adjacency matrix.

7. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 6, characterized in that, Perform adjacency matrix transformation on the first fine-grained association similarity set to obtain the first adjacency matrix, including: The first fine-grained association similarity set is standardized to obtain the first fine-grained association similarity standard value set; The first set of fine-grained association similarity standard values ​​is added to the initially empty matrix to construct the first adjacency matrix.

8. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 1, characterized in that, Based on the status characteristics of the multi-source dynamic sterilization operating condition signal, the switching action mapping of the digital potentiometer array is analyzed and distributed to the digital potentiometer array for dynamic sterilization optimization control, including: A pre-constructed switching action mapping resolver is used to transmit the state characteristics of the multi-source dynamic sterilization condition signal to the switching action mapping resolver for analysis, thereby obtaining the digital potentiometer switching action array. The digital potentiometer switching action array is distributed to the digital potentiometer array for dynamic sterilization optimization control.

9. The method for dynamic sterilization optimization control of sterilized milk based on high-efficiency production as described in claim 1, characterized in that, The digital potentiometer has a resolution of 8 bits or more and uses an end-to-end resistor of 10kΩ–50kΩ.

10. A dynamic sterilization optimization control device for sterilized milk based on high-efficiency production, characterized in that, The apparatus is used to implement the dynamic sterilization optimization control method for sterilized milk based on efficient production as described in any one of claims 1-9, and the apparatus comprises: The digital potentiometer array acquisition module is used to connect the temperature control signal link, flow control signal link, valve feedback control signal link and steam control signal link of the target sterilized milk production line in series with digital potentiometers through the SPI control interface to obtain a digital potentiometer array. The dynamic sterilization condition signal timing monitoring module is used to traverse the temperature control signal link, flow control signal link, valve feedback control signal link and steam control signal link to perform dynamic sterilization condition signal timing monitoring and determine the multi-source dynamic sterilization condition signal set sequence. The temporal memory integration module is used to perform temporal memory integration on the set sequence of multi-source dynamic sterilization condition signals to determine the state characteristics of the multi-source dynamic sterilization condition signals. The dynamic sterilization optimization control module is used to perform switching action mapping and analysis of the digital potentiometer array based on the status characteristics of the multi-source dynamic sterilization working condition signal, and distribute it to the digital potentiometer array for dynamic sterilization optimization control.

Images