A real-time monitoring method and system for lactic acid bacteria fermentation process based on blockchain

Abstract

The present invention discloses a real-time monitoring method and system for a lactic acid bacteria fermentation process based on blockchain, which specifically relates to the technical field of production monitoring, and is used to solve the problem of fermentation inconsistency caused by oxygen concentration gradient in the lactic acid bacteria fermentation process; collect oxygen concentration data in a fermentation container and transmit it to a blockchain network; analyze the dynamic change of the acid reaction hysteresis rate by Kalman filtering, and use maximum a posteriori probability estimation to identify the area with significant acid reaction delay; perform tensor decomposition on fermentation process monitoring parameters, evaluate the parameter influence at different positions to identify the area with significant oxygen concentration difference; mark the oxygen concentration gradient risk area based on the area with significant acid reaction delay and the area with significant oxygen concentration difference; dynamically adjust the blockchain data compression rate based on the proportion of risk areas; judge the local instability abnormality by analyzing the redox potential fluctuation rate through reverse differential entropy, and generate oxygen distribution adjustment instructions to control the environment adjustment device to adjust the oxygen distribution.

Description

Technical Field

[0001] The present invention relates to the field of production monitoring technology, and more specifically, to a blockchain-based real-time monitoring method and system for a lactic acid bacteria fermentation process. Background Art

[0002] In the lactic acid fermentation process, in order to ensure the quality of the fermented product and the growth state of the flora, the stability and consistency of the fermentation environment are crucial. The fermentation process based on facultative anaerobic and microaerobic lactic acid bacteria has a certain demand for oxygen supply, but excessive oxygen will have an adverse effect on the anaerobic flora and affect the fermentation efficiency. Therefore, it is particularly important to control the distribution of oxygen during the fermentation process. However, in the prior art, the oxygen in the fermentation environment often has an uneven distribution phenomenon, that is, an oxygen concentration gradient may be formed in the fermentation container.

[0003] The uneven distribution of oxygen causes the bacteria in different locations in the fermentation environment to be exposed to different oxygen concentrations, which in turn affects their metabolic rates. Especially under facultative anaerobic and microaerobic conditions, this difference can significantly affect the metabolic activity of the bacteria, leading to an uneven fermentation process and affecting the consistency and quality of the product.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a blockchain-based method and system for real-time monitoring of a lactic acid bacteria fermentation process to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A real-time monitoring method for lactic acid bacteria fermentation process based on blockchain, comprising the following steps:

[0008] Collect oxygen concentration data in the fermentation container in real time and transmit the oxygen concentration data to the blockchain network;

[0009] The dynamic changes of the acid reaction hysteresis rate were analyzed by Kalman filtering, and the significant acid reaction delay area in the fermentation vessel was identified using maximum a posteriori probability estimation.

[0010] Comprehensively analyze the monitoring parameters of the fermentation process through tensor decomposition, evaluate the influence of each monitoring parameter at different locations in the fermentation vessel, and identify areas with obvious differences in oxygen concentration;

[0011] The area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container is marked as the oxygen concentration gradient risk area;

[0012] Dynamically adjust the compression rate of data on the blockchain based on the proportion of oxygen concentration gradient risk areas;

[0013] The redox potential fluctuation rate in the oxygen concentration gradient risk area is analyzed by the reverse differential entropy algorithm to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal;

[0014] When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated.

[0015] In a preferred embodiment, the oxygen concentration data in the fermentation container is collected in real time, and the oxygen concentration data is transmitted to the blockchain network, specifically:

[0016] Arrange multiple oxygen concentration sensors in the fermentation container to collect oxygen concentration data at each location in real time;

[0017] De-noising and anomaly detection are performed on the collected oxygen concentration data;

[0018] The processed oxygen concentration data is transmitted to the blockchain network in batches, and a timestamp is generated and stored for each set of data on the blockchain;

[0019] The storage node location and timestamp of each set of oxygen concentration data are recorded on the blockchain.

[0020] In a preferred embodiment, the dynamic changes of the acid reaction hysteresis rate are analyzed by Kalman filtering, and the maximum a posteriori probability estimation is used to identify the significant acid reaction delay area in the fermentation container, specifically:

[0021] Using the oxygen concentration data and pH value data stored on the blockchain, the acid reaction hysteresis rate at each location in the fermentation container is calculated;

[0022] Construct a Kalman filter model for the dynamic change of the hysteresis rate of the acid reaction, set the initial state, observation equation and process noise parameters;

[0023] The acid reaction hysteresis rate data at each position are processed by Kalman filtering to obtain a smooth hysteresis rate dynamic change curve;

[0024] Based on the smoothed hysteresis rate curve, the probability value of delayed acidic reaction at each position was calculated using the maximum a posteriori probability estimation method;

[0025] According to the probability value of the acid reaction delay at each position, the area with significant acid reaction delay in the fermentation container is determined.

[0026] In a preferred embodiment, the monitoring parameters of the fermentation process are comprehensively analyzed by tensor decomposition to evaluate the influence of each monitoring parameter at different positions in the fermentation container to identify areas with obvious differences in oxygen concentration, specifically:

[0027] Obtain monitoring parameters at different locations in the fermentation container from the blockchain; the monitoring parameters include oxygen concentration, temperature, pH value and redox potential;

[0028] The extracted monitoring parameters are organized into multidimensional tensors according to spatial positions and time series to form a high-order data structure to represent the multiple correlations of the monitoring parameters;

[0029] High-order singular value decomposition is used to decompose the multidimensional tensor into a set of factor matrices and core tensors, and the influence weights of the main characteristic patterns and monitoring parameters at different positions are extracted;

[0030] According to the factor matrix obtained by tensor decomposition and the influence weights of the monitoring parameters at different positions, the influence of each monitoring parameter at different positions in the fermentation container was evaluated, and the areas with obvious differences in oxygen concentration were identified.

[0031] In a preferred embodiment, the compression rate of the data on the blockchain is dynamically adjusted based on the proportion of the oxygen concentration gradient risk area, specifically:

[0032] In all monitoring areas of the fermentation container, the ratio of the volume of the oxygen concentration gradient risk area to the volume of the total monitoring area of ​​the fermentation container is marked as the risk area ratio;

[0033] The larger the risk area ratio, the lower the compression rate of the data on the blockchain is adjusted.

[0034] In a preferred embodiment, the redox potential fluctuation rate in the oxygen concentration gradient risk area is analyzed by a reverse differential entropy algorithm to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal, specifically:

[0035] Obtain redox potential data for each location within the oxygen concentration gradient risk area and organize it in time series;

[0036] The extracted data were subjected to time difference processing to calculate the redox potential fluctuation rate at each location at different time points;

[0037] The redox potential fluctuation rate is input into the reverse differential entropy model, and the reverse differential entropy value of each position is calculated to obtain a quantitative index of the local instability.

[0038] According to the comparison result of the reverse differential entropy value at each position and the threshold, it is judged whether the local instability degree of the oxygen concentration gradient risk area is abnormal.

[0039] In a preferred embodiment, when the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated, specifically:

[0040] Determine the oxygen concentration gradient characteristics of the oxygen concentration gradient risk area, including concentration distribution and change trend;

[0041] Generate specific oxygen concentration adjustment parameters according to the characteristics of the oxygen concentration gradient, the oxygen concentration adjustment parameters including oxygen flow rate and oxygen flow direction, and transmit the oxygen distribution adjustment instruction to the environment adjustment device in the fermentation container;

[0042] After receiving the oxygen distribution adjustment instruction, the environment adjustment device adjusts the oxygen distribution in the fermentation container according to the oxygen concentration adjustment parameter.

[0043] On the other hand, the present invention provides a real-time monitoring system for lactic acid bacteria fermentation process based on blockchain, including a data acquisition and uploading module, an acid reaction analysis module, a tensor decomposition and analysis module, a risk area marking module, a data compression and adjustment module, a fluctuation anomaly assessment module, and an instruction generation and adjustment module;

[0044] Data collection and upload module: collects oxygen concentration data in the fermentation container in real time and transmits the oxygen concentration data to the blockchain network;

[0045] Acid reaction analysis module: Kalman filtering is used to analyze the dynamic changes of the acid reaction hysteresis rate, and the maximum a posteriori probability estimation is used to identify the area with significant acid reaction delay in the fermentation container;

[0046] Tensor decomposition analysis module: Comprehensively analyze the monitoring parameters of the fermentation process through tensor decomposition, evaluate the influence of each monitoring parameter at different locations in the fermentation container, and identify areas with obvious differences in oxygen concentration;

[0047] Risk area marking module: marking the area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container as the oxygen concentration gradient risk area;

[0048] Data compression adjustment module: dynamically adjusts the compression rate of data on the blockchain based on the proportion of oxygen concentration gradient risk areas;

[0049] Fluctuation anomaly assessment module: The reverse differential entropy algorithm is used to analyze the redox potential fluctuation rate in the oxygen concentration gradient risk area to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal;

[0050] Instruction generation and adjustment module: When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated.

[0051] The technical effects and advantages of the blockchain-based real-time monitoring method and system of lactic acid bacteria fermentation process of the present invention are as follows:

[0052] 1. By combining blockchain technology with real-time monitoring of the lactic acid bacteria fermentation process, the data of the fermentation process is recorded securely, in real time, and with high precision. By transmitting the real-time collected oxygen concentration data to the blockchain network, a stable and secure data foundation is provided for subsequent data analysis and dynamic monitoring. The dynamic changes of the acid reaction hysteresis rate are analyzed with the help of the Kalman filter method, and the areas with significant acid reaction delay in the fermentation container are identified using the maximum a posteriori probability estimation, ensuring that the changes in the acid reaction during the fermentation process can be detected and captured in real time. By tensor decomposition of multidimensional monitoring parameters such as oxygen concentration, temperature, and pH value, the influence of these parameters at different locations can be comprehensively evaluated to effectively identify areas with significant differences in oxygen concentration. Marking the union of the areas with significant acid reaction delay and the areas with significant differences in oxygen concentration as oxygen concentration gradient risk areas helps to accurately locate the key areas that may affect the consistency of fermentation, facilitating subsequent precise control. In addition, by dynamically monitoring the proportion of oxygen concentration gradient risk areas and adjusting the compression rate of data on the blockchain, the efficiency and security of data storage are achieved while ensuring data accuracy.

[0053] 2. After marking the risk area of ​​oxygen concentration gradient, the fluctuation rate of the redox potential in the area is analyzed by the reverse differential entropy algorithm to determine whether the local instability is abnormal. This analysis can sensitively capture the regional changes caused by oxygen fluctuations, help identify the instability caused by oxygen concentration fluctuations, and further ensure the stability and consistency of the fermentation process. When an abnormal degree of instability is detected, the system generates an oxygen distribution adjustment instruction and transmits the instruction to the environmental adjustment device in the fermentation container. The adjustment device dynamically adjusts the oxygen distribution parameters, including oxygen flow rate and flow direction, according to the instruction, so as to quickly respond to and correct the fermentation environment fluctuations caused by uneven oxygen distribution; this process constructs real-time closed-loop management of oxygen concentration during the fermentation process through distributed monitoring, dynamic analysis and automatic adjustment, which significantly improves the uniformity of the lactic acid bacteria fermentation process, product quality stability, and ensures the safety of the fermentation process and data integrity. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a schematic diagram of a real-time monitoring method for a lactic acid bacteria fermentation process based on blockchain according to the present invention;

[0055] Figure 2This is a schematic diagram of the structure of a real-time monitoring system for lactic acid bacteria fermentation process based on blockchain in the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Example 1

[0058] Figure 1 The present invention provides a real-time monitoring method for a lactic acid bacteria fermentation process based on blockchain, which comprises the following steps:

[0059] Collect oxygen concentration data in the fermentation container in real time and transmit the oxygen concentration data to the blockchain network.

[0060] The dynamic changes of the acid reaction hysteresis rate were analyzed by Kalman filtering, and the maximum a posteriori probability estimation was used to identify the significant acid reaction delay areas in the fermentation vessel.

[0061] The monitoring parameters of the fermentation process were comprehensively analyzed through tensor decomposition, and the influence of each monitoring parameter at different positions in the fermentation vessel was evaluated to identify areas with obvious differences in oxygen concentration.

[0062] The area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container is marked as the oxygen concentration gradient risk area.

[0063] Based on the proportion of oxygen concentration gradient risk areas, the compression rate of data on the blockchain is dynamically adjusted.

[0064] The redox potential fluctuation rate in the oxygen concentration gradient risk area is analyzed by the reverse differential entropy algorithm to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal.

[0065] When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated.

[0066] Collect oxygen concentration data in the fermentation container in real time and transmit the oxygen concentration data to the blockchain network, specifically:

[0067] Arrange multiple oxygen concentration sensors in the fermentation container to collect oxygen concentration data at each location in real time:

[0068] According to the size and shape of the fermentation container, the container is divided into multiple areas. To avoid blind areas, the sensor layout points are designed according to the three-dimensional spatial coordinates to ensure coverage of the surface, middle layer and bottom of the fermentation container. After the layout points are selected, the locations are clearly recorded by marking to facilitate subsequent data analysis and sensor maintenance.

[0069] According to the size of the fermentation container, the rate of change of oxygen concentration and the frequency of collection, select the appropriate number of oxygen concentration sensors to ensure monitoring accuracy. The sensor type should be a high-precision electrochemical or optical oxygen sensor to obtain accurate oxygen concentration data.

[0070] The sampling frequency is set according to the oxygen consumption characteristics during the lactic acid bacteria fermentation process, usually once per second or more frequently, to capture subtle changes in oxygen concentration. The sampling frequency needs to be adjusted in a timely manner according to the fermentation process and monitoring accuracy requirements.

[0071] The collected oxygen concentration data is subjected to denoising and anomaly detection:

[0072] A denoising algorithm based on wavelet transform is used to perform multi-level denoising on the data to eliminate the impact of environmental noise and sensor errors on data accuracy. Specifically, wavelet transform is used to decompose the data into multiple frequency levels, filter out high-frequency noise signals, and retain only valid data in the lower frequency bands of the signal.

[0073] A reasonable fluctuation range of oxygen concentration is set, and when the data deviates significantly, it is marked by an anomaly detection algorithm. An anomaly detection algorithm based on distribution estimation is used to detect possible sensor reading distortion or data anomalies, filter out these abnormal data, and ensure that the data transmitted to the blockchain is accurate.

[0074] The data after denoising and anomaly detection is reorganized and stored in a structured form (such as a time series table) in the cache, waiting to be uploaded to the blockchain in batches in the next step. The integrity of each piece of data is verified to ensure that the data is not lost or duplicated.

[0075] The processed oxygen concentration data is transmitted to the blockchain network in batches, and a timestamp is generated and stored on the blockchain for each set of data:

[0076] According to the set time interval or number of data, the oxygen concentration data in the cache is packaged into data batches. Each batch of data contains several oxygen concentration records, each record contains a timestamp, sensor location and data value to ensure the effectiveness of data transmission.

[0077] Before each batch of data is transmitted, the connection status of the blockchain node is verified to ensure that the data transmission channel is unobstructed. If a node response timeout or network failure is found, the backup node is enabled to upload data to the chain.

[0078] To ensure the security of transmitted data, encryption processing based on asymmetric encryption algorithms (such as RSA) is used to protect the transmission and storage of oxygen concentration data. Each batch of data is automatically encrypted before uploading to ensure the security of the data on the chain.

[0079] The encrypted data batches are uploaded to the blockchain through the API. After the transmission is completed, the blockchain will feedback the confirmation information of the successful data upload and generate a unique hash value for subsequent tracking.

[0080] The storage node location and timestamp of each set of oxygen concentration data is recorded on the blockchain:

[0081] After each set of data is uploaded to the blockchain, the generated timestamp and storage node location will be recorded on the blockchain to facilitate complete recording and subsequent tracing of the data.

[0082] A hash identifier is generated for each set of oxygen concentration data to ensure the uniqueness and immutability of the data. After each set of data is stored, the blockchain will feedback the storage status and hash identifier to provide a reliable basis for subsequent verification and query.

[0083] The dynamic changes of the acid reaction hysteresis rate were analyzed by Kalman filtering, and the maximum a posteriori probability estimation was used to identify the significant acid reaction delay area in the fermentation container, specifically:

[0084] Using the oxygen concentration data and pH value data stored on the blockchain, the acid reaction hysteresis rate at each location in the fermentation container is calculated:

[0085] The stored oxygen concentration data and pH value data are extracted from the blockchain. According to the different positions of the fermentation container, the oxygen concentration data and pH value data are associated with the corresponding time to calculate the acid reaction hysteresis rate at each position. The acid reaction hysteresis rate reflects the time difference between the change in oxygen concentration and the response of pH value.

[0086] When there is a certain time lag between the change in oxygen concentration and the response of pH value, the time difference can be expressed by the acid reaction hysteresis rate to measure the degree of lag between the change in oxygen concentration and the change in pH value.

[0087] The hysteresis rate of the acid reaction at a certain position in the fermentation vessel is calculated as follows: ;in, It indicates the hysteresis rate of acid reaction at that position; is a time-lagged variable, indicating the lag time difference for finding the maximum value; Indicates time; is the time series of oxygen concentration; is the pH time series; It means to find the lag time that makes the expression reach the maximum value. The value of .

[0088] For different positions in the fermentation container, the respective acid reaction hysteresis rates are calculated according to the above method to form an acid reaction hysteresis rate data set.

[0089] Construct a Kalman filter model for the dynamic change of the acid reaction hysteresis rate, and set the initial state, observation equation and process noise parameters:

[0090] A Kalman filter model of acid reaction hysteresis rate was constructed based on the acid reaction hysteresis rate data set to smooth the hysteresis rate data and eliminate the influence of random fluctuations.

[0091] The construction of the Kalman filter model includes the following:

[0092] The state variable of the acid reaction hysteresis rate is set to , indicating that at time Estimated value of the hysteresis rate of the acid reaction at .

[0093] Assuming that the change of the acid reaction hysteresis rate is a random process, the state equation can be expressed as: ;in, is the process noise, which is used to characterize the random changes of the hysteresis rate.

[0094] The observed acid reaction hysteresis rate data is used to establish the observation equation, which is expressed as: ;in, represents the observed hysteresis value, Observation noise is used to characterize the random deviation caused by the external environment or sensor error in the measurement process.

[0095] Through this observation equation, the actual observation data is combined with the Kalman filter model to smooth the noise.

[0096] Set the initial state and initial state covariance of the Kalman filter.

[0097] The acid reaction hysteresis rate data at each position is processed by Kalman filtering to obtain a smooth hysteresis rate dynamic change curve:

[0098] The above-mentioned Kalman filter model is used to smooth the acid reaction hysteresis rate data set to obtain a smoothed hysteresis rate dynamic change curve. The specific implementation includes the following steps:

[0099] According to the state equation of Kalman filter, the acid reaction hysteresis rate at each moment is predicted.

[0100] Calculate the predicted state, the expression is: ;in, is the predicted state, indicating that at time The lag rate prediction value, that is, based on time The estimated value of the state at time ; Indicates at time The estimated value of the hysteresis rate by the Kalman filter at this time represents the current hysteresis rate state after filtering.

[0101] Calculate the forecast covariance, the expression is: ;in, Indicates time The forecast covariance at time quantifies the uncertainty of the lagged rate forecast at the next time point; Indicates time covariance of the time-lagged rate estimates; is the variance of the process noise, describing the random fluctuation of the lag rate during the state update process.

[0102] Combined with the observation equation, the predicted value is updated using the observed hysteresis value to obtain the corrected state estimate and the corrected covariance.

[0103] Calculate the Kalman gain: ;in, is the Kalman gain, which is used to balance the weights of predicted values ​​and observed values; It is the variance of the observation noise, which indicates the magnitude of the error caused by the equipment or external environment during the measurement process.

[0104] The corrected state estimate is: ;in, is the corrected state estimate, indicating that The corrected lag rate estimate at .

[0105] The corrected covariance is: ;in, is the corrected covariance, which represents the uncertainty of the corrected state.

[0106] The updated corrected state estimate is the smoothed hysteresis rate estimate.

[0107] The above prediction and update steps are performed sequentially to obtain a smooth curve of the acid reaction hysteresis rate at each position.

[0108] Based on the smoothed hysteresis rate curve, the probability value of acid reaction delay at each position was calculated using the maximum a posteriori probability estimation method:

[0109] The probability value of the delayed acid reaction is calculated as follows: ;in, is the probability value of delayed acid reaction; Indicates that the acid reaction hysteresis rate exceeds the threshold The probability of occurrence of the smoothed lag rate estimate under the condition; Indicates that the acid reaction hysteresis rate exceeds the threshold The prior probability of , that is, the probability of a significant delay in acidic reaction; represents the marginal probability of the smoothed lagged rate estimate.

[0110] By statistically analyzing the distribution of hysteresis rate data, the probability values ​​of acid reaction delay at different locations are calculated, and the probability value of acid reaction delay at each location is used as the result of maximum a posteriori probability estimation to form a preliminary mark of the area with significant acid reaction delay.

[0111] Among them, the threshold The setting of hysteresis rate is based on historical data analysis and experimental verification. The statistical characteristics of hysteresis rate are usually obtained from the long-term monitoring of acid reaction hysteresis rate data. The critical point of significant change of hysteresis rate is determined by analyzing the distribution of hysteresis rate of different fermentation batches. In addition, combined with the specific requirements of the fermentation process, the hysteresis rate is set within a reasonable range so that the threshold Reflects the delayed relationship between oxygen concentration and pH response during fermentation, ensuring sensitive identification of abnormal hysteresis conditions.

[0112] According to the probability value of acid reaction delay at each position, the significant area of ​​acid reaction delay in the fermentation container is determined:

[0113] Setting the probability threshold for delay saliency judgment , the probability value of acid reaction delay and the probability threshold For comparison:

[0114] When the probability value of acid reaction delay is greater than the probability threshold When the temperature is 0.1°, the position is marked as an area with significant acid reaction delay.

[0115] The spatial location and time stamp of the area with significant delay are stored on the blockchain to ensure data consistency in subsequent analysis.

[0116] Among them, the probability threshold The setting is based on the risk analysis of the areas with significant delays in the acid reaction during the fermentation process. By statistically analyzing the delay probabilities at different locations, the probability distribution characteristics representing abnormal delays are identified, usually taking the 95% or 99% quantile that exceeds the normal fluctuation range. In addition, the probability threshold is also combined with the process requirements of the fermentation process to Reflect delayed significance to ensure timely marking and identification of areas that may affect fermentation quality.

[0117] The monitoring parameters of the fermentation process were comprehensively analyzed through tensor decomposition, and the influence of each monitoring parameter at different locations in the fermentation vessel was evaluated to identify areas with obvious differences in oxygen concentration, specifically:

[0118] Obtain monitoring parameters at different locations in the fermentation container from the blockchain:

[0119] The monitoring parameter data of various locations in the fermentation container are extracted from the blockchain network. The monitoring parameters include but are not limited to oxygen concentration, temperature, pH value and redox potential.

[0120] The acquired monitoring parameter data are matched according to the position and time in the fermentation container to form a complete data set.

[0121] The extracted monitoring parameters are organized into multidimensional tensors according to spatial position and time series to form a high-order data structure to represent the multiple correlations of the monitoring parameters:

[0122] Construct a fourth-order tensor where the four dimensions correspond to the three axes of spatial position and time.

[0123] For each position and time, the corresponding monitoring parameter value is taken as the element of the tensor, expressed as: ;in, Representing a tensor In Location and time The elements at Indicates the oxygen concentration value, Indicates the temperature value, Indicates pH value, Represents the redox potential value, is the name or symbol of the entire tensor (representing a multidimensional data structure containing the numerical set of all monitoring parameters).

[0124] in, represents the first spatial dimension within the fermentation vessel (e.g., position on the horizontal axis);

[0125] represents the second spatial dimension within the fermentation vessel (e.g., position on the longitudinal axis);

[0126] Representing the third spatial dimension within the fermentation vessel (e.g., position on the depth axis);

[0127] Represents the time dimension, which is used to record data at different time points during the fermentation process;

[0128] Represented in spatial dimensions Position index on (e.g., different horizontal positions);

[0129] Represented in spatial dimensions Position index on (e.g., different vertical positions);

[0130] Represented in spatial dimensions position index on (e.g., different depth positions);

[0131] Representing the time dimension The index is used to mark the specific time point of the data in the fermentation process.

[0132] Through the above method, all monitoring parameter data are organized into a complete four-dimensional tensor for subsequent tensor decomposition analysis.

[0133] High-order singular value decomposition is used to decompose the multidimensional tensor into a set of factor matrices and core tensors, and the influence weights of the main characteristic modes and monitoring parameters at different positions are extracted:

[0134] The tensor Perform pattern expansion in each dimension to obtain the corresponding matrix expansion form.

[0135] Mode-1 expansion (spatial dimension ): Move the tensor along Expand the axis to get the matrix .

[0136] Mode-2 expansion (spatial dimension ): Along Expand the axis to get the matrix .

[0137] Mode-3 expansion (spatial dimension ): Along Expand the axis to get the matrix .

[0138] Mode-4 expansion (time dimension ): Along Expand the axis to get the matrix .

[0139] For each pattern, expand the matrix (where ) performs singular value decomposition (SVD).

[0140] The singular value decomposition formula is: ;in, Represents the tensor Along the The matrix obtained by expanding the dimensions is Representation Mode The factor matrix of Representation Mode The singular value matrix of Representation Mode The right singular vector matrix of , is the transpose symbol.

[0141] The tensor Projected onto the singular vector space of each mode, we get the kernel tensor .

[0142] The core tensor calculation formula is: ;in, Represents the tensor in The product of the matrix in the formula and wait), For Mode The transpose of the factor matrix .

[0143] The main eigenvectors of each mode are extracted from the factor matrix, representing the main change direction of the data under this mode.

[0144] The elements in the kernel tensor are used in combination with the singular value matrix to calculate the influence weights of the monitoring parameters at different locations and times.

[0145] According to the factor matrix obtained by tensor decomposition and the influence weights of the monitoring parameters at different positions, the influence of each monitoring parameter at different positions in the fermentation container is evaluated, and the areas with obvious differences in oxygen concentration are identified:

[0146] For each spatial dimension, the influence of each position is calculated using the corresponding factor matrix.

[0147] Calculate the position influence, the expression is: ;in, Representation Mode The influence of location.

[0148] Using factor moments Calculate the temporal trends of monitored parameters.

[0149] The kernel tensor and singular value matrix are combined to calculate the weight of each monitoring parameter at different locations and times.

[0150] The weight formula of monitoring parameters is: ;in, is the weight matrix of monitoring parameters, represents the Kronecker product.

[0151] Set the threshold corresponding to the influence of oxygen concentration .

[0152] Identify areas with significant differences in oxygen concentration using the identification formula: ;in, is a collection of areas with obvious differences in oxygen concentration. Indicates at location The influence of oxygen concentration.

[0153] The influence of oxygen concentration at each location It is obtained by weighted combination of the influence of each position obtained by pattern expansion, reflecting the influence of oxygen concentration at different positions. The weight of the weighted combination comes from .

[0154] Among them, the threshold The setting is based on the analysis of the distribution characteristics of the oxygen concentration influence data, usually taking a high quantile (such as 95% or 99% quantile) that exceeds the normal range of variation. Combined with the specific needs of the fermentation process, the threshold Ensure that areas with significant differences in oxygen concentration can be accurately identified, allowing monitoring to focus on key areas that may affect fermentation quality, thereby improving monitoring sensitivity and accuracy.

[0155] The locations of the identified areas with significant differences in oxygen concentration and the corresponding impact data are recorded on the blockchain to ensure the traceability and non-tamperability of the data.

[0156] The area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container is marked as the oxygen concentration gradient risk area:

[0157] By taking the union of the areas with significant acid reaction delay and the areas with significant oxygen concentration difference as the oxygen concentration gradient risk areas, we can comprehensively consider the two key indicators to ensure the efficient identification of potential risk areas. The areas with significant acid reaction delay reflect the delayed response of oxygen to the acid reaction, indicating the possibility of adverse effects of oxygen supply on the fermentation process; the areas with significant oxygen concentration difference directly indicate the uneven distribution of oxygen in the container. Marking these two areas as oxygen concentration gradient risk areas by union helps to accurately capture the impact of abnormal oxygen supply on the consistency of the fermentation process, enhance the sensitivity and coverage of overall monitoring, and facilitate quick measures to improve the fermentation environment.

[0158] Based on the proportion of oxygen concentration gradient risk areas, the compression rate of data on the blockchain is dynamically adjusted, specifically:

[0159] In all monitoring areas of the fermentation container, the ratio of the volume of the oxygen concentration gradient risk area to the volume of the total monitoring area of ​​the fermentation container is marked as the risk area ratio.

[0160] The compression rate of data on the blockchain refers to the ratio of the volume of compressed data to the volume of the original data during the blockchain storage process. Specifically, it indicates the degree to which the amount of data stored in the blockchain is reduced compared to the original data volume, in order to optimize storage space and improve data processing efficiency.

[0161] The larger the risk area ratio, the lower the compression rate of the data on the blockchain is adjusted. This is because as the risk area ratio increases, the accuracy requirements of the data increase, and a lower compression rate can retain more detailed information for more comprehensive monitoring and analysis of oxygen distribution.

[0162] A specific adjustment method is given below:

[0163] Set the initial compression rate of the data on the blockchain to , then the dynamic adjustment formula of compression ratio can be: ;in, is the compression ratio after dynamic adjustment; is the adjustment coefficient, It is usually set to be greater than 0 to control the impact of the risk area ratio on the compression rate; is the risk area ratio.

[0164] When the risk area ratio is zero, the compression ratio is maintained at , that is, the data compression degree is high.

[0165] Determines the impact of the risk area ratio on the compression rate. The larger the value, the more significant the change in risk area ratio is on the compression ratio.

[0166] When the volume of the oxygen concentration gradient risk area increases relative to the total monitoring area volume, it means that the potential risk increases, so the compression rate needs to be reduced to retain more data details.

[0167] When the risk area increases, the compression rate is reduced to ensure data integrity and detail retention, which is convenient for subsequent analysis and tracing; when the risk decreases, the compression rate is increased to reduce storage space usage and improve the overall data processing efficiency and resource utilization of the blockchain.

[0168] The reverse differential entropy algorithm is used to analyze the redox potential fluctuation rate in the oxygen concentration gradient risk area to determine whether the local instability caused by oxygen concentration fluctuations in the oxygen concentration gradient risk area is abnormal. Specifically:

[0169] Obtain the redox potential data for each location within the oxygen concentration gradient risk area and organize it in time series:

[0170] Extract redox potential data from the identified oxygen concentration gradient risk areas, ensuring that the data at all time points are complete and arranged in chronological order. The data is formed into a redox potential time series by time for subsequent fluctuation rate calculation and analysis.

[0171] The redox potential data is collected by the redox potential sensor installed in the fermentation vessel, which can monitor the oxidation or reduction state of the fermentation medium in real time, generate potential signals and convert them into numerical data. The redox potential reflects the changes in chemical properties such as oxygen supply and microbial metabolic activity in the fermentation environment.

[0172] The extracted data were processed by time difference to calculate the redox potential fluctuation rate at each location at different time points:

[0173] The redox potential values ​​at adjacent time points were obtained respectively, and the redox potential fluctuation rate was calculated. The expression is: ;in, Indicates at time The redox potential fluctuation rate is represents the first A point in time, Indicated in The redox potential value measured at each time point, The index of the time point. represents the difference in redox potential between adjacent moments, Indicates the time interval between adjacent moments.

[0174] The redox potential fluctuation rate is input into the reverse differential entropy model, and the reverse differential entropy value of each position is calculated to obtain a quantitative index of the local instability:

[0175] Inverse differential entropy is used to quantify the complexity of the fluctuation rate, helping to identify local instabilities caused by oxygen concentration fluctuations.

[0176] The calculation formula of reverse differential entropy is: ;in, Reverse differential entropy value representing the risk area of ​​oxygen concentration gradient; is the probability of occurrence of redox potential fluctuation rate, which is estimated based on the statistical distribution of the fluctuation rate sequence; Represents the total number of time points in the redox potential time series.

[0177] The larger the reverse differential entropy value of the oxygen concentration gradient risk area is, the more complex the change of the fluctuation rate sequence is and the higher the degree of local instability is. The entropy value HHH of this area is used as a quantitative local instability indicator.

[0178] According to the comparison results of the reverse differential entropy value at each position and the threshold, it is judged whether the local instability of the oxygen concentration gradient risk area is abnormal:

[0179] Set the abnormal threshold corresponding to the reverse differential entropy, and compare the reverse differential entropy value with its corresponding abnormal threshold to determine whether the instability reaches the abnormal standard. The abnormal threshold corresponding to the reverse differential entropy is determined based on historical data and experiments to distinguish between normal fluctuations and abnormal instability.

[0180] When the reverse differential entropy value is greater than its corresponding abnormal threshold, the local instability of the oxygen concentration gradient risk area is judged to be abnormal; this indicates that the redox potential in this area fluctuates greatly and changes complexly, and may be more significantly affected by oxygen concentration fluctuations. There is a risk of abnormal or unstable conditions, and further monitoring or adjustment measures are required.

[0181] When the reverse differential entropy value is less than or equal to its corresponding abnormal threshold, the local instability level of the oxygen concentration gradient risk area is judged to be normal, indicating that the redox potential fluctuates slightly and changes relatively smoothly, which is in line with the expectations of a normal fermentation process and is not affected by obvious fluctuations in oxygen concentration, and no additional intervention is required.

[0182] When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated, specifically:

[0183] Determine the oxygen concentration gradient characteristics of the oxygen concentration gradient risk area. The oxygen concentration gradient characteristics include concentration distribution and change trend:

[0184] Concentration distribution: Obtain the specific spatial distribution of oxygen concentration in the risk area and identify the differences in oxygen concentration at different locations.

[0185] Change trend: Analyze the change trend of oxygen concentration in the area, clarify the increase or decrease pattern of concentration over time, and determine the flow direction and diffusion rate of oxygen in the area.

[0186] Generate specific oxygen concentration adjustment parameters based on the characteristics of oxygen concentration gradient, including oxygen flow rate and oxygen flow direction, and transmit oxygen distribution adjustment instructions to the environmental adjustment device in the fermentation container:

[0187] Oxygen flow rate: According to the concentration distribution and change trend, set the oxygen flow rate in the fermentation container to ensure that the oxygen distribution can be balanced within a reasonable time.

[0188] Oxygen flow direction: Determine the direction of oxygen flow based on the distribution characteristics of oxygen in the risk area so that oxygen can flow into the low-oxygen area more efficiently and reduce the impact of uneven distribution.

[0189] After receiving the oxygen distribution adjustment instruction, the environmental adjustment device adjusts the oxygen distribution in the fermentation container according to the oxygen concentration adjustment parameters:

[0190] Flow rate control: The environmental conditioning device will adjust the output flow rate of the oxygen supply system according to the flow rate parameters in the instruction to ensure that oxygen can quickly cover the risk area and alleviate the oxygen concentration gradient.

[0191] Flow direction control: According to the flow direction parameters of the instruction, the distribution angle and direction of the oxygen flow device are adjusted so that oxygen can flow to the low oxygen area more effectively.

[0192] Among them, the real-time oxygen concentration data in the fermentation container is transmitted and stored in the blockchain network, providing a data basis for all subsequent analysis steps. The decentralized and tamper-proof characteristics of blockchain technology ensure the integrity and reliability of the data, so that these data can be further analyzed and processed directly on the blockchain. Including using Kalman filtering to analyze the acid reaction lag rate, maximum a posteriori probability estimation to identify areas with significant acid reaction delays, evaluating the influence of monitoring parameters at different locations through tensor decomposition, identifying areas with obvious differences in oxygen concentration, and applying the reverse differential entropy algorithm to determine the degree of instability, etc. can all be completed based on the monitoring data stored on the blockchain, ensuring the security and continuity of data processing, while improving the accuracy and response efficiency of real-time analysis, and facilitating the targeted generation of oxygen distribution adjustment instructions.

[0193] Example 2

[0194] The difference between Example 2 of the present invention and Example 1 is that this example introduces a real-time monitoring system for lactic acid bacteria fermentation process based on blockchain.

[0195] Figure 2 A structural schematic diagram of a real-time monitoring system for a lactic acid bacteria fermentation process based on blockchain is given in the present invention. A real-time monitoring system for a lactic acid bacteria fermentation process based on blockchain includes a data acquisition and uploading module, an acid reaction analysis module, a tensor decomposition and analysis module, a risk area marking module, a data compression and adjustment module, a fluctuation anomaly assessment module and an instruction generation and adjustment module.

[0196] Data collection and upload module: collects oxygen concentration data in the fermentation container in real time and transmits the oxygen concentration data to the blockchain network.

[0197] Acid reaction analysis module: The dynamic changes of the acid reaction hysteresis rate are analyzed through Kalman filtering, and the maximum a posteriori probability estimation is used to identify the areas with significant acid reaction delay in the fermentation container.

[0198] Tensor decomposition analysis module: Through tensor decomposition, the monitoring parameters of the fermentation process are comprehensively analyzed to evaluate the influence of each monitoring parameter at different positions in the fermentation container to identify areas with obvious differences in oxygen concentration.

[0199] Risk area marking module: the area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container is marked as the oxygen concentration gradient risk area.

[0200] Data compression adjustment module: Dynamically adjust the compression rate of data on the blockchain based on the proportion of oxygen concentration gradient risk areas.

[0201] Fluctuation anomaly assessment module: The redox potential fluctuation rate in the oxygen concentration gradient risk area is analyzed through the reverse differential entropy algorithm to determine whether the local instability caused by oxygen concentration fluctuations in the oxygen concentration gradient risk area is abnormal.

[0202] Instruction generation and adjustment module: When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated.

[0203] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0204] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0205] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0206] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0207] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0208] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0209] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0210] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0211] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0212] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time monitoring method for lactic acid bacteria fermentation process based on blockchain, characterized in that: The steps include: Collect oxygen concentration data in the fermentation container in real time and transmit the oxygen concentration data to the blockchain network; The dynamic changes of the acid reaction hysteresis rate were analyzed by Kalman filtering, and the significant acid reaction delay area in the fermentation vessel was identified using maximum a posteriori probability estimation. Comprehensively analyze the monitoring parameters of the fermentation process through tensor decomposition, evaluate the influence of each monitoring parameter at different locations in the fermentation vessel, and identify areas with obvious differences in oxygen concentration; The area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container is marked as the oxygen concentration gradient risk area; Dynamically adjust the compression rate of data on the blockchain based on the proportion of oxygen concentration gradient risk areas; The redox potential fluctuation rate in the oxygen concentration gradient risk area is analyzed by the reverse differential entropy algorithm to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal; When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated; The dynamic changes of the acid reaction hysteresis rate were analyzed by Kalman filtering, and the maximum a posteriori probability estimation was used to identify the significant acid reaction delay area in the fermentation container, specifically: Using the oxygen concentration data and pH value data stored on the blockchain, the acid reaction hysteresis rate at each location in the fermentation container is calculated; Construct a Kalman filter model for the dynamic change of the hysteresis rate of the acid reaction, set the initial state, observation equation and process noise parameters; The acid reaction hysteresis rate data at each position are processed by Kalman filtering to obtain a smooth hysteresis rate dynamic change curve; Based on the smoothed hysteresis rate curve, the probability value of delayed acidic reaction at each position was calculated using the maximum a posteriori probability estimation method; Determine the significant acid reaction delay area in the fermentation container according to the probability value of the acid reaction delay at each position; The monitoring parameters of the fermentation process were comprehensively analyzed through tensor decomposition, and the influence of each monitoring parameter at different locations in the fermentation vessel was evaluated to identify areas with obvious differences in oxygen concentration, specifically: Obtain monitoring parameters at different locations in the fermentation container from the blockchain; the monitoring parameters include oxygen concentration, temperature, pH value and redox potential; The extracted monitoring parameters are organized into multidimensional tensors according to spatial positions and time series to form a high-order data structure to represent the multiple correlations of the monitoring parameters; High-order singular value decomposition is used to decompose the multidimensional tensor into a set of factor matrices and core tensors, and the influence weights of the main characteristic patterns and monitoring parameters at different positions are extracted; According to the factor matrix obtained by tensor decomposition and the influence weights of the monitoring parameters at different positions, the influence of each monitoring parameter at different positions in the fermentation container is evaluated, and the areas with obvious differences in oxygen concentration are identified; The reverse differential entropy algorithm is used to analyze the redox potential fluctuation rate in the oxygen concentration gradient risk area to determine whether the local instability caused by oxygen concentration fluctuations in the oxygen concentration gradient risk area is abnormal. Specifically: Obtain redox potential data for each location within the oxygen concentration gradient risk area and organize it in time series; The extracted data were subjected to time difference processing to calculate the redox potential fluctuation rate at each location at different time points; The redox potential fluctuation rate is input into the reverse differential entropy model, and the reverse differential entropy value of each position is calculated to obtain a quantitative index of the local instability. According to the comparison result of the reverse differential entropy value at each position and the threshold, it is judged whether the local instability degree of the oxygen concentration gradient risk area is abnormal.

2. According to a blockchain-based real-time monitoring method for lactic acid bacteria fermentation process according to claim 1, it is characterized in that: Collect oxygen concentration data in the fermentation container in real time and transmit the oxygen concentration data to the blockchain network, specifically: Arrange multiple oxygen concentration sensors in the fermentation container to collect oxygen concentration data at each location in real time; De-noising and anomaly detection are performed on the collected oxygen concentration data; The processed oxygen concentration data is transmitted to the blockchain network in batches, and a timestamp is generated and stored for each set of data on the blockchain; The storage node location and timestamp of each set of oxygen concentration data are recorded on the blockchain.

3. According to claim 2, a real-time monitoring method for lactic acid bacteria fermentation process based on blockchain, characterized in that: Based on the proportion of oxygen concentration gradient risk areas, the compression rate of data on the blockchain is dynamically adjusted, specifically: In all monitoring areas of the fermentation container, the ratio of the volume of the oxygen concentration gradient risk area to the volume of the total monitoring area of ​​the fermentation container is marked as the risk area ratio; The larger the risk area ratio, the lower the compression rate of the data on the blockchain is adjusted.

4. A method for real-time monitoring of lactic acid bacteria fermentation process based on blockchain according to claim 3, characterized in that: When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated, specifically: Determine the oxygen concentration gradient characteristics of the oxygen concentration gradient risk area, including concentration distribution and change trend; Generate specific oxygen concentration adjustment parameters according to the characteristics of the oxygen concentration gradient, the oxygen concentration adjustment parameters including oxygen flow rate and oxygen flow direction, and transmit the oxygen distribution adjustment instruction to the environment adjustment device in the fermentation container; After receiving the oxygen distribution adjustment instruction, the environment adjustment device adjusts the oxygen distribution in the fermentation container according to the oxygen concentration adjustment parameter.

5. A real-time monitoring system for lactic acid bacteria fermentation process based on blockchain, used to implement a real-time monitoring method for lactic acid bacteria fermentation process based on blockchain according to any one of claims 1 to 4, characterized in that: It includes data acquisition and uploading module, acid reaction analysis module, tensor decomposition analysis module, risk area marking module, data compression adjustment module, fluctuation anomaly assessment module and instruction generation adjustment module; Data collection and upload module: collects oxygen concentration data in the fermentation container in real time and transmits the oxygen concentration data to the blockchain network; Acid reaction analysis module: Kalman filtering is used to analyze the dynamic changes of the acid reaction hysteresis rate, and the maximum a posteriori probability estimation is used to identify the area with significant acid reaction delay in the fermentation container; Tensor decomposition analysis module: Comprehensively analyze the monitoring parameters of the fermentation process through tensor decomposition, evaluate the influence of each monitoring parameter at different locations in the fermentation container, and identify areas with obvious differences in oxygen concentration; Risk area marking module: marking the area corresponding to the union of the area with significant acid reaction delay and the area with significant oxygen concentration difference in the fermentation container as the oxygen concentration gradient risk area; Data compression adjustment module: dynamically adjusts the compression rate of data on the blockchain based on the proportion of oxygen concentration gradient risk areas; Fluctuation anomaly assessment module: The reverse differential entropy algorithm is used to analyze the redox potential fluctuation rate in the oxygen concentration gradient risk area to determine whether the local instability caused by oxygen concentration fluctuation in the oxygen concentration gradient risk area is abnormal; Instruction generation and adjustment module: When the local instability of the oxygen concentration gradient risk area is abnormal, an oxygen distribution adjustment instruction is generated.

Images