A flour production monitoring and management system based on big data

Abstract

The present invention relates to the technical field of flour production, and in particular to a flour production monitoring and management system based on big data, including a data acquisition module, a data processing module, a data analysis module, an enzyme activity real-time monitoring module, an optimization control module, and a user interface and report generation module, wherein: the data acquisition module collects real-time production data of flour production; the data processing module preprocesses the collected real-time production data; the data analysis module analyzes the preprocessed real-time production data; the enzyme activity real-time monitoring module monitors the enzyme activity in flour in real time; the optimization control module adjusts and optimizes the flour production process in real time; and the user interface and report generation module provide a graphical user interface. The present invention provides a scientific basis for the optimization and adjustment of the production process, ensuring the stability of the production process and the improvement of product quality.

Description

Technical Field

[0001] The present invention relates to the technical field of flour production, and in particular to a flour production monitoring and management system based on big data. Background Art

[0002] With the rapid development of the food industry, flour, as a basic food raw material, is particularly important to monitor and optimize its production process. Flour production involves multiple links, including raw material handling, processing, packaging, etc. Parameter changes in each link will directly affect product quality and production efficiency. Therefore, real-time monitoring and adjustment of key parameters in the production process are crucial to ensuring the high quality of flour and the efficient and stable production process. With the rapid development of big data and intelligent manufacturing technology, how to use these technologies to improve the intelligence level of flour production is the focus of current industry attention.

[0003] The existing flour production monitoring system mainly relies on manual operation and traditional monitoring methods, and has problems such as untimely data collection, single monitoring parameters, insufficient data processing capabilities, and difficulty in achieving real-time adjustment and optimization. Specifically, the traditional system can usually only monitor a single or limited parameter, and lacks the ability to comprehensively monitor and analyze multi-dimensional data in a complex production environment. In addition, the existing technology lacks real-time monitoring and in-depth analysis of key biochemical parameters such as enzyme activity, and is unable to capture and process abnormal changes in the production process in a timely manner, resulting in difficulty in ensuring product quality and improving production efficiency. Faced with large amounts of data and complex production environments, the monitoring and processing capabilities of traditional systems are stretched to the limit and cannot meet the high standards of modern flour production.

[0004] The purpose of the present invention is to provide a flour production monitoring and management system based on big data, to solve the deficiencies in the prior art and to achieve real-time monitoring and intelligent optimization of the entire flour production process. Summary of the invention

[0005] Based on the above objectives, the present invention provides a flour production monitoring and management system based on big data.

[0006] A flour production monitoring and management system based on big data, comprising a data acquisition module, a data processing module, a data analysis module, an enzyme activity real-time monitoring module, an optimization control module, and a user interface and report generation module, wherein;

[0007] The data acquisition module collects real-time production data of each link in the flour production process, including temperature, humidity, pressure, flow rate, and equipment status;

[0008] The data processing module pre-processes the collected real-time production data, including data cleaning and format conversion;

[0009] The data analysis module analyzes the pre-processed real-time production data to mine the law of changes in the real-time production data;

[0010] The enzyme activity real-time monitoring module monitors the enzyme activity in flour in real time and identifies the law and trend of enzyme activity changes, specifically including:

[0011] Enzyme activity monitoring: real-time detection of enzyme activity data in flour;

[0012] Enzyme activity data processing: preliminary processing of the monitored enzyme activity data, including data cleaning and format conversion;

[0013] Model construction: Establish an enzyme activity prediction model based on historical enzyme activity data;

[0014] Enzyme activity analysis: Use the enzyme activity prediction model to analyze the enzyme activity data after preliminary processing to identify the rules and trends of enzyme activity changes;

[0015] The optimization control module adjusts and optimizes the flour production process in real time based on the analysis results of real-time production data and enzyme activity data, including production data monitoring, abnormal situation warning, equipment fault diagnosis and production data adjustment;

[0016] The user interface and report generation module provides a graphical user interface for operators to view production data, monitoring results and control schemes in real time, supports manual intervention, and generates production reports, including daily, weekly and monthly reports, for managers to make reference for decision-making.

[0017] Furthermore, the data acquisition module includes:

[0018] Temperature monitoring: Temperature sensors installed on the production line and equipment surface monitor the temperature changes in the flour production process in real time;

[0019] Humidity monitoring: Monitor the humidity level of the production environment through humidity sensors installed in the production environment;

[0020] Pressure monitoring: monitor the required pressure parameters during the production process through pressure sensors connected to flour processing equipment;

[0021] Flow monitoring: Monitor the flow rate of flour and related raw materials through flow sensors installed on material conveying pipelines;

[0022] Equipment condition monitoring: Monitor the equipment's operating status, workload, and fault conditions through condition sensors integrated into production equipment.

[0023] Furthermore, the data processing module includes:

[0024] Data cleaning: by removing noise data, filling missing values, and processing abnormal data;

[0025] Format conversion: Convert cleaned real-time production data into a unified format and structure.

[0026] Furthermore, the data analysis module includes:

[0027] Feature extraction: Extract features from pre-processed real-time production data through principal component analysis (PCA);

[0028] Pattern recognition: Use the support vector machine (SVM) classification algorithm to perform pattern recognition on production data;

[0029] Trend Analysis: Model and forecast time series of production data using Bayesian change point detection models.

[0030] Furthermore, the Bayesian change point detection model includes:

[0031] Initialization: Set the initial state r0 and initial parameter θ0, and set the prior probability of the change point P(r1=1)=λ and the probability of no change P(r1>1)=1-λ;

[0032] Multi-dimensional data collection and processing: Collect multi-dimensional production data x t =[x t,1 ,x t,2 ,...,x t,n ], and normalize each data dimension. The calculation formula is:

[0033]

[0034] Among them, μ i and σ i are the mean and standard deviation of the i-th dimension respectively;

[0035] Calculate the posterior probability of the change point: Calculate the posterior probability of the change point P(r t |x 1:t ), the calculation formula is:

[0036]

[0037] in,

[0038] Calculate the likelihood of the observed data: based on the current model parameters θ t-1 , calculate the likelihood P(x t |r t ,θ t-1 ), the calculation formula is:

[0039]

[0040] Among them, μ t-1,i and are the mean and variance of the i-th dimension respectively;

[0041] Update model parameters: Dynamically update model parameters θ based on the latest data t , the calculation formula is:

[0042]

[0043] Among them, w k is the weight;

[0044] Detect change point: Determine whether a change point exists. If P(r t =1|x 1:t )>threshold, it is considered that a change point occurs at time t, expressed as:

[0045]

[0046] Further, the enzyme activity monitoring includes:

[0047] Biosensor data collection: Biosensors are placed on the flour production line to collect enzyme activity data in flour samples in real time. Biosensors detect the activity of enzymes (such as α-amylase, protease, etc.) and generate electrical signals through the reaction between enzymes and substrates.

[0048] Electrical signal conversion and amplification: The electrical signal collected by the biosensor is converted into a digital signal through a converter and amplified through a signal amplifier;

[0049] Data transmission: Transmit the amplified digital signal through a wired or wireless network.

[0050] Furthermore, the enzyme activity prediction model adopts a fluctuation analysis model, and the fluctuation analysis model includes:

[0051] Wavelet decomposition: Perform multi-scale wavelet decomposition on the historical enzyme activity data to obtain detail coefficients D at different scales j and the approximation coefficient A J ;

[0052] The detail coefficient D j The calculation formula is:

[0053]

[0054] Among them, D j is the detail coefficient of the jth layer, ψ j is the wavelet function, a and b are the scale and translation parameters, and x(t) is the enzyme activity signal;

[0055] The approximation coefficient A J The calculation formula is:

[0056]

[0057] Among them, A J is the approximation coefficient of the Jth layer, is the scaling function;

[0058] Adaptive denoising: Introducing adaptive threshold λ j The wavelet coefficients are denoised and the calculation formula is:

[0059]

[0060] Among them, the threshold λ j Dynamically adjust according to the characteristics of enzyme activity data;

[0061]

[0062] Among them, σ j is the standard deviation of the detail coefficient, N is the number of data points;

[0063] Feature extraction: Extract key features from the detail coefficients and approximate coefficients after denoising, including energy features and frequency domain features. The calculation formula is:

[0064]

[0065] Among them, E j is the energy characteristic of the jth layer;

[0066] Reconstruct signal: Reconstruct the processed wavelet coefficients into time domain signals through inverse wavelet transform. The calculation formula is:

[0067]

[0068] Among them, C ψ is the normalization constant of the wavelet function;

[0069] Time series prediction: Use the Prophet model to predict the reconstructed signal, including:

[0070] Trend function: g(t) = (k + a(t) T δ)t+(m+a(t) T γ);

[0071] Where k is the slope, m is the intercept, a(t) is the indicator function, and δ and γ are the amplitudes of the change points;

[0072] Seasonal ingredients:

[0073] Among them, A i and B i is the Fourier coefficient, P is the period;

[0074] Model: y(t)=g(t)+s(t)+h(t)+∈ t ;

[0075] Where h(t) is the holiday effect, ∈ t is the error term.

[0076] Further, the enzyme activity analysis includes:

[0077] Data input: Input the preliminarily processed enzyme activity data into the enzyme activity prediction model;

[0078] Model application: The constructed enzyme activity prediction model is used to predict the input data. The prediction formula is:

[0079] y(t)=g(t)+s(t)+h(t)+∈ t ;

[0080] Among them, g(t) represents the trend component, s(t) represents the seasonal component, h(t) represents the holiday effect, ∈ t represents the error term;

[0081] Trend analysis: Identify the patterns and trends of enzyme activity changes by analyzing the predicted results y(t).

[0082] Furthermore, the trend analysis includes:

[0083] Trend component analysis: Analyze the trend component g(t) in the prediction result y(t), and identify the change direction and speed of the enzyme activity data by calculating the slope k and change point of the trend component;

[0084] Seasonal component analysis: Analyze the seasonal component s(t) in the prediction result y(t), and identify the periodicity and seasonal fluctuations of enzyme activity data by calculating the amplitude and period of the seasonal component;

[0085] Anomaly detection: The error term ∈ in the prediction result y(t) t Perform analysis to detect sudden changes and abnormal fluctuations in enzyme activity data by setting thresholds to identify outliers.

[0086] Furthermore, the optimization control module includes:

[0087] Production data monitoring: real-time monitoring of collected production data;

[0088] Abnormal situation warning: set the threshold value X of key parameter X according to production process and historical data threshold The key parameters X include temperature T, humidity H, pressure P, flow F, equipment status E and enzyme activity EA. When monitoring data X t Exceeding the set threshold X threshold When the alarm is triggered, it will automatically send out an early warning signal;

[0089] The threshold value X threshold The calculation formula is:

[0090] X threshold =μ X +3σ X ;

[0091] Among them, μ X and σ X are the mean and standard deviation of historical data respectively;

[0092] Equipment fault diagnosis: Use support vector machine (SVM) for fault diagnosis;

[0093] Production data adjustment: Based on the analysis results of real-time production data and enzyme activity data, calculate the adjustment plan of production parameters, and set the adjustment target to keep the key parameter X at the target value X. opt , the adjustment formula is:

[0094] X adjust =X current +α(X opt -X current );

[0095] Among them, α is the adjustment coefficient, X current is the current key parameter value, X opt is the target value.

[0096] Beneficial effects of the present invention:

[0097] The present invention realizes all-round real-time monitoring and optimization control of the flour production process by integrating data collection, data processing, data analysis, real-time monitoring of enzyme activity, optimization control, and user interface and report generation modules, ensures the accuracy and real-time nature of data collection, provides a comprehensive production data foundation, effectively extracts and identifies features and patterns in production data, can capture change points in the production process in real time, and improves the sensitivity and response speed of the system.

[0098] The present invention adopts a biosensor to detect the enzyme activity data in flour in real time, and uses a fluctuation analysis model to perform in-depth analysis and prediction of the enzyme activity data, identifies the law and trend of enzyme activity change, and ensures the efficiency and accuracy of data processing through adaptive denoising and feature extraction technology. The prediction model constructed in combination with historical enzyme activity data can accurately predict the enzyme activity change in the production process, provide a scientific basis for the optimization and adjustment of the production process, and ensure the stability of the production process and the improvement of product quality.

[0099] The present invention ensures the high efficiency and stability of the production process by real-time monitoring of production data and enzyme activity data, performing abnormal situation warning, equipment fault diagnosis and production data adjustment. The module uses a support vector machine to perform equipment fault diagnosis, combines the analysis results of real-time data and enzyme activity data, uses an optimization algorithm to calculate and automatically adjust production parameters, and improves production efficiency and product quality. The user interface and report generation module provide a friendly operation interface and detailed production reports, support the decision-making of operators and managers, and ensure the transparency and controllability of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0101] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;

[0102] Figure 2 Schematic diagram of a real-time enzyme activity monitoring module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0103] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0104] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0105] like Figure 1-Figure 2 As shown, a flour production monitoring and management system based on big data includes a data acquisition module, a data processing module, a data analysis module, an enzyme activity real-time monitoring module, an optimization control module, and a user interface and report generation module, wherein;

[0106] The data acquisition module collects real-time production data of each link in the flour production process, including temperature, humidity, pressure, flow, and equipment status. It uses high-sensitivity sensors to ensure the accuracy and real-time nature of data acquisition.

[0107] The data processing module pre-processes the collected real-time production data, including data cleaning and format conversion, to prepare for subsequent data analysis and ensure efficient data processing and accuracy;

[0108] The data analysis module analyzes the pre-processed real-time production data, explores the changing rules of real-time production data, and guides the optimization and adjustment of production data;

[0109] The real-time enzyme activity monitoring module monitors the enzyme activity in flour in real time and identifies the patterns and trends of enzyme activity changes, including:

[0110] Enzyme activity monitoring: Real-time detection of enzyme activity data in flour to ensure the accuracy of data collection;

[0111] Enzyme activity data processing: preliminary processing of the monitored enzyme activity data, including data cleaning and format conversion;

[0112] Model construction: Establish an enzyme activity prediction model based on historical enzyme activity data;

[0113] Enzyme activity analysis: Use the enzyme activity prediction model to analyze the enzyme activity data after preliminary processing to identify the rules and trends of enzyme activity changes;

[0114] The optimization control module makes real-time adjustments and optimizations to the flour production process based on the analysis results of real-time production data and enzyme activity data, including production data monitoring, abnormal situation warning, equipment fault diagnosis and production data adjustment, to ensure efficient and stable production process and excellent product quality.

[0115] The user interface and report generation module provides a graphical user interface for operators to view production data, monitoring results and control schemes in real time, supports manual intervention, and generates production reports, including daily, weekly and monthly reports, for managers to make reference for decision-making;

[0116] Through the above content, accurate optimization and control of the production process can be achieved to improve production efficiency and product quality. The system also provides a friendly user interface and detailed production reports to support the decision-making of operators and managers, ensuring efficient and stable production processes and excellent product quality.

[0117] The data acquisition module includes:

[0118] Temperature monitoring: Temperature sensors installed on the production line and equipment surface monitor the temperature changes in the flour production process in real time;

[0119] Humidity monitoring: Monitor the humidity level of the production environment through humidity sensors installed in the production environment;

[0120] Pressure monitoring: monitor the required pressure parameters during the production process through pressure sensors connected to flour processing equipment;

[0121] Flow monitoring: Monitor the flow rate of flour and related raw materials through flow sensors installed on material conveying pipelines;

[0122] Equipment status monitoring: Monitor the equipment's operating status, workload, and fault conditions through status sensors integrated into production equipment;

[0123] Through the above content, the sensor network transmits the collected data to the data processing module in real time through wired or wireless communication, ensuring that various parameters of the production process can be monitored accurately and in real time.

[0124] The data processing module includes:

[0125] Data cleaning: Ensure data accuracy and completeness by removing noise data, filling missing values, and processing abnormal data;

[0126] To remove noise data, the moving average filter method is used to smooth the sensor data. The calculation formula is:

[0127]

[0128] Among them, y(t) is the smoothed data, x(ti) is the original data, and N is the window size;

[0129] Filling missing values ​​uses linear interpolation to fill missing data. The calculation formula is:

[0130]

[0131] Where x(t) is the missing value, x(t1) and x(t2) are the known data points on both sides of the missing value, t is the position of the missing value, and t1 and t2 are the positions of the known data points;

[0132] To process abnormal data, the Z-score method is used to detect outliers and the median is corrected. The calculation formula is:

[0133]

[0134] Where Z is the standard score, x is the observed value, μ is the mean of the data, σ is the standard deviation of the data, and when |Z|>3, the data is regarded as an outlier and replaced with the median of the data;

[0135] Format conversion: convert the cleaned real-time production data into a unified format and structure to facilitate subsequent data analysis;

[0136] Through the above preprocessing steps, efficient processing and accuracy of real-time production data can be ensured, providing a basis for subsequent data analysis and optimized control.

[0137] Data analysis modules include:

[0138] Feature extraction: Extract features from pre-processed real-time production data through principal component analysis (PCA);

[0139] The calculation formula of principal component analysis (PCA) is:

[0140] Z = XW;

[0141] Among them, Z is the data matrix after dimension reduction, X is the original data matrix, and W is the eigenvector matrix;

[0142] Pattern recognition: Use the support vector machine (SVM) classification algorithm to perform pattern recognition on production data;

[0143] The support vector machine (SVM) classification algorithm includes:

[0144] Optimization goal:

[0145] Under the constraints: y i (w·x i +b)≥1-ξi ,ξ i ≥0;

[0146] Where w is the normal vector of the hyperplane, b is the bias, ξ i is the slack variable, C is the penalty parameter, y i is the label, x i is the feature vector;

[0147] Trend analysis: Modeling and forecasting time series of production data using Bayesian change point detection models;

[0148] Through the above content, the use of principal component analysis can effectively reduce data dimensions and extract key features, thereby reducing computational complexity and improving analysis efficiency. The support vector machine classification algorithm can accurately identify patterns and anomalies in production data and provide highly accurate classification results. Combined with the Bayesian change point detection model for trend analysis, it can capture change points in the production process in real time, detect anomalies in a timely manner, optimize the production process, improve the system's sensitivity and response speed, and ensure the efficiency and stability of the production process and the improvement of product quality.

[0149] The Bayesian change point detection model includes:

[0150] Initialization: Set the initial state r0 and initial parameter θ0, and set the prior probability of the change point P(r1=1)=λ and the probability of no change P(r1>1)=1-λ;

[0151] Multidimensional data collection and processing: Collect multidimensional production data xt = [x t,1 ,x t,2 ,...,x t,n ], and normalize each data dimension. The calculation formula is:

[0152]

[0153] Among them, μ i and σ i are the mean and standard deviation of the i-th dimension respectively;

[0154] Calculate the posterior probability of the change point: Calculate the posterior probability of the change point P(r t |x 1:t ), the calculation formula is:

[0155]

[0156] in,

[0157] Calculate the likelihood of the observed data: based on the current model parameters θ t-1 , calculate the likelihood P(xt |r t ,θ t-1 ), the calculation formula is:

[0158]

[0159] Among them, μ t-1,i and are the mean and variance of the i-th dimension respectively;

[0160] Update model parameters: Dynamically update model parameters θ based on the latest data t , the calculation formula is:

[0161]

[0162] Among them, w k is the weight, reflecting the importance of the data point, and w is set k =exp(-α(tk));

[0163] Detect change point: Determine whether a change point exists. If P(r t =1|x 1:t )>threshold, it is considered that a change point occurs at time t, expressed as:

[0164]

[0165] Through the above content, the present invention can monitor and analyze the change points in the production process in real time, improve the sensitivity and accuracy of the system, and ensure the high efficiency and stability of the production process and the improvement of product quality.

[0166] Enzyme activity monitoring includes:

[0167] Biosensor data collection: Biosensors are placed on the flour production line to collect enzyme activity data in flour samples in real time. Biosensors detect the activity of enzymes (such as α-amylase, protease, etc.) and generate electrical signals through the reaction between enzymes and substrates.

[0168] Electrical signal conversion and amplification: The electrical signal collected by the biosensor is converted into a digital signal through a converter and amplified through a signal amplifier to ensure the accuracy and stability of the signal;

[0169] Data transmission: Transmit the amplified digital signal through a wired or wireless network to ensure the real-time and integrity of data transmission;

[0170] Through the above steps, the enzyme activity real-time monitoring module can efficiently and accurately detect the enzyme activity data in flour in real time, providing key support for the optimization control of the production process.

[0171] The enzyme activity prediction model adopts a fluctuation analysis model, which includes:

[0172] Wavelet decomposition: Perform multi-scale wavelet decomposition on the historical enzyme activity data to obtain detail coefficients D at different scales j and the approximation coefficient A J ;

[0173] Detail coefficient D j The calculation formula is:

[0174]

[0175] Among them, D j is the detail coefficient of the jth layer, ψ j is the wavelet function, a and b are the scale and translation parameters, and x(t) is the enzyme activity signal;

[0176] Wavelet function ψ j Calculated by recursive filter, the calculation formula is:

[0177]

[0178] Among them, c k are the wavelet filter coefficients, is the scaling function;

[0179] Approximation coefficient A J The calculation formula is:

[0180]

[0181] Among them, A J is the approximation coefficient of the Jth layer, is the scaling function;

[0182] Scaling function Calculated by recursive filter, the calculation formula is:

[0183]

[0184] Among them, c k are the scaling filter coefficients;

[0185] Adaptive denoising: Introducing adaptive threshold λ j The wavelet coefficients are denoised and the calculation formula is:

[0186]

[0187] Among them, the threshold λ j Dynamically adjust according to the characteristics of enzyme activity data;

[0188]

[0189] Among them, σ j is the standard deviation of the detail coefficient, N is the number of data points;

[0190] Feature extraction: Extract key features from the detail coefficients and approximate coefficients after denoising, including energy features and frequency domain features. The calculation formula is:

[0191]

[0192] Among them, E j is the energy characteristic of the jth layer;

[0193] Reconstruct signal: Reconstruct the processed wavelet coefficients into time domain signals through inverse wavelet transform. The calculation formula is:

[0194]

[0195] Among them, C ψ is the normalization constant of the wavelet function;

[0196] Time series prediction: Use the Prophet model to predict the reconstructed signal, including:

[0197] Trend function: g(t) = (k + a(t) T δ)t+(m+a(t) T γ);

[0198] Where k is the slope, m is the intercept, a(t) is the indicator function, and δ and γ are the amplitudes of the change points;

[0199] Seasonal ingredients:

[0200] Among them, A i and B i is the Fourier coefficient, P is the period;

[0201] Model: y(t)=g(t)+s(t)+h(t)+∈ t ;

[0202] Where h(t) is the holiday effect, ∈ t is the error term;

[0203] Through the above steps, the enzyme activity prediction of the present invention can efficiently and accurately process and analyze enzyme activity data, ensuring the high efficiency and stability of the production process and the improvement of product quality.

[0204] Enzyme activity assays include:

[0205] Data input: Input the preliminarily processed enzyme activity data into the enzyme activity prediction model;

[0206] Model application: The constructed enzyme activity prediction model is used to predict the input data. The prediction formula is:

[0207] y(t)=g(t)+s(t)+h(t)+∈ t ;

[0208] Among them, g(t) represents the trend component, s(t) represents the seasonal component, h(t) represents the holiday effect, ∈ t represents the error term;

[0209] Trend analysis: Identify the patterns and trends of enzyme activity changes by analyzing the predicted results y(t);

[0210] Through the above steps, the enzyme activity prediction model can be effectively used to conduct in-depth analysis of the enzyme activity data after preliminary processing, identify the patterns and trends of enzyme activity changes, and provide a scientific basis for the optimization and adjustment of the production process.

[0211] Trend analysis includes:

[0212] Trend component analysis: Analyze the trend component g(t) in the prediction result y(t), and identify the change direction and speed of the enzyme activity data by calculating the slope k and change point of the trend component;

[0213] Seasonal component analysis: Analyze the seasonal component s(t) in the prediction result y(t), and identify the periodicity and seasonal fluctuations of enzyme activity data by calculating the amplitude and period of the seasonal component;

[0214] Anomaly detection: The error term ∈ in the prediction result y(t) t Perform analysis to detect sudden changes and abnormal fluctuations in enzyme activity data by setting thresholds to identify outliers.

[0215] The optimization control module includes:

[0216] Production data monitoring: real-time monitoring of collected production data;

[0217] Abnormal situation warning: set the threshold value X of key parameter X according to production process and historical data threshold The key parameters X include temperature T, humidity H, pressure P, flow F, equipment status E and enzyme activity EA. When monitoring data X t Exceeding the set threshold X threshold When the alarm is triggered, it will automatically send out an early warning signal;

[0218] Threshold X threshold The calculation formula is:

[0219] X threshold =μ X+3σ X ;

[0220] Among them, μ X and σ X are the mean and standard deviation of historical data respectively;

[0221] Equipment fault diagnosis: Use support vector machine (SVM) for fault diagnosis;

[0222] Production data adjustment: Based on the analysis results of real-time production data and enzyme activity data, calculate the adjustment plan of production parameters, and set the adjustment target to keep the key parameter X at the target value X. opt , the adjustment formula is:

[0223] X adjust =X current +α(X opt -X current );

[0224] Among them, α is the adjustment coefficient, X current is the current key parameter value, X opt is the target value;

[0225] Through the above steps, the flour production process can be adjusted and optimized in real time based on the analysis results of real-time production data and enzyme activity data, thereby achieving efficient and stable production process and improved product quality.

[0226] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

Claims

1. A flour production monitoring and management system based on big data, characterized in that: It includes a data acquisition module, a data processing module, a data analysis module, an enzyme activity real-time monitoring module, an optimization control module, and a user interface and report generation module, wherein; The data acquisition module collects real-time production data of each link in the flour production process, including temperature, humidity, pressure, flow rate, and equipment status; The data processing module pre-processes the collected real-time production data, including data cleaning and format conversion; The data analysis module analyzes the pre-processed real-time production data to mine the law of changes in the real-time production data; The enzyme activity real-time monitoring module monitors the enzyme activity in flour in real time and identifies the law and trend of enzyme activity changes, specifically including: Enzyme activity monitoring: real-time detection of enzyme activity data in flour; Enzyme activity data processing: preliminary processing of the monitored enzyme activity data, including data cleaning and format conversion; Model construction: Establish an enzyme activity prediction model based on historical enzyme activity data; Enzyme activity analysis: Use the enzyme activity prediction model to analyze the enzyme activity data after preliminary processing to identify the rules and trends of enzyme activity changes; The optimization control module adjusts and optimizes the flour production process in real time based on the analysis results of real-time production data and enzyme activity data, including production data monitoring, abnormal situation warning, equipment fault diagnosis and production data adjustment; The user interface and report generation module provides a graphical user interface for operators to view production data, monitoring results and control schemes in real time, supports manual intervention, and generates production reports, including daily, weekly and monthly reports, for managers to make reference for decision-making.

2. A flour production monitoring and management system based on big data according to claim 1, characterized in that: The data acquisition module comprises: Temperature monitoring: Temperature sensors installed on the production line and equipment surface monitor the temperature changes in the flour production process in real time; Humidity monitoring: Monitor the humidity level of the production environment through humidity sensors installed in the production environment; Pressure monitoring: monitor the required pressure parameters during the production process through pressure sensors connected to flour processing equipment; Flow monitoring: Monitor the flow rate of flour and related raw materials through flow sensors installed on material conveying pipelines; Equipment condition monitoring: Monitor the equipment's operating status, workload, and fault conditions through condition sensors integrated into production equipment.

3. A flour production monitoring and management system based on big data according to claim 1, characterized in that: The data processing module comprises: Data cleaning: by removing noise data, filling missing values, and handling abnormal data; Format conversion: Convert cleaned real-time production data into a unified format and structure.

4. A flour production monitoring and management system based on big data according to claim 1, characterized in that: The data analysis module includes: Feature extraction: Extract features from preprocessed real-time production data through principal component analysis; Pattern recognition: Use support vector machine classification algorithm to perform pattern recognition on production data; Trend Analysis: Model and forecast time series of production data using Bayesian change point detection models.

5. A flour production monitoring and management system based on big data according to claim 4, characterized in that: The Bayesian change point detection model includes: Initialization: Set the initial state r0 and initial parameter θ0, and set the prior probability of the change point P(r1=1)=λ and the probability of no change P(r1>1)=1-λ; Multi-dimensional data collection and processing: Collect multi-dimensional production data x t =[x t,1 ,x t,2 ,...,x t,n ], and normalize each data dimension. The calculation formula is: Among them, μ i and σ i are the mean and standard deviation of the i-th dimension respectively; Calculate the posterior probability of the change point: Calculate the posterior probability of the change point P(r t |x 1:t ), the calculation formula is: in, Calculate the likelihood of the observed data: based on the current model parameters θ t-1 , calculate the likelihood P(x t |r t ,θ t-1 ), the calculation formula is: Among them, μ t-1,i and are the mean and variance of the i-th dimension respectively; Update model parameters: Dynamically update model parameters θ based on the latest data t , the calculation formula is: Among them, w k is the weight; Detect change point: Determine whether a change point exists. If P(r t =1|x 1:t )>threshold, it is considered that a change point occurs at time t, expressed as:

6. The flour production monitoring and management system based on big data according to claim 1, characterized in that: The enzyme activity monitoring includes: Biosensor data collection: Biosensors are placed on the flour production line to collect enzyme activity data in flour samples in real time. Biosensors detect enzyme activity and generate electrical signals through the reaction between enzyme and substrate. Electrical signal conversion and amplification: The electrical signal collected by the biosensor is converted into a digital signal through a converter and amplified through a signal amplifier; Data transmission: Transmit the amplified digital signal through a wired or wireless network.

7. A flour production monitoring and management system based on big data according to claim 6, characterized in that: The enzyme activity prediction model adopts a fluctuation analysis model, and the fluctuation analysis model includes: Wavelet decomposition: Perform multi-scale wavelet decomposition on the historical enzyme activity data to obtain detail coefficients D at different scales j and the approximation coefficient A J ; The detail coefficient D j The calculation formula is: Among them, D j is the detail coefficient of the jth layer, ψ j is the wavelet function, a and b are the scale and translation parameters, and x(t) is the enzyme activity signal; The approximation coefficient A J The calculation formula is: Among them, A J is the approximation coefficient of the Jth layer, is the scaling function; Adaptive denoising: Introducing adaptive threshold λ j The wavelet coefficients are denoised and the calculation formula is: Among them, the threshold λ j Dynamically adjust according to the characteristics of enzyme activity data; Among them, σ j is the standard deviation of the detail coefficient, N is the number of data points; Feature extraction: Extract key features from the detail coefficients and approximate coefficients after denoising, including energy features and frequency domain features. The calculation formula is: Among them, E j is the energy characteristic of the jth layer; Reconstruct signal: Reconstruct the processed wavelet coefficients into time domain signals through inverse wavelet transform. The calculation formula is: Among them, C ψ is the normalization constant of the wavelet function; Time series prediction: Use the Prophet model to predict the reconstructed signal, including: Trend function: g(t) = (k + a(t) T δ)t+(m+a(t) T γ); Where k is the slope, m is the intercept, a(t) is the indicator function, and δ and γ are the amplitudes of the change points; Seasonal ingredients: Among them, A i and B i is the Fourier coefficient, P is the period; Function: y(t)=g(t)+s(t)+h(t)+∈ t ; Where h(t) is the holiday effect, ∈ t is the error term.

8. The flour production monitoring and management system based on big data according to claim 7 is characterized in that: The enzyme activity analysis includes: Data input: Input the preliminarily processed enzyme activity data into the enzyme activity prediction model; Model application: The constructed enzyme activity prediction model is used to predict the input data. The prediction formula is: y(t)=g(t)+s(t)+h(t)+∈ t ; Among them, g(t) represents the trend component, s(t) represents the seasonal component, h(t) represents the holiday effect, ∈ t represents the error term; Trend analysis: Identify the patterns and trends of enzyme activity changes by analyzing the predicted results y(t).

9. The flour production monitoring and management system based on big data according to claim 8, characterized in that: The trend analysis includes: Trend component analysis: Analyze the trend component g(t) in the prediction result y(t), and identify the change direction and speed of the enzyme activity data by calculating the slope k and change point of the trend component; Seasonal component analysis: Analyze the seasonal component s(t) in the prediction result y(t), and identify the periodicity and seasonal fluctuations of enzyme activity data by calculating the amplitude and period of the seasonal component; Anomaly detection: The error term ∈ in the prediction result y(t) t Perform analysis to detect sudden changes and abnormal fluctuations in enzyme activity data by setting thresholds to identify outliers.

10. The flour production monitoring and management system based on big data according to claim 1, characterized in that: The optimization control module comprises: Production data monitoring: real-time monitoring of collected production data; Abnormal situation warning: set the threshold value X of key parameter X according to production process and historical data threshold The key parameters X include temperature T, humidity H, pressure P, flow F, equipment status E and enzyme activity EA. When monitoring data X t Exceeding the set threshold X threshold When the alarm is triggered, it will automatically send out a warning signal; The threshold value X threshold The calculation formula is: X threshold =μ X +3s X ; Among them, μ X and σ X are the mean and standard deviation of historical data respectively; Equipment fault diagnosis: Use support vector machines for fault diagnosis; Production data adjustment: Based on the analysis results of real-time production data and enzyme activity data, calculate the adjustment plan of production parameters, and set the adjustment target to keep the key parameter X at the target value X. opt , the adjustment formula is: X adjust =X current +α(X opt -X current ); Among them, α is the adjustment coefficient, X current is the current key parameter value, X opt is the target value.

Images