Pet soft food production data management method and system

Abstract

The invention relates to the technical field of data management, in particular to a pet soft food production data management method and system.The method comprises the following steps of analyzing heat energy and rhythm synchronization based on damp and hot section operation data, judging batch data continuity, extracting fluctuation data to recognize quality inspection deviation, and matching packaging and discharging to analyze response consistency. And abnormal batches are extracted to locate an alarm concentration area. According to the method, by monitoring the heat energy flow and the conveying period, synchronization of the heat load and the putting rhythm is achieved, data consistency and collaboration are improved, data interruption and overlapping are avoided in combination with the batch time difference, fluctuation of indexes such as PH is periodically recognized, the quality inspection gathering time period is determined, packaging and discharging response offset is analyzed, and the equipment collaboration efficiency is quantified; energy records and alarm logs are crossly screened, abnormal batches are accurately positioned, a multi-parameter time sequence coupling analysis chain is constructed, cross-section quality response and traceability are enhanced, and centralized troubleshooting and early warning of the abnormal batches are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for managing pet soft food production data. Background Art

[0002] The field of data management technology encompasses the entire process of managing information collection, storage, organization, transmission, and maintenance. The core of this technology is to achieve the orderly, efficient, and secure flow of data throughout its entire lifecycle. This includes the acquisition of raw data, the organization of structured and unstructured data, the development of data standards, the establishment of data access and sharing strategies, and control mechanisms for data consistency and integrity. In practical applications, data management technology is widely used in a variety of fields, including industrial production, corporate operations, medical services, financial transactions, and internet services. Its characteristics are reflected in the full-chain management and technical integration from the source of data to the terminal that utilizes it, ensuring the traceability, accuracy, and timeliness of data during use and adapting to ever-changing business and management needs.

[0003] Among them, the pet soft food production data management method refers to a method for systematically managing data related to raw material ratio records, production batch tracking, processing parameter collection, process temperature and humidity monitoring, quality inspection item registration, warehousing and outbound time records, and relevant personnel operation logs during the production process of pet food, especially soft food. This method establishes a unified data collection standard, numbers and classifies various types of production data in different processes, uses the database to sort and associate information in time series, and uses the data interface to upload and synchronously update production equipment information, thereby forming a manageable, reviewable and analyzable data set. This management method also includes a binding mechanism for data record responsible persons, the setting of manual review processes for abnormal values, and the definition of different data access rights, etc., to ensure the standardization and controllability of the data collection and registration process.

[0004] While existing technologies cover multi-step information collection, from raw material recording to warehouse delivery registration, they lack a means to dynamically compare data status consistency across core production stages, leading to time misalignment and record disconnection between different stages. Regarding batch management, existing methods rely too heavily on timestamps or manual labeling, making it difficult to accurately track batch information in fast-moving production lines and prone to data overlap or omissions. In the application of quality inspection data, they focus solely on the instantaneous value of a single indicator, failing to extract stable segments from cyclical fluctuations, resulting in a one-sided understanding of quality control status. Regarding equipment response analysis, existing technologies fail to effectively connect equipment operating sequences with material behavior characteristics, making it difficult to identify and verify inter-equipment coordination issues. Furthermore, the correlation of alarm logs and energy data remains at the stage of recording isolated events, lacking a systematic anomaly location mechanism. This results in inefficient troubleshooting and the risk of batch quality issues spreading to subsequent stages. For example, if an energy anomaly occurs in a heating silo, without cross-stage data verification logic, quality deviations cannot be promptly detected in the cooling or packaging stages, impacting overall product consistency and safety. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a pet soft food production data management method and system.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: a method for managing pet soft food production data, comprising the following steps:

[0007] S1: Based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rail, and pre-feeding silo, the system detects the changing trends of heat flow values and conveying cycles within a continuous operation cycle, identifies the degree of synchronization between the material heating load and the feeding rhythm, and obtains the synchronization status value of the wet-heat section data;

[0008] S2: Based on the synchronization status value of the wet and hot segment data, the continuity and time difference of the pet food processing node batch number are compared to determine whether the data is interrupted or overlapped between segments, and the batch data continuity matching degree is generated;

[0009] S3: Call the batch number corresponding to the continuous matching degree of the batch data, extract the pH, nutrient composition and water content fluctuation data, analyze the floating trend of the data points within the detection period, identify the concentrated offset of the quality inspection period, and obtain the stable segment of the quality inspection value sequence;

[0010] S4: Based on the stable segment of the quality inspection value sequence, match the packaging segment identification and the discharge link data, analyze the interaction time and equipment response offset, filter the batches in the fluctuation period and compare the response consistency between the devices, and output the collaborative data response offset set.

[0011] As a further solution of the present invention, the synchronization status value of the wet and hot section data includes the heat flow intensity change amplitude, the delivery rhythm synchronization rate, and the heating rhythm consistency index; the batch data continuity matching degree includes the batch number connection rate, the timing interval error value, and the data paragraph matching degree; the stable section of the quality inspection value sequence includes the pH value stable range, the nutritional index fluctuation amplitude, and the water content stability coefficient; the collaborative data response offset set includes the packaging label response delay, the discharge trigger consistency deviation, and the equipment interaction time difference.

[0012] As a further solution of the present invention, the step of obtaining the synchronization status value of the wet and hot section data is specifically as follows:

[0013] S111: Based on data from the pet food wet heat production process, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos, the system identifies and compares the heat output values and conveyor rail operating rates within consecutive cycles to obtain a heat cycle matching trend value.

[0014] S112: Based on the heat energy cycle matching trend value, the material delivery cycle and synchronization interval data are called, rhythm difference comparison is performed on the corresponding sections, and the amplitude distribution range is identified to generate the heating rhythm offset amplitude interval;

[0015] S113: Based on the heating rhythm offset amplitude range, extract the start and end time and variation amplitude of the synchronization section of the heat energy trend and material delivery, identify the fluctuation section, and use the formula:

[0016]

[0017] Compare the offset density value with the thermal stability reference value to obtain the synchronization status value of the wet and hot section data;

[0018] Among them, S represents the synchronization status value of the wet and hot section data, Th i Represents the time length of the i-th section of the thermal energy flow, Tm i Represents the time length of the i-th section of the material delivery section, H i Represents the thermal energy density of the heat flow value in the i-th section, M i Represents the density of material delivery in the i-th section, A i Represents the dynamic load value of the conveyor rail in the i-th section, and n represents the total number of sections.

[0019] As a further solution of the present invention, the steps for obtaining the continuous matching degree of the batch data are specifically as follows:

[0020] S211: extracting the batch number of the processing node within the time period according to the synchronization status value of the wet and hot section data, analyzing the relationship between the batch number and the time sequence, identifying the offset between the batch number and the collection time, and obtaining the batch time sequence deviation structure;

[0021] S212: Based on the batch time series deviation structure, identify number gaps and time overlaps, analyze the cumulative quantity and classify them, evaluate the summary distribution of number-related and time-related anomalies, and obtain the number-time offset distribution characteristics;

[0022] S213: Call the number time offset distribution feature to perform a quantitative analysis of number differences, time fluctuations, and pairing integrity for each type of anomaly using the formula:

[0023]

[0024] Identify the batch number and time coupling consistency index, perform segmented comparison based on the preset threshold range, and obtain the continuous matching degree of batch data;

[0025] Among them, Ψ represents the continuous matching degree of batch data, Δb j Represents the batch number offset of the j-th batch pair, Δt j represents the squared time offset of the j-th batch pair, δ j Represents the identification value of the j-th batch number and time pair, Γ j represents the number of pairs in the jth batch, Θ j represents the number of time tags for the j-th batch pair, Ω j represents the total number of material record events for the jth batch pair, and m represents the total number of batch pairs in the analysis time period.

[0026] As a further solution of the present invention, the steps for obtaining the stable segment of the quality inspection value sequence are specifically as follows:

[0027] S311: Calling the batch number corresponding to the batch data continuous matching degree, extracting pH, nutrient composition and water content fluctuation data, monitoring the fluctuation trend of the data within the detection period, and obtaining a preliminary data sequence of the fluctuation trend;

[0028] S312: Analyze the fluctuation trend based on the preliminary data sequence of the fluctuation trend, identify the concentrated offset within the quality inspection period by analyzing the change of the fluctuation value, mark the offset interval, and obtain the offset abnormal fluctuation interval;

[0029] S313: Filter the data according to the abnormal fluctuation range of the offset to remove irrelevant data, using the formula:

[0030]

[0031] Calculate the stability value of the quality inspection value sequence and obtain the stable segment of the quality inspection value sequence;

[0032] Among them, Z represents the stability value of the quality inspection value sequence, P k represents the detection value of the kth data point, represents the mean of the detected values, K represents the total number of data points, and V represents the estimated standard deviation of the fluctuation values.

[0033] As a further solution of the present invention, the steps of obtaining the collaborative data response offset set are specifically as follows:

[0034] S411: Based on the stable segment of the quality inspection value sequence, matching the packaging segment identifier with the discharge link data, identifying the discharge batch number corresponding to the packaging sequence, extracting the discharge time point and the packaging start time point, calculating the time difference between the two time points, and obtaining the packaging and discharge interaction time;

[0035] S412: Extracting device response data during the interaction period based on the packaging and discharging interaction time, analyzing the time interval between the interaction time and the signal triggering time, identifying the response offset starting point and duration, screening batches that meet the conditions, and obtaining a device response offset batch set;

[0036] S413: For the device response offset batch set, identify the device data of the corresponding batch, extract the motor start and stop time, temperature control signal data, discharge port displacement value and feeder speed value, and use the formula:

[0037]

[0038] Calculate the data response difference value, identify the batch data associated with the devices with inconsistent responses in the same cycle based on the difference value, aggregate them into the structure set, and output the coordinated data response offset set;

[0039] Among them, R represents the data response difference value, u represents the total number of periodic signals collected by device A, Q represents the total number of periodic signals collected by device B, q represents the change in the forward signal of device A, x represents the change in the forward signal of device B, r represents the change in the reverse signal of device A, and y represents the change in the reverse signal of device B.

[0040] As a further embodiment of the present invention, the method further comprises step S5:

[0041] S5: Calling the collaborative data response offset set, extracting the energy records and alarm logs of abnormal batches in the feeding and mixing area, drying cabin, and cooling transmission line, and locating the data abnormality concentration area by cross-screening the record frequency and offset batches, and obtaining the concentrated detection area of abnormal batches in soft grain production;

[0042] The centralized detection area for abnormal batches of soft grain production includes abnormal energy consumption frequency bands, early warning log centralized nodes, and offset overlap areas between work sections.

[0043] As a further solution of the present invention, the steps for obtaining the centralized detection area for abnormal batches of soft grain production are specifically as follows:

[0044] S511: Call the collaborative data response offset set to extract the batch numbers and equipment process identifiers of abnormal batches in the feeding and mixing area, drying chamber, and cooling transmission line, capture the corresponding electricity, steam, and water flow data, count the record frequency of each batch in the area, and generate a section energy record frequency set;

[0045] S512: Extracting alarm logs corresponding to abnormal batches based on the energy record frequency set of the work section, classifying and counting them by regional equipment and alarm level, cross-screening the alarm batches of energy records, and obtaining an alarm data screening result set;

[0046] S513: Filter the result set based on the alarm data, count the frequency and distribution ratio of repeated abnormal batches in the equipment area, match the area identifier and the batch coordinate index, filter the section area with a frequency higher than the positioning benchmark, integrate the positioning label and coordinate mapping data, and obtain the concentrated detection area for abnormal batches in soft grain production.

[0047] The pet soft food production data management system is used to implement the above-mentioned pet soft food production data management method, and the system includes:

[0048] The thermal energy synchronization module is based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos. It extracts matching segments between the thermal energy flow and the conveying cycle within a continuous cycle, counts the number of matching points within the cycle, determines the direction of change in the number of matching points and the segment density, locates the positive and negative frequency conversion segments, and establishes the synchronization status value of the wet-heat segment data.

[0049] The batch continuity module extracts the corresponding processing batch number based on the synchronization status value of the wet and hot section data, compares the time difference and number spacing of adjacent batches, identifies the gap and overlap segments, and establishes the batch data continuity matching degree;

[0050] The quality inspection stability module extracts the pH value, nutrient composition, and water content data of the batch based on the continuous matching degree of the batch data, analyzes the trend of the range and mean difference in the continuous period, screens the concentrated section of the stable value, and establishes the stable section of the quality inspection value sequence;

[0051] The response offset module extracts the associated packaging segment identification time and discharge signal time based on the stable segment of the quality inspection value sequence, calculates and sorts the distribution of equipment response time differences, locates the numbering positions of response inversion and interval mutation, tracks the corresponding equipment combination and matching offset frequency, and establishes a collaborative data response offset set;

[0052] Based on the collaborative data response offset set, the anomaly positioning module extracts the energy logs and alarm logs of the associated batches in the feeding and mixing area, drying cabin, and cooling transmission line, filters the intersection of frequency and alarm period, and establishes a centralized detection area for abnormal batches of soft grain production.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are:

[0054] In the present invention, by continuously monitoring the heat flow value and conveying cycle of the core link of the wet heat section, real-time synchronous identification between the heat load and the material delivery rhythm is achieved, the time consistency and stage coordination of the section operation status data are improved, and further combined with the time difference analysis of the batch number, the data is effectively avoided from being interrupted and overlapped during the process connection, so that the production flow information has a clear time chain. The introduction of the periodic floating identification of the test data in terms of pH, nutrient content and moisture helps to clarify the fluctuation concentration period of the quality inspection data and strengthen the timeliness and accuracy of the quality control stability analysis. The correlation analysis of the response offset between the discharge and packaging interaction links makes the collaborative efficiency of the equipment operation quantifiable and easy to accurately screen the potential response lag in the fluctuating batches. Combined with the cross-screening method of energy records and abnormal alarm logs, the ability to accurately locate the relationship between abnormal batches and energy consumption changes is improved, so that abnormal behavior can be identified and defined in advance at the key nodes of the process. Through the multi-parameter, multi-node time series coupling analysis method, a logical chain that can be penetrated and traced is formed between various types of data, the response agility and traceability depth of cross-section quality management are enhanced, and the systematic monitoring and centralized investigation of abnormal batches in the production chain are supported. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0056] Figure 2 This is a flow chart for obtaining the synchronization status value of the wet and hot section data in the present invention;

[0057] Figure 3 This is a flow chart for obtaining the continuous matching degree of batch data in the present invention;

[0058] Figure 4 This is a flow chart for obtaining the stable segment of the quality inspection value sequence in the present invention;

[0059] Figure 5 This is a flowchart for obtaining a collaborative data response offset set in the present invention;

[0060] Figure 6 This is a flow chart for obtaining the centralized detection area for abnormal batches of soft grain production in the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0063] Example 1

[0064] See also Figure 1 The present invention provides a technical solution: a method for managing pet soft food production data, comprising the following steps:

[0065] S1: Based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rail, and pre-feeding silo, the system detects the changing trends of heat flow values and conveying cycles within a continuous operation cycle, identifies the degree of synchronization between the material heating load and the feeding rhythm, and obtains the synchronization status value of the wet-heat section data;

[0066] S2: Based on the synchronization status value of the wet and hot segment data, the continuity and time difference of the pet food processing node batch numbers are compared to determine whether the data is interrupted or overlapped between segments, and the batch data continuity matching degree is generated;

[0067] S3: Call the batch number corresponding to the continuous matching degree of batch data, extract the pH, nutrient composition and water content fluctuation data, analyze the floating trend of data points within the detection period, identify the concentrated offset of the quality inspection period, and obtain the stable segment of the quality inspection value sequence;

[0068] S4: Based on the stable segments of the quality inspection value sequence, match the packaging segment identifier with the discharge link data, analyze the interaction time and device response offset, filter the batches during the fluctuation period, compare the response consistency between devices, and output the collaborative data response offset set;

[0069] S5: Call the collaborative data response offset set to extract the energy records and alarm logs of abnormal batches in the feeding and mixing area, drying chamber and cooling transmission line. By cross-screening the record frequency and offset batches, the data anomaly concentration area is located to obtain the concentrated detection area of abnormal batches in soft grain production.

[0070] The synchronization status values of the wet and hot section data include the amplitude of the heat flow intensity change, the synchronization rate of the conveying rhythm, and the consistency index of the heating rhythm. The continuous matching degree of batch data includes the batch number connection rate, the time interval error value, and the data paragraph matching degree. The stable section of the quality inspection value sequence includes the pH value stable range, the fluctuation amplitude of the nutritional index, and the moisture content stability coefficient. The collaborative data response offset set includes the packaging label response delay, the discharge trigger consistency deviation, and the equipment interaction time difference. The centralized detection area for abnormal batches of soft grain production includes the abnormal energy consumption frequency band, the early warning log concentration node, and the offset overlap area between work sections.

[0071] See also Figure 2 The specific steps for obtaining the synchronization status value of the wet and hot section data are as follows:

[0072] S111: Based on data from the pet food wet heat production process, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos, the system identifies and compares the heat output values and conveyor rail operating rates within consecutive cycles to obtain a heat cycle matching trend value.

[0073] In the production of soft pet food, controlling thermal energy output and material conveying rate is key. For example, suppose a production line needs to match thermal energy output and conveying rate to ensure the continuity of the production process and the quality of the material. During specific operations, sensors are called to monitor the thermal energy output value of the heating treatment chamber and the operating rate of the material conveying track. The data acquisition process may include regularly sending data to the control configuration through temperature and speed sensors. The control configuration performs preliminary analysis of the data, such as calculating the thermal energy value and conveying speed at each time point. Through continuous monitoring of the data, the heating temperature or conveying rate can be adjusted in time to ensure the efficient operation of the production line. During the comparison process, the changing trends of thermal energy and conveying cycle in different time periods will be calculated. For example, the linear regression analysis method is used to determine the trend line, so as to determine whether the production parameters need to be adjusted to adapt to the changes in raw material characteristics and generate a thermal energy cycle matching trend value.

[0074] S112: Based on the heat energy cycle matching trend value, the material delivery cycle and synchronization interval data are called, rhythm difference comparison is performed on the corresponding sections, and the amplitude distribution range is identified to generate the heating rhythm offset amplitude interval;

[0075] Compare this data with the material delivery cycle data to ensure synchronization between the two during the production process. For example, in actual production, if a deviation is found between the material delivery cycle and the heat supply, it can be corrected by adjusting the feeder rate or changing the heating strategy. The specific execution process includes calling the synchronization interval data, such as obtaining the material delivery data for the time period corresponding to the heat matching trend through database query, and then using data analysis software to compare the rhythm difference between the two and calculate the deviation amplitude. This calculation includes finding the maximum and minimum deviation values within each time period and drawing a deviation distribution chart so that operators can intuitively understand the synchronization status of the entire production line and generate the heating rhythm deviation amplitude interval.

[0076] S113: Based on the heating rhythm deviation amplitude range, extract the start and end time and change amplitude of the synchronization section of the heat energy trend and material delivery, identify the fluctuation section, and use the formula:

[0077]

[0078] Compare the offset density value with the thermal stability reference value to obtain the synchronization status value of the wet and hot section data;

[0079] Among them, S represents the synchronization status value of the wet and hot section data, Th i Represents the time length of the i-th section of the thermal energy flow, Tm i Represents the time length of the i-th section of the material delivery period, H i Represents the thermal energy density of the heat flow value in the i-th section, M i Represents the density of material delivery in the i-th section, A i represents the dynamic load value of the conveyor rail in the i-th section, and n represents the total number of sections;

[0080] Call the start and end time and change amplitude of all synchronization segments in the heat energy cycle matching trend value and the material delivery cycle to identify the synchronization segments with offset fluctuations. First, explain the parameters involved in the formula and the acquisition process.

[0081] Th i is the duration of the thermal energy change in the i-th section, which can be recorded by the thermal energy data acquisition system. The starting and ending time of the thermal energy flow change is in seconds;

[0082] Tm i The duration of the corresponding material delivery in the same section is obtained through the material delivery record, and the start and end time of the continuous delivery section is obtained in seconds;

[0083] H i is the thermal energy density in joules per second, which is obtained by dividing the total amount of thermal energy output per unit time by the length of the time period;

[0084] M iThe material density is expressed in kg / s and can be obtained by dividing the total amount of material put into the time period by the length of time;

[0085] A i is the dynamic load value of the conveyor rail in the i-th section, in kg / m, which is calculated by converting the material weight and the current section length of the conveyor rail. It is obtained by PLC collecting data from the load weight sensor and the conveyor rail position sensor. All parameters have been standardized and unified into basic SI units.

[0086] Assume that the synchronization segment is divided into three segments, and the specific collected values are as follows:

[0087] First section: Th1 = 40 s, Tm1 = 30 s, H1 = 800 J / s, M1 = 5 kg / s, A1 = 40 kg / m;

[0088] Second section: Th2 = 60 s, Tm2 = 65 s, H2 = 950 J / s, M2 = 4.5 kg / s, A2 = 42 kg / m;

[0089] The third section: Th3 = 50 s, Tm3 = 55 s, H3 = 900 J / s, M3 = 5.2 kg / s, A3 = 41 kg / m;

[0090] Substitute into the formula and expand the calculation terms:

[0091] Section 1:

[0092] Section 2:

[0093] Section 3:

[0094] molecular:

[0095] Denominator:

[0096] Segment 1: 800+5+40=845;

[0097] Section 2: 950 + 4.5 + 42 = 996.5;

[0098] Section 3: 900 + 5.2 + 41 = 946.2;

[0099] sum:

[0100] Final calculation results:

[0101] The results show that the synchronization status value of the wet and hot section data is about 0.467. If the empirical benchmark value under the historical stable state is 0.35, the current value has exceeded the benchmark, indicating that there is a significant offset between the synchronization sections. The material delivery rhythm or heat output should be recalibrated to control the synchronization error. By combining the absolute amount of the time difference with the coupling product of heat energy density and material density to form the numerator, and the total material-heat energy-load factor to form the denominator, the synchronization status value can fully reflect the dynamically changing coupling strength and load stability in the production process, and has a strong recognition ability for monitoring and identifying abnormal sections.

[0102] See also Figure 3 , the specific steps for obtaining the continuous matching degree of batch data are:

[0103] S211: Extract the batch number of the processing node within the time period based on the synchronization status value of the wet and hot section data, analyze the relationship between the number and the time sequence, identify the offset between the batch number and the collection time, and obtain the batch time sequence deviation structure;

[0104] In the production of soft pet food, ensuring the continuity and data integrity of batch numbers is a key step in optimizing the production process. Relevant time series are extracted from the synchronized status values of the wet and hot section data, and further mapped to the batch numbers of the processing nodes, visually displaying the distribution of each batch in the time series. For example, assuming that within a certain time window, the batch number is observed to increase from 1001 to 1005, and the timestamps of each batch number are recorded, and the timestamp differences between two consecutive batches are compared, this method can intuitively identify possible time interval anomalies or number jumps in data collection, and refine the analysis of the matching degree between batch numbers and timestamps. For example, if the timestamp of batch 1002 is found to be 30 minutes later than that of batch 1001, and the timestamp of batch 1003 is found to be 10 minutes earlier than that of batch 1002, it indicates that there is an anomaly in the data recording or production process. Through such detailed data comparison, a batch time series deviation structure is generated, providing a quantitative monitoring method for production management.

[0105] S212: Based on the batch time sequence deviation structure, identify number gaps and time overlaps, analyze the cumulative quantity and classify them, evaluate the summary distribution of number and time anomalies, and obtain the number time offset distribution characteristics;

[0106] Based on the analysis results of the batch timing deviation structure, we further delve into the specific deviations between batch numbers and timestamps. If the analysis results show that the number of a batch suddenly jumps from 102 to 106, and its timestamp has only a slight increase compared to the previous batch, this indicates that an error occurred in the numbering or data entry process. By setting a threshold, we can determine the degree of number jump or time overlap that is acceptable. Exceeding the threshold is considered an anomaly. Through aggregate analysis, we can identify all batches that violate the normal production sequence and calculate the proportion of these abnormal batches in the overall production, thereby generating number time offset distribution characteristics. This feature not only helps production managers quickly locate problem areas, but also provides data support for subsequent process optimization.

[0107] S213: Call the number time offset distribution feature to perform a quantitative analysis of the number difference, time fluctuation and pairing integrity for each type of anomaly using the formula:

[0108]

[0109] Identify the batch number and time coupling consistency index, perform segmented comparison based on the preset threshold range, and obtain the continuous matching degree of batch data;

[0110] Among them, Ψ represents the continuous matching degree of batch data, Δb j Represents the batch number offset of the j-th batch pair, Δt j represents the squared time offset of the j-th batch pair, δ j Represents the identification value of the j-th batch number and time pair, Γ j represents the number of pairs in the jth batch, Θ j represents the number of time tags for the j-th batch pair, Ω j represents the total number of material record events for the jth batch pair, and m represents the total number of batch pairs in the analysis time period;

[0111] After obtaining the numbering time offset distribution characteristics, we further constructed a computational model for evaluating the continuity matching degree of batch data, and introduced statistical indicators to comprehensively quantify the numbering differences and time offsets.

[0112] Ψ represents the degree of batch data continuity matching, and its value range is from 0 to 1. The closer the value is to 1, the stronger the batch continuity;

[0113] Δb j : The batch number offset of the jth segment, in dimensionless units, can be obtained by subtracting 1 from the difference between the current batch number and the previous batch number;

[0114] Δt j: The time offset value of the jth segment, in minutes, is obtained by subtracting the standard beat interval from the difference between the current batch recording time and the previous batch time;

[0115] δ j : The marker item for the segment where the number and time are completely matched, if it meets the conditions, it is recorded as 1, otherwise it is recorded as 0;

[0116] Γ j : The number of valid numbered records in the jth segment;

[0117] Θ j : The number of valid time stamps of the jth segment;

[0118] Ω j : The feeding record event corresponding to the jth segment;

[0119] Taking a period of actual pet food processing time as an example, three consecutive batches of data were selected for calculation:

[0120] Section 1: The numbers are from 1001 to 1003, which are actually 3 batches. The numbers should be continuous. The measured number difference is 2. The offset value Δb1 = (1003-1001)-2 = 0. The time differences are 8 minutes and 12 minutes respectively. The standard beat is 10 minutes. The pairing is normal δ1=1, the valid record is Γ1=3, Θ1=3, and the feeding event Ω1=2;

[0121] Section 2: Numbers are from 1004 to 1006, with number 1005 missing. The offset Δb2 = (1006-1004)-2 = 0. However, 1005 does not actually appear, so let δ2 = 0. The time offset is 15 minutes and 10 minutes. Γ2=3, Θ2=2, Ω2=3;

[0122] Segment 3: The numbers are consecutively 1007 to 1008, the number offset Δb3 = (1008-1007)-1 = 0, the time offset is 9 minutes, and the standard beat is 10 minutes. δ3=1, Γ3=2, Θ3=2, Ω3=2;

[0123] Substitute the above values into the formula for calculation:

[0124] Molecular part:

[0125]

[0126] Numerator sum:

[0127] Denominator:

[0128] Final calculation results:

[0129] The result shows that the batch data continuity matching degree is 0.818, which is below the standard ratio threshold of 0.85, indicating that some numbers are missing or time is misaligned in the current sampling section, and the numbers are not fully aligned with the acquisition rhythm. This result can be directly used to determine whether the data segment needs to be traced or re-collected. In subsequent steps, the operation threshold or correction strategy should be set based on this matching value. In combination with the three factors of number difference, time fluctuation and pairing integrity, it is normalized through multi-level operations such as square root, absolute value and summation to avoid excessive influence of a single item, reflecting the overall offset strength of data continuity, thereby effectively improving the accuracy of judging data matching integrity.

[0130] See also Figure 4 The specific steps for obtaining the stable segment of the quality inspection value sequence are as follows:

[0131] S311: Call the batch number corresponding to the batch data continuous matching degree, extract the pH, nutrient composition and water content fluctuation data, monitor the fluctuation trend of the data within the detection period, and obtain a preliminary data sequence of the fluctuation trend;

[0132] First, the batch number is obtained, and then the fluctuation data of pH, nutrients and water content are extracted. The data is then monitored and analyzed in detail. The data sources used in the analysis process include various time nodes in the production process. For pH value, nutrients and water content, data points are first obtained through sampling, and the consistency of the sampling cycle is ensured. For example, if data is collected once an hour, then by comparing the changes in pH value over multiple time periods, the fluctuation of the production environment can be observed. The fluctuation trend of the data points is analyzed based on the acquired data, especially using statistical methods such as standard deviation and variance to describe the changes in the data. Taking pH value as an example, assuming that the fluctuation range of pH value is between 6.5 and 7.2 in a production cycle, the fluctuation data can be used to analyze the variation range of pH value in the cycle, and obtain a preliminary data series of the fluctuation trend, that is, the trend of the fluctuation range and fluctuation frequency calculated by monitoring data.

[0133] S312: Based on the preliminary data sequence of the fluctuation trend, analyze the fluctuation trend, identify the concentrated offset within the quality inspection period by analyzing the change of the fluctuation value, mark the offset interval, and obtain the offset abnormal fluctuation interval;

[0134] To analyze the fluctuation data of pH, nutrients and water content, the fluctuation range of the data points is first refined. For example, assuming that the fluctuation range of pH value is from 6.5 to 7.2 in a certain period of time, and the fluctuation is periodic, then the average value of the periodic fluctuation can be used for further analysis. By comparing the fluctuations of pH, nutrients and water content in different time periods, combined with the production process parameters, the deviation of the fluctuation trend in a specific period can be found, and possible quality inspection problems can be marked. In the data analysis process, it is necessary to first select a suitable threshold to judge the abnormality of the fluctuation. For the fluctuation of pH value, if it exceeds the preset standard range (for example, pH value range 6.0 to 7.0), the fluctuation can be considered to be offset. If the pH value in a certain time period reaches 7.5, the data point is considered to deviate from the normal range and belong to the offset segment. Through analysis, the offset abnormal fluctuation interval is obtained, which provides the necessary data support for the subsequent quality inspection process.

[0135] S313: Filter the data based on the deviation abnormal fluctuation range to remove irrelevant data using the formula:

[0136]

[0137] Calculate the stability value of the quality inspection value sequence and obtain the stable segment of the quality inspection value sequence;

[0138] Among them, Z represents the stability value of the quality inspection value sequence, P k represents the detection value of the kth data point, represents the mean of the test values, K represents the total number of data points, and V represents the estimated value of the standard deviation of the fluctuation value;

[0139] Screen out data that does not meet the requirements and retain numerical data with strong stability. During the screening process, it is necessary to discard data with large fluctuations to ensure that the final result can accurately reflect the stability of the production process. To achieve this, the fluctuation of the screened data must be calculated first to ensure that the stability of the data can be accurately measured within different production cycles. To achieve this, the fluctuation calculation formula is used.

[0140] Z represents the stability of the stable segment of the quality inspection value sequence (dimensionless), which is used to measure the stability of data point fluctuations;

[0141] P k is the actual value of the kth data point, in the same units as the measurement of that data point (e.g., pH is in pH and water content is in %);

[0142] Represents the average of all data points, and the unit is P k same;

[0143] K is the total number of data points, indicating how many data points are involved in the calculation;

[0144] V represents the standard deviation of the fluctuation value, and its unit is the same as P k The same calculation formula is:

[0145] Assume there are the following five data points: pH values are 6.8, 7.1, 6.9, 7.0, and 6.7;

[0146] First, calculate the mean of the data points

[0147] Calculate the mean value for each data point and square the difference: (6.8-6.9) 2 =0.01, (7.1-6.9) 2 =0.04, (6.9-6.9) 2 =0, (7.0-6.9) 2 =0.01, (6.7-6.9) 2 =0.04;

[0148] Sum these squared differences:

[0149] Then, calculate the standard deviation V of the fluctuation values:

[0150] Calculate the volatility Z:

[0151] Calculate the absolute difference between each data point and the mean: |6.8-6.9|=0.1, |7.1-6.9|=0.2, |6.9-6.9|=0, |7.0-6.9|=0.1, |6.7-6.9|=0.2;

[0152] The sum of the absolute differences is:

[0153] Substitute the calculated value into the formula:

[0154] The calculation shows that the stability value of the quality inspection value sequence Z = 1.897 indicates that the data volatility is low and the overall data fluctuation is relatively stable. This stability value shows that the screened stable segment has low volatility, and it can be considered that the data can accurately reflect the stability of the production process.

[0155] See also Figure 5 , the specific steps for obtaining the collaborative data response offset set are:

[0156] S411: Based on the stable segment of the quality inspection value sequence, match the packaging segment identifier with the discharge link data, identify the discharge batch number corresponding to the packaging sequence, extract the discharge time point and the packaging start time point, calculate the time difference between the two time points, and obtain the packaging and discharge interaction time;

[0157] First, extract the discharge record number corresponding to the stable segment and determine its start and end index values in the production batch table. Then, retrieve the corresponding packaging segment identifier from the packaging data record table, including information such as the packaging number, packaging start and end times, and packaging equipment number. Each packaging record is then matched one by one to find the corresponding discharge batch. By comparing the packaging start time with the discharge end time recorded in the discharge batch, the time difference between packaging and discharge is calculated. For example, if the discharge end time is 09:30:00 and the packaging start time is 09:35:20, the interaction time is 320 seconds. This time difference forms the basis for the data to be obtained in this section. The batch and time fields are used to calculate and generate the time difference value for each matching record. All the time differences in the records are then sorted by batch number to form a complete interaction time dataset. To avoid the impact of packaging mismatches or discharge signal delays, only records with interaction times between 120 seconds and 480 seconds are retained. During the application process, the interaction time generated by different packaging equipment is counted. If the batch interaction time of a certain device is concentrated at the upper limit (such as exceeding 480 seconds), the device has a waiting or signal blocking problem, and the packaging discharge interaction time is obtained.

[0158] S412: Extract device response data during the interaction period based on the packaging and discharging interaction time, analyze the time interval between the interaction time and the signal trigger time, identify the response offset starting point and duration, filter out batches that meet the conditions, and obtain a device response offset batch set;

[0159] Extract the device response data for the time period of each interaction time data entry. Device response data typically includes fields such as temperature control start signals, motor start and stop records, and feed current trigger records. First, map the interaction time period to the device log data to obtain the device state change records for that time period. For example, if the interaction time for a batch is 320 seconds and the packaging start time is 09:35:20, the device response segment is between 09:30:00 and 09:35:20. Extract the device state change timestamps within this time range. If the device experiences a sudden speed change at 09:33:10 and triggers a temperature control alarm at 09:33:25, this indicates an abnormal device response within that time period. By constructing a time series comparison, the offset time between the time point of the device state change and the interaction time point is compared item by item. For example, if the speed change is 90 seconds different from the interaction center point, it is recorded as an offset value of 90. If the offset value exceeds the preset offset threshold of 180 seconds, it is considered that the device has a response offset in this batch. The offset threshold is set according to the type of device. For example, the threshold can be set to 120 seconds for small packaging equipment and 240 seconds for large equipment. Finally, all batches with offset values exceeding the threshold are screened out and summarized into a list to obtain the device response offset batch set.

[0160] S413: For the equipment response offset batch set, identify the equipment data of the corresponding batch, extract the motor start and stop time, temperature control signal data, discharge port displacement value and feeder speed value, and use the formula:

[0161]

[0162] Calculate the data response difference value, identify the batch data associated with the devices with inconsistent responses in the same cycle based on the difference value, aggregate them into the structure set, and output the coordinated data response offset set;

[0163] Where R represents the data response difference value, u represents the total number of periodic signals collected by device A, Q represents the total number of periodic signals collected by device B, q represents the change in the forward signal of device A, x represents the change in the forward signal of device B, r represents the change in the reverse signal of device A, and y represents the change in the reverse signal of device B.

[0164] Call the signal records of all devices involved in the batch to obtain the total number of signals, the number of positive jumps, and the number of negative jumps for each device in the same sampling period. To ensure dimensional consistency, all signal data must be uniformly converted according to the "number of signal events per minute". Taking batch number B0321 as an example, assume that the batch generates data between 10:00:00 and 10:05:00, with a sampling period of 5 minutes. The sampling period is fixed and the unit conversion is performed. During this time period, device A records 20 signals and device B records 16 signals, with a total of u = 20 and Q = 16 respectively. According to the sensor log records, device A generates 8 positive jumps and 4 negative jumps during this period, and device B generates 6 positive jumps and 5 negative jumps. Therefore, q = 8, r = 4, x = 6, and y = 5;

[0165] To perform the calculation, the steps are as follows:

[0166] The first step is to calculate the difference in the total number of signals as uQ = 20 - 16 = 4;

[0167] The second step is to calculate the total jump amount as q+r+x+y=8+4+6+5=23;

[0168] The third step is to calculate the positive jump difference as qx = 8-6 = 2;

[0169] Step 4: Substitute the values into the formula to get:

[0170] In this formula, all parameters have been converted to the "number of events / period" dimension under the same sampling period to avoid interference with the results caused by differences in sampling rates between different devices;

[0171] Where R represents the data response difference value (unitless), which is used to measure the degree of difference in the response of the control signal of two devices within the same time period. u represents the total number of periodic signals of device A (bars / period), Q represents the total number of periodic signals of device B (bars / period), q represents the number of positive signal jumps of device A (times / period), x represents the number of positive signal jumps of device B (times / period), r represents the number of negative signal jumps of device A (times / period), and y represents the number of negative signal jumps of device B (times / period).

[0172] By performing linkage multiplication processing on the two key signal difference items, the response differences between devices can be clearly quantified and corresponded to a specific batch range, thereby enhancing the ability to explain the source of the differences. The results show that the response difference value R is 0.3478, which exceeds the preset benchmark value of 0.3, indicating that there is a response difference problem between device A and device B in batch B0321 within the same cycle. This batch should be marked as an abnormal batch and included in the collaborative data response offset set.

[0173] See also Figure 6 The specific steps for obtaining the centralized detection area for abnormal batches of soft grain production are as follows:

[0174] S511: Call the collaborative data response offset set to extract the batch numbers and equipment process identifiers of abnormal batches in the feeding and mixing area, drying chamber, and cooling transmission line, capture the corresponding electricity, steam, and water flow data, count the record frequency of each batch in the area, and generate the section energy record frequency set;

[0175] The equipment process identifiers associated with each batch in the feeding and mixing area, drying chamber, and cooling transmission line were obtained based on the batch number. Energy records during equipment operation were extracted. Energy records include three parameters: electricity meter readings, cumulative steam valve on / off time, and cooling water flow meter cycle flow values. For each equipment area, the number of records for each abnormal batch within a 5-minute sampling period was counted. The statistical results were summarized and organized into a frequency matrix using a "batch-area" secondary index. Data recording frequency is defined as the number of times each type of energy data is recorded per unit time. If an abnormal batch has 12 water flow records in the cooling transmission line, 6 steam records in the drying chamber, and 3 electricity records in the feeding and mixing area, its frequency vector is [3, 6, 12]. This three-dimensional frequency structure is formed for all abnormal batches and used as input for subsequent analysis to obtain the energy record frequency set for the work section.

[0176] S512: Extract alarm logs corresponding to abnormal batches based on the work section energy record frequency set, classify and count them by regional equipment and alarm level, cross-screen the alarm batches of energy records, and obtain an alarm data screening result set;

[0177] The alarm log data generated by the corresponding batch in the three equipment areas is extracted. The alarm type, trigger time and equipment number contained in the data package are extracted by area, and the data is organized into a unified format into a time-aligned alarm index table. Each log needs to be cross-judged with the corresponding energy recording time window, and alarm records that only occur during the energy recording period are retained. If the steam energy consumption of a batch is recorded from 10:00 to 10:05, the alarm data triggered within this time period must be matched, and invalid alarms with no time period intersection are filtered out. The alarm level is divided into three levels according to the settings: 1, 2, and 3, representing minor, moderate, and severe alarms, respectively. Each type of alarm is counted separately, and each batch data forms a total alarm record for the three types of equipment, such as 2 feeding and mixing areas, 4 drying chambers, and 1 cooling line. Through this screening and merging process, the regional alarm statistical structure of each batch is obtained, and the alarm quantity mapping between batches and areas is established to obtain the alarm data screening result set.

[0178] S513: Filter the result set based on the alarm data, count the frequency and distribution ratio of repeated abnormal batches within the equipment area, match the area identifier with the batch coordinate index, select the section area with a frequency higher than the positioning benchmark, integrate the positioning label and coordinate mapping data, and obtain the concentrated detection area for abnormal batches in soft grain production;

[0179] Abnormal batch numbers are extracted by equipment area classification and frequency statistics are performed. The statistical indicator is the number of alarms for each batch number in a single area. If batch B021 appears 4 times in the feeding and mixing area, 2 times in the drying chamber, and 6 times in the cooling line, the statistical frequencies are 4, 2, and 6 respectively. The positioning benchmark frequency value is set to 3 as the criterion for judging the density of batch abnormalities. The frequency values of all batches appearing in the three types of areas are compared with the benchmark values, and the regional labels with frequencies greater than the benchmark values are marked. For example, if B021 exceeds the benchmark in both the feeding area and the cooling line, it is marked as a "feeding + cooling" label. The coordinate field attached to the batch is then called for mapping, and the label and batch coordinates are merged into a two-dimensional area recognition unit. A regional coordinate index table is established, and the structured data is organized with the label as the index to obtain a concentrated detection area for abnormal batches in soft grain production.

[0180] The pet soft food production data management system is used to implement the above-mentioned pet soft food production data management method, and the system includes:

[0181] The thermal energy synchronization module is based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos. It extracts matching segments between the thermal energy flow and the conveying cycle within a continuous cycle, counts the number of matching points within the cycle, determines the direction of change in the number of matching points and the segment density, locates the positive and negative frequency conversion segments, and establishes the synchronization status value of the wet-heat segment data.

[0182] The batch continuity module extracts the corresponding processing batch number based on the synchronization status value of the wet and hot section data, compares the time difference and number spacing of adjacent batches, identifies the gap and overlap sections, and establishes the batch data continuity matching degree;

[0183] The quality inspection stability module extracts the pH value, nutrient composition, and water content data of each batch based on the continuous matching degree of batch data, analyzes the trend of the range and mean difference within the continuous period, selects the concentrated section of stable values, and establishes the stable section of the quality inspection value sequence;

[0184] The response offset module extracts the associated packaging segment identification time and discharge signal time based on the stable segment of the quality inspection value sequence, calculates and sorts the distribution of equipment response time differences, locates the numbering positions of response inversion and interval mutation, tracks the corresponding equipment combination and matching offset frequency, and establishes a collaborative data response offset set;

[0185] Based on the collaborative data response offset set, the anomaly positioning module extracts the energy logs and alarm logs of related batches in the feeding and mixing area, drying chamber, and cooling transmission line, filters the intersection of frequency and alarm period, and establishes a centralized detection area for abnormal batches in soft grain production.

[0186] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for managing pet soft food production data, characterized in that: The following steps are involved: S1: Based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rail, and pre-feeding silo, the system detects the changing trends of heat flow values and conveying cycles within a continuous operation cycle, identifies the degree of synchronization between the material heating load and the feeding rhythm, and obtains the synchronization status value of the wet-heat section data; S2: Based on the synchronization status value of the wet and hot segment data, the continuity and time difference of the pet food processing node batch number are compared to determine whether the data is interrupted or overlapped between segments, and the batch data continuity matching degree is generated; S3: Call the batch number corresponding to the continuous matching degree of the batch data, extract the pH, nutrient composition and water content fluctuation data, analyze the floating trend of the data points within the detection period, identify the concentrated offset of the quality inspection period, and obtain the stable segment of the quality inspection value sequence; S4: Based on the stable segment of the quality inspection value sequence, match the packaging segment identification and the discharge link data, analyze the interaction time and equipment response offset, filter the batches in the fluctuation period and compare the response consistency between the devices, and output the collaborative data response offset set.

2. The pet food production data management method according to claim 1, characterized in that: The synchronization status value of the wet and hot section data includes the heat flow intensity change amplitude, the delivery rhythm synchronization rate, and the heating rhythm consistency index; the batch data continuous matching degree includes the batch number connection rate, the timing interval error value, and the data paragraph matching degree; the stable section of the quality inspection value sequence includes the pH value stable range, the nutritional index fluctuation amplitude, and the moisture content stability coefficient; the collaborative data response offset set includes the packaging label response delay, the discharge trigger consistency deviation, and the equipment interaction time difference.

3. The pet food production data management method according to claim 1, characterized in that: The steps for obtaining the synchronization status value of the wet and hot section data are specifically as follows: S111: Based on data from the pet food wet heat production process, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos, the system identifies and compares the heat output values and conveyor rail operating rates within consecutive cycles to obtain a heat cycle matching trend value. S112: Based on the heat energy cycle matching trend value, the material delivery cycle and synchronization interval data are called, rhythm difference comparison is performed on the corresponding sections, and the amplitude distribution range is identified to generate the heating rhythm offset amplitude interval; S113: Based on the heating rhythm offset amplitude range, extract the start and end time and variation amplitude of the synchronization section of the heat energy trend and material delivery, identify the fluctuation section, and use the formula: Compare the offset density value with the thermal stability reference value to obtain the synchronization status value of the wet and hot section data; Among them, S represents the synchronization status value of the wet and hot section data, Th i Represents the time length of the i-th section of the thermal energy flow, Tm i Represents the time length of the i-th section of the material delivery period, H i Represents the thermal energy density of the heat flow value in the i-th section, M i Represents the density of material delivery in the i-th section, A i Represents the dynamic load value of the conveyor rail in the i-th section, and n represents the total number of sections.

4. The pet soft food production data management method according to claim 3, characterized in that: The steps for obtaining the continuous matching degree of the batch data are specifically as follows: S211: extracting the batch number of the processing node within the time period according to the synchronization status value of the wet and hot section data, analyzing the relationship between the batch number and the time sequence, identifying the offset between the batch number and the collection time, and obtaining the batch time sequence deviation structure; S212: Based on the batch time series deviation structure, identify number gaps and time overlaps, analyze the cumulative quantity and classify them, evaluate the summary distribution of number-related and time-related anomalies, and obtain the number-time offset distribution characteristics; S213: Call the number time offset distribution feature to perform a quantitative analysis of number differences, time fluctuations, and pairing integrity for each type of anomaly using the formula: Identify the batch number and time coupling consistency index, perform segmented comparison based on the preset threshold range, and obtain the continuous matching degree of batch data; Among them, Ψ represents the continuous matching degree of batch data, Δb j Represents the batch number offset of the j-th batch pair, Δt j represents the squared time offset of the j-th batch pair, δ j Represents the identification value of the j-th batch number and time pair, Γ j represents the number of pairs in the jth batch, Θ j represents the number of time tags for the j-th batch pair, Ω j represents the total number of material record events for the jth batch pair, and m represents the total number of batch pairs in the analysis time period.

5. The pet food production data management method according to claim 4, characterized in that: The steps for obtaining the stable segment of the quality inspection value sequence are specifically as follows: S311: Calling the batch number corresponding to the batch data continuous matching degree, extracting pH, nutrient composition and water content fluctuation data, monitoring the fluctuation trend of the data within the detection period, and obtaining a preliminary data sequence of the fluctuation trend; S312: Analyze the fluctuation trend based on the preliminary data sequence of the fluctuation trend, identify the concentrated offset within the quality inspection period by analyzing the change of the fluctuation value, mark the offset interval, and obtain the offset abnormal fluctuation interval; S313: Filter the data according to the abnormal fluctuation range of the offset to remove irrelevant data, using the formula: Calculate the stability value of the quality inspection value sequence and obtain the stable segment of the quality inspection value sequence; Among them, Z represents the stability value of the quality inspection value sequence, P k represents the detection value of the kth data point, represents the mean of the detected values, K represents the total number of data points, and V represents the estimated standard deviation of the fluctuation values.

6. The pet food production data management method according to claim 5, characterized in that: The steps for obtaining the collaborative data response offset set are specifically as follows: S411: Based on the stable segment of the quality inspection value sequence, matching the packaging segment identifier with the discharge link data, identifying the discharge batch number corresponding to the packaging sequence, extracting the discharge time point and the packaging start time point, calculating the time difference between the two time points, and obtaining the packaging and discharge interaction time; S412: Extracting device response data during the interaction period based on the packaging and discharging interaction time, analyzing the time interval between the interaction time and the signal triggering time, identifying the response offset starting point and duration, screening batches that meet the conditions, and obtaining a device response offset batch set; S413: For the device response offset batch set, identify the device data of the corresponding batch, extract the motor start and stop time, temperature control signal data, discharge port displacement value and feeder speed value, and use the formula: Calculate the data response difference value, identify the batch data associated with the devices with inconsistent responses in the same cycle based on the difference value, aggregate them into the structure set, and output the coordinated data response offset set; Among them, R represents the data response difference value, u represents the total number of periodic signals collected by device A, Q represents the total number of periodic signals collected by device B, q represents the change in the forward signal of device A, x represents the change in the forward signal of device B, r represents the change in the reverse signal of device A, and y represents the change in the reverse signal of device B.

7. The pet food production data management method according to claim 1, characterized in that: The method further comprises step S5: S5: Calling the collaborative data response offset set, extracting the energy records and alarm logs of abnormal batches in the feeding and mixing area, drying cabin, and cooling transmission line, and locating the data abnormality concentration area by cross-screening the record frequency and offset batches, and obtaining the concentrated detection area of abnormal batches in soft grain production; The centralized detection area for abnormal batches of soft grain production includes abnormal energy consumption frequency bands, early warning log centralized nodes, and offset overlap areas between work sections.

8. The pet soft food production data management method according to claim 7, characterized in that: The specific steps for obtaining the centralized detection area for abnormal batches of soft grain production are as follows: S511: Call the collaborative data response offset set to extract the batch numbers and equipment process identifiers of abnormal batches in the feeding and mixing area, drying chamber, and cooling transmission line, capture the corresponding electricity, steam, and water flow data, count the record frequency of each batch in the area, and generate a section energy record frequency set; S512: Extracting alarm logs corresponding to abnormal batches based on the energy record frequency set of the work section, classifying and counting them by regional equipment and alarm level, cross-screening the alarm batches of energy records, and obtaining an alarm data screening result set; S513: Filter the result set based on the alarm data, count the frequency and distribution ratio of repeated abnormal batches in the equipment area, match the area identifier and the batch coordinate index, filter the section area with a frequency higher than the positioning benchmark, integrate the positioning label and coordinate mapping data, and obtain the concentrated detection area for abnormal batches in soft grain production.

9. A pet soft food production data management system, characterized in that: The system is used to implement the pet soft food production data management method according to any one of claims 1 to 8, and the system includes: The thermal energy synchronization module is based on data from the wet-heat production section of pet soft food, including online operating information from the heating treatment chamber, material conveyor rails, and pre-feeding silos. It extracts matching segments between the thermal energy flow and the conveying cycle within a continuous cycle, counts the number of matching points within the cycle, determines the direction of change in the number of matching points and the segment density, locates the positive and negative frequency conversion segments, and establishes the synchronization status value of the wet-heat segment data. The batch continuity module extracts the corresponding processing batch number based on the synchronization status value of the wet and hot section data, compares the time difference and number spacing of adjacent batches, identifies the gap and overlap segments, and establishes the batch data continuity matching degree; The quality inspection stability module extracts the pH value, nutrient composition, and water content data of the batch based on the continuous matching degree of the batch data, analyzes the trend of the range and mean difference in the continuous period, screens the concentrated section of the stable value, and establishes the stable section of the quality inspection value sequence; The response offset module extracts the associated packaging segment identification time and discharge signal time based on the stable segment of the quality inspection value sequence, calculates and sorts the distribution of equipment response time differences, locates the numbering positions of response inversion and interval mutation, tracks the corresponding equipment combination and matching offset frequency, and establishes a collaborative data response offset set; Based on the collaborative data response offset set, the anomaly positioning module extracts the energy logs and alarm logs of the associated batches in the feeding and mixing area, drying cabin, and cooling transmission line, filters the intersection of frequency and alarm period, and establishes a centralized detection area for abnormal batches of soft grain production.

Images