Intelligent comparison system and method for food material procurement and acceptance based on internet of things

Abstract

The application discloses an intelligent comparison system and method for food material purchase and acceptance based on the Internet of Things, and relates to the technical field of intelligent comparison for food material purchase and acceptance.The system comprises the following steps: collecting environmental temperature and humidity and cargo surface temperature to generate a dew mark through a distributed sensing node; collecting a weighing sequence and determining an initial steady-state value and a segmented point through a networked weighing device; calculating an external condensation mass deduction amount according to the initial steady-state value and the segmented point; calculating a moisture absorption mass deduction amount in combination with a field prior parameter read by a data gateway; and finally calculating a deviation amount through the deduction amounts and performing a compliance judgment.The application can improve the acceptance accuracy of food material purchase and realize intelligent and efficient comparison.

Description

Technical Field

[0001] This invention relates to the field of intelligent comparison technology for food procurement and acceptance, and in particular to an intelligent comparison system and method for food procurement and acceptance based on the Internet of Things. Background Technology

[0002] In cold chain procurement scenarios involving fresh produce and prepared meals, goods are transferred from a low-temperature transportation environment to a receiving platform and checkweighing station with higher temperature and humidity. The cycle time for loading, unloading, depalletizing, weighing, and judgment is usually limited to a few minutes. At this time, the ambient air temperature and relative humidity rise significantly. If the surface temperature of the goods is below the dew point, moisture in the air will quickly condense on the surface to form a water film or droplets, resulting in a weight gain on a minute-by-minute scale. At the same time, cellulose materials such as cardboard boxes and corrugated liners will absorb moisture due to the increased relative humidity, raising the balance moisture content and adding extra weight. Receiving goods also requires rapid comparison with the order's net weight to form a traceable record. On-site equipment is often equipped with temperature and humidity sensors, surface temperature measurement devices, networked scales, and transport recorders. However, how to separate the two types of environmentally driven weight gain from the weighing results within a short time window and make compliance judgments accordingly has become the core challenge affecting the fairness and efficiency of settlement.

[0003] Existing technologies mainly employ methods such as extending waiting time, manual wiping, empirical reduction factors, direct comparison by single-point weighing, and independent air drying or heating. These methods generally suffer from the following drawbacks: First, waiting and wiping rely on manual labor, resulting in poor repeatability and impacting turnover efficiency; second, empirical reduction is not clearly established on the physical boundary between dew point and material isotherms, making it prone to systematic deviations due to environmental changes; third, single-point weighing ignores the rapid changes in external condensation and the slow changes in moisture absorption, failing to provide stable results within a fixed time window; and fourth, independent air drying or heating alters the state of the goods, making it unsuitable for food delivery acceptance. In summary, existing methods struggle to achieve accurate, repeatable, and traceable procurement acceptance within a short time window. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies where, during the acceptance process of food arrival, the weighing value is too high due to condensation on the surface of the goods and moisture absorption by the packaging materials, resulting in distorted acceptance results. Therefore, this invention proposes an intelligent comparison system and method for food procurement and acceptance based on the Internet of Things.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: The intelligent comparison method for food procurement and acceptance based on the Internet of Things includes: S1. Collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger flag based on the ambient air temperature, ambient relative humidity and cargo surface temperature; S2. Collect the weighing time series through networked weighing equipment, and determine the initial steady-state value and segmentation point based on the weighing time series; S3. Calculate the external condensation mass deduction based on the condensation triggering flag, initial steady-state value, and weighing values ​​at the segment points of the weighing time series; S4. Read the on-site prior parameters through the data gateway, and determine the moisture absorption mass deduction amount based on the condensation trigger flag and the on-site prior parameters; S5. Calculate the deviation based on the deduction amount for external condensation quality and the deduction amount for moisture absorption quality, and make a compliance judgment on the deviation amount.

[0006] Preferably, ambient air temperature, ambient relative humidity, and cargo surface temperature are collected through distributed sensing nodes, and a condensation trigger flag is generated based on the ambient air temperature, ambient relative humidity, and cargo surface temperature, including: The ambient air temperature, ambient relative humidity, and cargo surface temperature are collected synchronously by a temperature and humidity sensor and an infrared surface temperature sensor. The collected data is uploaded to the local data aggregation node via a wireless transmission protocol. Based on the collected ambient air temperature and relative humidity, the ambient thermal and humid state is converted into dew point temperature using the Magnus approximation. A condensation trigger flag is generated based on the difference between the dew point temperature and the surface temperature of the cargo. If the difference between the dew point temperature and the surface temperature of the cargo is greater than zero, the condensation trigger flag is set to one; otherwise, the condensation trigger flag is set to zero.

[0007] Preferably, the weighing time series is collected through a networked weighing device, and the initial steady-state value and segmentation points are determined based on the weighing time series, including: The networked weighing device outputs a weighing time sequence with a sampling period of one second. The sampling frequency is one hertz, the observation window is two hundred and forty seconds, and the data is transmitted to the edge computing node in real time via an industrial bus. Edge computing nodes obtain and cache the initial steady-state value by calculating the median of the weighing time series in the five to ten-second interval; The slope sequence of the weighing time series was calculated using the central difference method. Within a time interval of 30 to 240 seconds, the smallest time point that satisfies both the slope sequence value being less than a preset slope threshold and the rate of change of the slope sequence being less than zero is selected and determined as the segment point of the weighing time series.

[0008] Preferably, the external condensation mass deduction is calculated based on the condensation triggering indicator, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series, including: Read the weighing values ​​at the segment points of the weighing time series; If the condensation trigger flag is set to one, the difference between the weighing value at the segment point and the initial steady-state value will be used as the deduction amount for external condensation mass. If the condensation trigger flag is set to zero, then the external condensation quality deduction will be set to zero.

[0009] Preferably, the on-site prior parameters are read through the data gateway, and the moisture absorption mass deduction is determined based on the condensation trigger flag and the on-site prior parameters, including: The on-site pre-test parameters include: on-site ambient temperature, on-site ambient relative humidity, ambient relative humidity before delivery, and dry weight of paper packaging. The ambient relative humidity before delivery is transmitted back to the data platform by the cold chain monitoring terminal via a wide area network. The data platform normalizes the relative humidity of the on-site environment and the relative humidity of the environment before delivery; Based on the ambient temperature and the normalized relative humidity of the ambient environment, the first equilibrium humidity was calculated using the Hailwood–Horrobin isotherm. Based on the ambient temperature and the normalized relative humidity of the environment before delivery, the second equilibrium moisture content was calculated using the Hailwood–Horrobin isotherm. The dry basis moisture content increment is obtained based on the difference between the first equilibrium moisture content and the second equilibrium moisture content. The moisture absorption weight deduction is calculated based on the condensation trigger flag: if the condensation trigger flag is one, the product of the dry basis moisture content increment and the dry weight of the paper packaging is used as the moisture absorption weight deduction; otherwise, the moisture absorption weight deduction is set to zero.

[0010] Preferably, the deviation is calculated based on the deduction for external condensation mass and the deduction for moisture absorption mass, including: Read the tare weight and net content of the order; Read the final weighing value of the last second in the weighing time series; The comparable net weight is obtained by subtracting the external condensation mass deduction, the moisture absorption mass deduction, and the tare weight from the final weighing value. The deviation is calculated based on the difference between the comparable net weight and the net content of the order.

[0011] Preferably, the deviation is judged to be compliant. If the deviation is greater than the preset relative tolerance, it is judged to be unqualified; otherwise, it is judged to be qualified.

[0012] To address the aforementioned problems, this invention also provides an intelligent comparison system for food procurement and acceptance based on the Internet of Things, the system comprising: The condensation indicator module is used to collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger indicator based on the ambient air temperature, ambient relative humidity and cargo surface temperature. The weighing sequence module is used to collect the weighing time series through networked weighing devices and determine the initial steady-state value and segmentation points based on the weighing time series; The external condensation quality module is used to calculate the external condensation quality deduction based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series; The moisture absorption quality module is used to read the on-site prior parameters through the data gateway and determine the moisture absorption quality deduction based on the condensation trigger flag and the on-site prior parameters; The deviation determination module is used to calculate the deviation based on the external condensation quality deduction and the moisture absorption quality deduction, and to determine the compliance of the deviation.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention collects key temperature parameters of the environment and goods through distributed sensing nodes to generate condensation trigger flags. It obtains the weighing time series with the help of networked weighing equipment and determines the initial steady-state value and segment points. Based on the condensation state and weighing parameters, it calculates the external condensation weight deduction and combines the on-site prior parameters to calculate the moisture absorption weight deduction. Finally, after removing environmental interference by deduction, it calculates the deviation and completes the compliance judgment. In this way, within a limited window of minutes after arrival, the condensation weight gain and paper moisture absorption weight gain caused by the environment are separated from the weighing results, avoiding the comparison distortion caused by overweighing and improving the accuracy, repeatability and auditability of acceptance.

[0014] 2. This invention replaces empirical waiting with dew point calculation, ensuring that condensation judgment is based on the thermodynamic boundary of moist air; it obtains the initial steady-state value with a one-second sampling period, a 240-second observation window, and the median of the five- to ten-second interval, suppressing the influence of placement disturbances and extreme values; it obtains the slope sequence using the central difference method and determines the inflection point within 30 to 240 seconds, so that the rapid weight gain in the external condensation stage and the slow weight gain in the subsequent moisture absorption stage are distinguished and quantified over time; it maps the ambient temperature and relative humidity at the site and the relative humidity of the environment before delivery to the equilibrium moisture content at both ends, takes the difference, and combines it with the dry weight of the paper packaging to obtain the moisture absorption weight deduction, which only takes effect when the condensation trigger flag is set to one, thereby reducing the systematic influence of environmental differences on the symmetry of the quantity from the root. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 A flowchart illustrating an IoT-based intelligent comparison method for food procurement and acceptance, provided in an embodiment of the present invention. Figure 2 This is a functional module diagram of an IoT-based intelligent comparison system for food procurement and acceptance, provided as an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0017] Example: This example provides an intelligent comparison method for food procurement and acceptance based on the Internet of Things. See [link to example]. Figure 1 Specifically, including: S1. Collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger flag based on the ambient air temperature, ambient relative humidity and cargo surface temperature; In embodiments of the present invention, ambient air temperature, ambient relative humidity, and cargo surface temperature are collected by distributed sensing nodes. A condensation trigger flag is generated based on the ambient air temperature, ambient relative humidity, and cargo surface temperature, including: The ambient air temperature, ambient relative humidity, and cargo surface temperature are collected synchronously by a temperature and humidity sensor and an infrared surface temperature sensor. Specifically, a temperature and humidity sensor is used to collect ambient air temperature and relative humidity, while an infrared surface temperature sensor is used to collect the surface temperature of the cargo. Both sensors collect data synchronously to obtain time-consistent data on ambient air temperature, relative humidity, and cargo surface temperature. The temperature and humidity sensor and the infrared surface temperature sensor are deployed at designated collection locations in the cargo acceptance area. The temperature and humidity sensor collects ambient air temperature and relative humidity in real time, while the infrared surface temperature sensor collects cargo surface temperature in real time. Through a time synchronization mechanism, the temperature and humidity sensor and the infrared surface temperature sensor simultaneously initiate the data acquisition process at a preset sampling frequency under the same time reference, continuously acquiring real-time data on ambient air temperature, relative humidity, and cargo surface temperature. The collected data is processed by the signal processing unit inside the sensor and then sent to the data processing module in real time via a data transmission link, thus providing time-synchronized multi-source sensing data for the subsequent generation of condensation trigger indicators.

[0018] The collected data is uploaded to the local data aggregation node via a wireless transmission protocol. Based on the collected ambient air temperature and relative humidity, the ambient thermal and humid state is converted into dew point temperature using the Magnus approximation. Specifically, ambient air temperature and relative humidity data collected by temperature and humidity sensors are uploaded to the local data aggregation node via a wireless transmission protocol. The choice of wireless transmission protocol can be determined based on the actual requirements of data transmission distance and power consumption in industrial food acceptance scenarios. The local data aggregation node performs timestamp verification and synchronization integration on the received ambient air temperature and relative humidity data to ensure consistency between the two in the time dimension. Subsequently, the local data aggregation node calls the calculation logic based on Magnus approximation to convert the ambient thermal and humidity state into dew point temperature, providing the core temperature criterion for the subsequent generation of condensation trigger flags. The dew point temperature calculation is divided into two steps. The first step is to calculate the intermediate variable, and the formula for calculating the intermediate variable is:

[0019] In the formula, As an intermediate variable, For ambient relative humidity, The ambient air temperature, , Let be an empirical constant, where The value is 17.62. The value is 243.12; The formula for the second step is:

[0020] In the formula, Dew point temperature, As an intermediate variable, , Let be an empirical constant, where The value is 17.62. The value is 243.12; Specifically, this invention derives the calculation formula based on the thermodynamic definition of dew point and the Magnus approximation: saturated water vapor pressure increases exponentially with temperature. The Magnus empirical formula is used to characterize this monotonic relationship in logarithmic form, where the constant is... and Based on extensive field measurements and fitting, it is used for engineering calculations in near-surface, ambient temperature and humidity environments. The value is 17.62. The value is 243.12 degrees Celsius. The actual water vapor pressure of the air is equal to the saturated water vapor pressure at the same temperature multiplied by the relative humidity. Let the ambient temperature be T and the relative humidity be RH. Substituting this equilibrium relationship into the Magnus equation and taking the logarithm, we can obtain the intermediate variable. This variable combines the logarithmic term of relative humidity with the temperature term, equivalently describing the actual water vapor pressure at a given ambient temperature and relative humidity. The dew point temperature means that air reaches saturation at a certain temperature, i.e., the actual water vapor pressure equals the saturated water vapor pressure at that temperature. Substituting this condition back and solving algebraically, we obtain the dew point temperature as... Multiply Divide by minus The dew point temperature is analytically derived from the water vapor pressure equilibrium condition and the Magnus saturated water vapor pressure formula, and can stably determine whether condensation has occurred in the normal temperature and high humidity scenario of food receiving.

[0021] A condensation trigger flag is generated based on the difference between the dew point temperature and the surface temperature of the cargo. If the difference between the dew point temperature and the surface temperature of the cargo is greater than zero, the condensation trigger flag is set to one; otherwise, the condensation trigger flag is set to zero. Specifically, the local data aggregation node retrieves the dew point temperature and cargo surface temperature after time synchronization. The local data aggregation node uses its built-in difference calculation module to subtract the cargo surface temperature from the dew point temperature to obtain the temperature difference. Based on the physical law of water vapor condensation in engineering thermodynamics, when the ambient dew point temperature is higher than the object's surface temperature, water vapor in the air will inevitably condense on the object's surface. Based on this, a judgment rule is set: if the calculated temperature difference is greater than zero, the condensation condition is met, and the condensation trigger flag is set to one; if the temperature difference is less than or equal to zero, the condensation condition is not met, and the condensation trigger flag is set to zero. The setting of the condensation trigger flag to one or zero conforms to the conventional method of binary status identification in industrial control, facilitating subsequent logical judgments on whether external condensation mass deduction and moisture absorption mass deduction are required.

[0022] S2. Collect the weighing time series through networked weighing equipment, and determine the initial steady-state value and segmentation point based on the weighing time series; In an embodiment of the present invention, a weighing time series is collected through a networked weighing device, and the initial steady-state value and segmentation points are determined based on the weighing time series, including: The networked weighing device outputs a weighing time sequence with a sampling period of one second. The sampling frequency is one hertz, the observation window is two hundred and forty seconds, and the data is transmitted to the edge computing node in real time via an industrial bus. Specifically, the networked weighing device is deployed at the weighing station in the food acceptance area. Its weighing sensors, after calibration, possess metrological performance matching the accuracy requirements of food acceptance. The device's internal clock is synchronized with the system time, initiating weighing data acquisition at a sampling cycle of one second. This cycle setting is determined based on the physical characteristics of the external condensation process (rapid changes on the order of minutes) and the moisture absorption process (slow changes on the order of ten minutes), which can accurately capture the dynamic changes in the weighing value without data redundancy due to excessive sampling. Corresponding to the sampling cycle, the sampling frequency is set to one hertz to ensure that the amount of data collected per unit time is compatible with the time scale of the change process. The observation window is set to 240 seconds, which is determined based on the time requirement that the external condensation stage of the food is basically completed and the initial characteristics of the moisture absorption stage can be identified in the actual acceptance scenario, and can fully cover the key change range of the external condensation peak and the moisture absorption step. The weighing time series collected by the networked weighing device is transmitted via an industrial bus. The selected industrial bus meets the anti-interference and real-time requirements of data transmission in an industrial environment, ensuring that the weighing data is sent to the edge computing node immediately after acquisition, providing continuous and complete raw data for subsequent calculation of initial steady-state values ​​and segment points.

[0023] Edge computing nodes calculate the median of the weighing time series within a five- to ten-second interval to obtain the initial steady-state value and cache it. Specifically, after receiving the continuous weighing time series transmitted by the networked weighing device, the edge computing node first performs time axis calibration on the series to ensure that the time stamp of each data point accurately corresponds to the sampling time. Then, it extracts a data segment with a time interval of five to ten seconds from the weighing time series. This time interval is determined based on the physical stability characteristics of the goods after they are placed on the weighing device. Typically, the goods have already achieved initial stability within five seconds of placement, and the weighing values ​​within this five-to-ten-second interval are minimally affected by instantaneous disturbances during placement, reflecting the initial weight state of the goods. The edge computing node then calls the median calculation module to process the extracted five-to-ten-second data segment. The median of all weighing values ​​within a ten-second interval is used for calculation. The median is chosen as the initial steady-state value because there may be individual outliers caused by minor sensor noise or mechanical vibration within this interval. Compared with the arithmetic mean, the median is more resistant to such extreme value interference, ensuring the stability and representativeness of the initial steady-state value. The calculated initial steady-state value is stored in the local cache unit of the edge computing node according to a preset data format. The cache unit uses non-volatile storage medium to ensure that the data can be quickly retrieved in subsequent calculations and will not be lost due to temporary power outages, providing a reliable initial reference value for the calculation of external coagulation mass deduction.

[0024] The slope sequence of the weighing time series was calculated using the central difference method. Specifically, after acquiring the complete weighing time series, the edge computing node first performs outlier detection and correction, eliminating jump values ​​caused by instantaneous vibration or electromagnetic interference of the weighing device to ensure the continuity and reliability of the series. Then, it determines the time window parameters for the central difference method, selecting weighing values ​​5 seconds before and after the current time point for calculation. This window length is determined based on the physical characteristics of the external condensation process (rapid changes on the order of minutes) and the moisture absorption process (slow changes on the order of ten minutes). The 5-second window effectively smooths high-frequency noise and accurately captures the trend of weighing values ​​over time, avoiding calculation fluctuations caused by an excessively short window or trend lag caused by an excessively long window. For each time point t in the weighing time series, the edge computing node calls the differential calculation module to... Take the weighing values ​​at t+5 seconds and t-5 seconds, calculate the difference between them, and divide it by the 10-second time interval to obtain the slope value at that time point. For the first 5 seconds and the last 5 seconds of the time series (i.e., t less than 5 seconds or t greater than 235 seconds), since complete data for the first and last 5 seconds cannot be obtained, forward differencing or backward differencing is used for supplementary calculation. Forward differencing uses the difference between the weighing values ​​at t+5 seconds and t seconds divided by 5 seconds, and backward differencing uses the difference between the weighing values ​​at t seconds and t-5 seconds divided by 5 seconds to ensure the integrity of the slope sequence. Through the above process, a slope sequence with the same length as the weighing time series is generated. This sequence is stored in the temporary data area of ​​the edge computing node to provide dynamic change characteristics for the determination of subsequent segmentation points.

[0025] Within a time interval of 30 to 240 seconds, the smallest time point that satisfies the condition that the slope sequence value is less than the preset slope threshold and the rate of change of the slope sequence is less than zero is selected and determined as the segment point of the weighing time series. Specifically, the edge computing node first determines the time interval for screening segmentation points to be 30 to 240 seconds. This interval is set based on the physical characteristics of the external condensation process, which changes rapidly on a minute-by-minute basis, and the moisture absorption process, which changes slowly on a ten-minute-by-ten-minute basis. Within the time interval before 30 seconds, the weighing value may still be affected by the instantaneous disturbance at the initial stage of cargo placement. After 30 seconds, the external condensation process gradually emerges, while the moisture absorption process has not yet entered a stable phase. Screening within this interval ensures the capture of the critical node where external condensation transitions to moisture absorption. Subsequently, a slope threshold of 0.3 grams per second is set. This threshold is determined based on experimental data from a large number of food inspection scenarios. During the external condensation stage, the weighing value increases rapidly due to rapid surface condensation, and the slope is usually greater than 0.3 grams per second. However, after entering the moisture absorption stage, the slope drops below 0.3 grams per second because the moisture absorption rate of the paper material slows down. This serves as the quantitative boundary distinguishing the two stages. The edge computing node then calls the variable... The rate of change calculation module calculates the rate of change for each time point in the slope sequence. Specifically, it subtracts the slope value of the previous adjacent time point (with a time interval of one second) from the slope value of the current time point to obtain the rate of change. A rate of change less than zero indicates that the slope is decreasing, reflecting that the rate of increase of the weighing value is slowing down, which is consistent with the physical characteristics of the start of the moisture absorption process after the external condensation process ends. Then, the edge computing node traverses all time points in the interval from 30 seconds to 240 seconds, verifying one by one whether each time point simultaneously meets the conditions that the slope sequence value is less than 0.3 grams per second and the rate of change of the slope sequence is less than zero. The first time point that meets the conditions is determined as the segment point of the weighing time sequence. Selecting the minimum time point as the segment point can ensure accurate capture of the turning point between the end of the external condensation stage and the beginning of the moisture absorption stage, providing a time reference for the accurate calculation of the subsequent external condensation mass deduction and moisture absorption mass deduction.

[0026] S3. Calculate the external condensation mass deduction based on the condensation triggering flag, initial steady-state value, and weighing values ​​at the segment points of the weighing time series; In embodiments of the present invention, the external condensation mass deduction is calculated based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series, including: Read the weighing values ​​at the segment points of the weighing time series; If the condensation trigger flag is set to one, the difference between the weighing value at the segment point and the initial steady-state value will be used as the deduction amount for external condensation mass. If the condensation trigger flag is set to zero, then the external condensation quality deduction will be set to zero. Specifically, the edge computing node retrieves the determined segment point information from its local cache and extracts the corresponding weighing value from the weighing time series based on the time identifier of the segment point. During the extraction process, timestamp comparison is used to ensure that the read weighing value and the segment point are accurately matched in the time dimension. At the same time, the edge computing node retrieves the initial steady-state value calculated earlier from the cache, as well as the condensation trigger flag transmitted by the local data aggregation node. If the condensation trigger flag is set to one, it indicates that the surface of the cargo meets the condensation conditions and the external condensation process has occurred. At this time, the weighing value at the segment point is subtracted from the initial steady-state value, and the difference is the external condensation mass deduction. This calculation is based on the fact that the external condensation stage is the main weight gain process before the segment point, and the weight gain in this stage is mainly formed by water vapor condensed on the surface. If the condensation trigger flag is set to zero, it indicates that the surface of the cargo does not meet the condensation conditions and the external condensation process has not occurred. At this time, the external condensation mass deduction is set to zero. This setting is based on the logic of judging the difference between the dew point temperature and the surface temperature of the cargo. When the condensation conditions are not met, there is no external condensation weight gain, and no deduction is required.

[0027] S4. Read the on-site prior parameters through the data gateway, and determine the moisture absorption mass deduction amount based on the condensation trigger flag and the on-site prior parameters; In an embodiment of the present invention, prior field parameters are read through a data gateway, and the moisture absorption mass deduction is determined based on the condensation trigger flag and the prior field parameters, including: The on-site pre-test parameters include: on-site ambient temperature, on-site ambient relative humidity, ambient relative humidity before delivery, and dry weight of paper packaging. The ambient relative humidity before delivery is transmitted back to the data platform by the cold chain monitoring terminal via a wide area network. Specifically, the on-site pre-verification parameters are the basic data set used to determine the moisture absorption weight deduction, including the on-site ambient temperature, on-site relative humidity, pre-delivery relative humidity, and dry weight of the paper packaging. The on-site ambient temperature and relative humidity are environmental heat and humidity parameters collected in real time at the acceptance site. The pre-delivery relative humidity is collected by the cold chain monitoring terminal during the shipment transportation process and then transmitted back to the data platform via the wide area network. The dry weight of the paper packaging is the weight of the paper packaging material in a dry state.

[0028] The data platform normalizes the relative humidity of the on-site environment and the relative humidity of the environment before delivery; Specifically, the data platform obtains relative humidity data of the on-site environment from edge computing nodes and extracts pre-delivery relative humidity data from historical data transmitted from cold chain monitoring terminals. Both types of data are accompanied by collection timestamps to ensure timeliness. The data platform calls the data verification module to perform physical range verification on the two types of relative humidity data. Since the physical definition of relative humidity is the percentage of the partial pressure of water vapor in the air to the partial pressure of saturated water vapor at the same temperature, its reasonable value range is 0 to 100. Therefore, if the data exceeds this range, the system automatically corrects it to boundary values, where values ​​below 0 are corrected to 0, and values ​​above 100 are corrected to [missing value]. The relative humidity is set to 100 to conform to the physical characteristics of relative humidity. Subsequently, the data platform initiates a normalization calculation process, dividing the verified on-site relative humidity and the pre-delivery relative humidity by 100 to obtain a normalized result with a value range of 0 to 1. This processing method is determined based on the input parameter requirements of the Hailwood–Horrobin isotherm model. The model calculation requires relative humidity to be expressed in decimal form to ensure the accuracy of moisture content calculation. The normalized on-site relative humidity and the pre-delivery relative humidity are marked as standardized data and stored in the parameter database of the data platform.

[0029] Based on the ambient temperature and the normalized relative humidity of the ambient environment, the first equilibrium humidity was calculated using the Hailwood–Horrobin isotherm. The formula for calculating the first equilibrium moisture content is:

[0030]

[0031]

[0032]

[0033]

[0034] In the formula The first equilibrium moisture content, The ambient temperature at the site. The relative humidity of the site environment after normalization. These are temperature-related feature parameters in the Hailwood–Horrobin model. It is the temperature-dependent adsorption coefficient in the Hailwood–Horrobin model. It is the first temperature-related correction factor in the Hailwood–Horrobin model. It is the second temperature-related correction factor in the Hailwood–Horrobin model; Specifically, this invention uses the Hailwood–Horrobin theory to describe the hygroscopic isothermal behavior of cellulose materials, dividing the moisture in paper fibers into dissolved water and bound water, which are respectively determined by coefficients. and coefficients With coefficient The equilibrium constant is characterized by W, which is a characteristic molecular weight term that varies with temperature, reflecting the influence of adsorption potential energy and polymer chain free volume on hygroscopic capacity. The first term in the formula is... Multiply by the normalized relative humidity h and divide by one, then subtract Multiplying by h indicates that dissolved water increases approximately according to Raoult's law with humidity, tending towards saturation at high humidity. The numerator and denominator of the second term contain first and second terms with respect to h, corresponding to monohydric and dihydric effects respectively, reflecting multilayer adsorption. This causes the equilibrium humidity to exhibit an S-shaped characteristic, with approximately linear values ​​at low humidity, an inflection point at medium humidity, and a rapid increase at high humidity. The addition term in the denominator reflects site occupancy and saturation effects, limiting unbounded growth and providing a physically reasonable upper limit. Dividing the leading coefficient 1800 by W converts the result into a dry basis mass percentage, unifying material structural differences into the scaling factor. Thus, temperature and relative humidity are related through k, , The temperature function of W is mapped to the equilibrium moisture content, which can accurately characterize the moisture absorption and weight gain of paper and paperboard under on-site conditions.

[0035] Specifically, the temperature correction concept of the Hailwood–Horrobin model is adopted, and the effect of temperature on the adsorption equilibrium constant is approximated by a quadratic polynomial. This is based on the fact that in thermodynamics, the equilibrium constant with respect to temperature usually follows a van der Hoff exponential relationship, which can be approximated by linear and quadratic terms from a Taylor expansion within the engineering temperature range, thus characterizing the monotonic change and necessary curvature with minimal parameters. The increase in temperature W reflects the combined effect of the increase in free volume of the cellulose-based material and the change in the effectiveness of adsorption sites; therefore, positive linear and quadratic coefficients are used to represent the slow increasing trend. k represents the equivalent equilibrium constant of dissolved water. Adsorption is an exothermic process; increasing temperature promotes desorption and reduces affinity, hence the constant term is relatively large, the quadratic coefficient is negative, and the linear coefficient is very small to reflect limited sensitivity. The equilibrium constant for monohydrate is represented by a negative linear coefficient and a negative quadratic coefficient, which indicate that the adsorption capacity decreases continuously with increasing temperature, thus describing the accelerated decay. The equilibrium constant for dihydration is represented by a positive linear coefficient, which initially increases due to site activation in the low to medium temperature range, and decreases again in the high temperature range due to thermal disturbances and saturation inhibition, corresponding to a negative quadratic coefficient. All of the above temperature functions are derived from least-squares fitting of a large amount of measured data within commonly used temperature ranges. This maintains the monotonic or inflection-point characteristics consistent with physical laws while facilitating real-time calculations on field edge equipment, thus ensuring the stability and engineering feasibility of the equilibrium moisture content solution.

[0036] Based on the ambient temperature and the normalized relative humidity of the environment before delivery, the second equilibrium moisture content was calculated using the Hailwood–Horrobin isotherm. Specifically, the on-site ambient temperature and the normalized relative humidity before delivery are substituted into the same calculation process as the first equilibrium moisture content to obtain the second equilibrium moisture content. The first equilibrium moisture content refers to the mass moisture content of the paper packaging material when it reaches a dynamic equilibrium of adsorption and desorption with the surrounding air under the conditions of on-site ambient temperature and relative humidity, and is used to characterize the equilibrium moisture content level of the packaging material under on-site conditions. The second equilibrium moisture content refers to the theoretically achievable mass moisture content of the paper packaging material under the conditions of on-site ambient temperature and relative humidity before delivery, and is used to characterize the equilibrium moisture content level of the packaging material before it is transferred from the transportation or warehousing stage to the on-site environment. The dry basis moisture content increment is obtained based on the difference between the first equilibrium moisture content and the second equilibrium moisture content. The moisture absorption weight deduction is calculated based on the condensation trigger flag: if the condensation trigger flag is one, the product of the dry basis moisture content increment and the dry weight of the paper packaging is used as the moisture absorption weight deduction; otherwise, the moisture absorption weight deduction is set to zero. Specifically, subtracting the second equilibrium moisture content from the first equilibrium moisture content yields the dry basis moisture content increment. This increment reflects the change in moisture content caused by the difference in humidity between the pre-delivery environment and the on-site environment. Subsequently, the dry weight of the paper packaging transmitted from the data platform is acquired. This parameter represents the mass of the paper packaging material in a completely dry state, measured through prior experiments and pre-stored in the system. Simultaneously, the edge computing node reads the condensation trigger flag generated by the local data aggregation node. Based on the correlation between condensation and the moisture absorption process—when the condensation trigger flag is one, it indicates the presence of condensation on the surface of the shipment. When condensation occurs, the paper packaging material will absorb moisture significantly due to the high humidity environment. Therefore, a moisture absorption weight deduction needs to be calculated. The increase in dry basis moisture content is multiplied by the dry weight of the paper packaging. The resulting product is the moisture absorption weight deduction, representing the equivalent weight increase corresponding to the rise in equilibrium moisture content of the paper packaging material under the conditions of ambient temperature and relative humidity, relative to the relative humidity before delivery. This amount is deducted from the weighing result. When the condensation trigger flag is zero, it indicates that there is no condensation on the surface of the goods, and the impact of changes in ambient humidity on the moisture content of the paper packaging is negligible. Therefore, the moisture absorption weight deduction is set to zero.

[0037] S5. Calculate the deviation based on the deduction amount for external condensation quality and the deduction amount for moisture absorption quality, and make a compliance judgment on the deviation amount; In embodiments of the present invention, the deviation is calculated based on the external condensation weight deduction and the moisture absorption weight deduction, and compliance determination is performed on the deviation, including: Read the tare weight and net content of the order; Read the final weighing value of the last second in the weighing time series; The comparable net weight is obtained by subtracting the external condensation mass deduction, the moisture absorption mass deduction, and the tare weight from the final weighing value. Specifically, the tare weight is retrieved from the materials management system via a data interface. This tare weight is the weight of the food packaging material (excluding the food itself), pre-stored in the system after previous weighing calibration to ensure consistency with the actual packaging specifications. Simultaneously, the net weight of the order is retrieved from the purchase order system. This value represents the standard net weight of the food as stipulated in the purchase contract and serves as the benchmark for acceptance comparison. The weighing value corresponding to the last second is extracted from the weighing time series as the final weighing value. The selection of the last second data is based on the observation window being set to 240 seconds, which covers the main changes in external condensation and moisture absorption. As the weighing value stabilizes, it reflects the final weighing status of the goods under the acceptance environment. The edge computing node starts the net weight calculation module, which subtracts the external condensation weight deduction (weight of surface condensation water), the moisture absorption weight deduction (weight of paper packaging due to moisture absorption), and the tare weight (weight of the packaging itself) from the final weighing value. The result is the comparable net weight. This calculation logic eliminates the influence of environmental factors and packaging weight on the measurement results, so that the final comparable net weight can truly reflect the actual net content of the food, providing a reliable basis for subsequent comparison with the net content of the order.

[0038] The deviation is calculated based on the difference between the comparable net weight and the net content of the order. The deviation is judged to be compliant. If the deviation is greater than the preset relative tolerance, it is judged as unqualified; otherwise, it is judged as qualified. Specifically, the edge computing node calls the difference calculation module to subtract the order's net content from the previously obtained comparable net weight. The result is the deviation, which directly reflects the degree of difference between the actual net content of the ingredients and the agreed standard in the order. Then, the compliance judgment process begins. First, a preset relative tolerance is determined, which is the product of the order's net content and a tolerance coefficient. The tolerance coefficient is determined based on national metrological technical specifications and food industry acceptance standards, typically set to 0.5 to 2 for fresh ingredients, with the specific value adjusted according to the perishable nature of the ingredients and industry practices. The edge computing node takes the absolute value of the deviation to simultaneously control both situations where the actual net content exceeds or falls below the order's net content. If the absolute value is greater than the relative tolerance, it is judged as unqualified, indicating that the difference between the actual net content of the ingredients and the agreed standard exceeds the acceptable range. If the absolute value is less than or equal to the relative tolerance, it is judged as qualified, indicating that the difference is within a reasonable range. After the judgment result is generated, a calculation process log is attached and pushed to the acceptance terminal in real time and synchronized to the supply chain management system, forming a traceable acceptance conclusion record.

[0039] like Figure 2 The diagram shown is a functional block diagram of an intelligent comparison system for food procurement and acceptance based on the Internet of Things provided in an embodiment of the present invention.

[0040] In this embodiment, the functions of each module / unit are as follows: The condensation indicator module is used to collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger indicator based on the ambient air temperature, ambient relative humidity and cargo surface temperature. The weighing sequence module is used to collect the weighing time series through networked weighing devices and determine the initial steady-state value and segmentation points based on the weighing time series; The external condensation quality module is used to calculate the external condensation quality deduction based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series; The moisture absorption quality module is used to read the on-site prior parameters through the data gateway and determine the moisture absorption quality deduction based on the condensation trigger flag and the on-site prior parameters. The deviation determination module is used to calculate the deviation based on the external condensation quality deduction and the moisture absorption quality deduction, and to determine the compliance of the deviation.

[0041] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart comparison method for food procurement and acceptance based on the Internet of Things, characterized in that, include: S1. Collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger flag based on the ambient air temperature, ambient relative humidity and cargo surface temperature; S2. Collect the weighing time series through networked weighing equipment, and determine the initial steady-state value and segmentation point based on the weighing time series; S3. Calculate the external condensation mass deduction based on the condensation triggering flag, initial steady-state value, and weighing values ​​at the segment points of the weighing time series; S4. Read the on-site prior parameters through the data gateway, and determine the moisture absorption mass deduction amount based on the condensation trigger flag and the on-site prior parameters; S5. Calculate the deviation based on the deduction amount for external condensation quality and the deduction amount for moisture absorption quality, and make a compliance judgment on the deviation amount; The process of acquiring weighing time series through networked weighing devices and determining initial steady-state values ​​and segmentation points based on the weighing time series includes: the networked weighing device outputting the weighing time series at a sampling period of one second, wherein the sampling frequency is one hertz, the observation window is two hundred and forty seconds, and the data is transmitted in real time to the edge computing node via an industrial bus; the edge computing node obtains the initial steady-state value and caches it by calculating the median of the weighing time series in the five to ten second interval; the slope sequence of the weighing time series is calculated using the central difference method; within the time interval of thirty seconds to two hundred and forty seconds, the smallest time point that satisfies the condition that the slope sequence value is less than a preset slope threshold and the rate of change of the slope sequence is less than zero is selected, and the smallest time point is determined as the segmentation point of the weighing time series; The step of calculating the external condensation mass deduction based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series includes: reading the weighing values ​​at the segment points of the weighing time series; if the condensation trigger flag is one, then the difference between the weighing value at the segment point and the initial steady-state value is used as the external condensation mass deduction; if the condensation trigger flag is zero, then the external condensation mass deduction is set to zero. The on-site preliminary parameters include: on-site ambient temperature, on-site ambient relative humidity, ambient relative humidity before delivery, and dry weight of paper packaging. The ambient relative humidity before delivery is transmitted back to the data platform from the cold chain monitoring terminal via a wide area network.

2. The intelligent comparison method for food procurement and acceptance based on the Internet of Things as described in claim 1, characterized in that, Ambient air temperature, relative humidity, and cargo surface temperature are collected through distributed sensor nodes. A condensation trigger flag is generated based on these parameters, including: The ambient air temperature, ambient relative humidity, and cargo surface temperature are collected synchronously by a temperature and humidity sensor and an infrared surface temperature sensor. The collected data is uploaded to the local data aggregation node via a wireless transmission protocol. Based on the collected ambient air temperature and relative humidity, the ambient thermal and humid state is converted into dew point temperature using the Magnus approximation. A condensation trigger flag is generated based on the difference between the dew point temperature and the surface temperature of the cargo. If the difference between the dew point temperature and the surface temperature of the cargo is greater than zero, the condensation trigger flag is set to one; otherwise, the condensation trigger flag is set to zero.

3. The intelligent comparison method for food procurement and acceptance based on the Internet of Things as described in claim 1, characterized in that, The data gateway reads the on-site prior parameters, and determines the moisture absorption weight deduction based on the condensation trigger flag and the on-site prior parameters, including: The data platform normalizes the relative humidity of the on-site environment and the relative humidity of the environment before delivery; Based on the ambient temperature and the normalized relative humidity of the ambient environment, the first equilibrium humidity was calculated using the Hailwood–Horrobin isotherm. Based on the ambient temperature and the normalized relative humidity of the environment before delivery, the second equilibrium moisture content was calculated using the Hailwood–Horrobin isotherm. The dry basis moisture content increment is obtained based on the difference between the first equilibrium moisture content and the second equilibrium moisture content. The moisture absorption weight deduction is calculated based on the condensation trigger flag: if the condensation trigger flag is one, the product of the dry basis moisture content increment and the dry weight of the paper packaging is used as the moisture absorption weight deduction; otherwise, the moisture absorption weight deduction is set to zero.

4. The intelligent comparison method for food procurement and acceptance based on the Internet of Things as described in claim 1, characterized in that, The deviation is calculated based on the deduction for external condensation mass and the deduction for moisture absorption mass, including: Read the tare weight and net content of the order; Read the final weighing value of the last second in the weighing time series; The comparable net weight is obtained by subtracting the external condensation mass deduction, the moisture absorption mass deduction, and the tare weight from the final weighing value. The deviation is calculated based on the difference between the comparable net weight and the net content of the order.

5. The intelligent comparison method for food procurement and acceptance based on the Internet of Things as described in claim 1, characterized in that, The deviation is judged to be compliant. If the deviation is greater than the preset relative tolerance, it is judged as unqualified; otherwise, it is judged as qualified.

6. An intelligent comparison system for food procurement and acceptance based on the Internet of Things, characterized in that, The system includes: The condensation indicator module is used to collect ambient air temperature, ambient relative humidity and cargo surface temperature through distributed sensing nodes, and generate a condensation trigger indicator based on the ambient air temperature, ambient relative humidity and cargo surface temperature. The weighing sequence module is used to collect the weighing time series through networked weighing devices and determine the initial steady-state value and segmentation points based on the weighing time series; The external condensation quality module is used to calculate the external condensation quality deduction based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series; The moisture absorption quality module is used to read the on-site prior parameters through the data gateway and determine the moisture absorption quality deduction based on the condensation trigger flag and the on-site prior parameters; The deviation determination module is used to calculate the deviation based on the external condensation quality deduction and the moisture absorption quality deduction, and to determine the compliance of the deviation. The process of acquiring weighing time series through networked weighing devices and determining initial steady-state values ​​and segmentation points based on the weighing time series includes: the networked weighing device outputting the weighing time series at a sampling period of one second, wherein the sampling frequency is one hertz, the observation window is two hundred and forty seconds, and the data is transmitted in real time to the edge computing node via an industrial bus; the edge computing node obtains the initial steady-state value and caches it by calculating the median of the weighing time series in the five to ten second interval; the slope sequence of the weighing time series is calculated using the central difference method; within the time interval of thirty seconds to two hundred and forty seconds, the smallest time point that satisfies the condition that the slope sequence value is less than a preset slope threshold and the rate of change of the slope sequence is less than zero is selected, and the smallest time point is determined as the segmentation point of the weighing time series; The step of calculating the external condensation mass deduction based on the condensation trigger flag, the initial steady-state value, and the weighing values ​​at the segment points of the weighing time series includes: reading the weighing values ​​at the segment points of the weighing time series; if the condensation trigger flag is one, then the difference between the weighing value at the segment point and the initial steady-state value is used as the external condensation mass deduction; if the condensation trigger flag is zero, then the external condensation mass deduction is set to zero. The on-site preliminary parameters include: on-site ambient temperature, on-site ambient relative humidity, ambient relative humidity before delivery, and dry weight of paper packaging. The ambient relative humidity before delivery is transmitted back to the data platform from the cold chain monitoring terminal via a wide area network.

Images