Infant nutrition intake analysis and diet intelligent management method

Abstract

The invention relates to the technical field of infant diet management, and discloses an infant nutrition intake analysis and diet intelligent management method. The method comprises the following steps: interactively obtaining physiological characteristic parameters and diet history parameters of an infant individual; determining nutritional requirement type selection, and obtaining a target diet plan; interactively acquiring target structure features of the target diet plan, initializing diet management conditions in combination with nutritional demand selection, and outputting initial management control constraints including a nutrition intake threshold and a maximum allowable deviation rate; taking the initial management control constraint as an optimization condition, optimizing the ingestion preparation speed according to the food ingestion record information and the preset time constraint, and positioning the ingestion control speed; analyzing a nutrition deviation trend of the target diet plan according to the intake control speed, and optimizing diet adjustment control parameters based on an analysis result; and when food intake is managed according to the intake control speed during diet execution, intermittently and automatically adjusting the target diet plan by using the diet adjustment control parameters.

Description

Technical Field

[0001] This invention relates to the field of early childhood diet management technology, specifically to a method for analyzing early childhood nutrient intake and intelligent dietary management. Background Technology

[0002] A balanced and appropriate intake of nutrients is crucial for a child's growth and development, directly impacting the normal development of bodily functions, the building of immunity, and the foundation for future health. Currently, early childhood dietary management largely relies on the subjective experience of parents or caregivers, lacking scientific, systematic, and personalized management methods, making it difficult to meet the individualized nutritional needs of young children. When planning meals for young children, parents often lack comprehensive professional nutritional knowledge and struggle to accurately combine a child's age, weight, and health status to create a meal plan that meets their actual needs. Furthermore, a child's health condition can influence their nutritional requirements; for example, when a child is experiencing a cold or indigestion, their need for certain nutrients may change, making it difficult for parents to accurately assess and adjust the diet accordingly. Regarding the use of dietary history data, current management methods often simply record what the child has eaten, failing to analyze the nutritional composition of the food in depth, let alone accurately match this information with the child's health status to their nutritional needs. This results in meal plans lacking specificity, potentially failing to provide the child with the necessary nutrients and thus impacting their growth and development. Even when some parents attempt to create a meal plan, they often struggle to effectively manage its implementation. Lacking clear nutrient intake thresholds and maximum allowable deviation rates as references, parents cannot determine whether their child's actual nutrient intake is within a reasonable range or whether the intake rate is appropriate. When a child's actual intake deviates from the plan, parents often rely on intuition to make adjustments rather than implementing precise dietary changes based on scientific analysis, significantly diminishing the effectiveness of the meal plan. This haphazard approach to meal management can easily lead to nutritional imbalances in young children, potentially causing malnutrition, obesity, or other health problems, which are detrimental to their healthy growth. Summary of the Invention

[0003] The purpose of this invention is to provide a method for analyzing and intelligently managing the nutritional intake of young children, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a method for analyzing and intelligently managing the nutritional intake of young children, the method comprising: The physiological characteristic parameters and dietary history parameters of an individual child are obtained interactively. The physiological characteristic parameters include age information, weight information and health status information, and the dietary history parameters include food intake record information and nutritional composition information. Nutritional needs are matched based on the health status information and nutritional composition information to obtain a nutritional requirement selection. Then, the target dietary plan is obtained by initializing the dietary plan based on the nutritional requirement selection. The target structural features of the target diet plan are obtained interactively, and the diet management conditions are initialized according to the target structural features and nutritional requirements. The initial management control constraints are output, including the nutritional intake threshold and the maximum allowable deviation rate. Using the initial management and control constraints as optimization conditions, intake pacing optimization is performed based on the food intake record information and preset time constraints to determine the intake control speed. Based on the intake control rate, a nutritional deviation trend analysis of the target diet plan is performed, and dietary adjustment control optimization is performed based on the analysis results to obtain dietary adjustment control parameters; During the implementation of the diet plan, when managing food intake at the intake control rate, the target diet plan is intermittently and automatically adjusted using the diet adjustment control parameters.

[0005] Preferably, nutritional needs are matched based on the health status information and nutritional composition information to obtain nutritional requirement selection. The method includes: Based on the nutritional type, the nutritional composition information is broken down to obtain the distribution characteristics of multiple nutritional contents for various nutritional types. The multiple nutrient content distribution characteristics are updated according to preset health standards to obtain multiple demand-matching distribution characteristics. By integrating the multiple demand matching distribution features, a global nutrient distribution feature is obtained; Based on the global nutrient distribution characteristics, multi-level nutrient demand matching is performed to obtain a multi-level diet plan component, wherein the multi-level diet plan component has multiple nutrient scales, plan duration and food type identifiers; Based on the multiple nutritional scales, a multi-layered diet plan component is sequentially assembled to obtain the nutritional requirement selection.

[0006] Preferably, the method involves interactively obtaining the target structural features of the target diet plan, initializing diet management conditions based on the target structural features and nutritional requirements, and outputting initial management control constraints. The method includes: Interactively obtain multiple component attribute information of the multi-layer diet plan component; The intake capacity is calculated based on the target structural features to obtain the effective intake capacity. Based on the attribute information of the multiple components and the effective intake capacity, network data is called to obtain multiple diet simulation models. Then, by fusing the multiple diet simulation models, a diet plan model is obtained. After retrieving local data from the diet plan to obtain the standard intake rate, the diet plan model is simulated using the standard intake rate to output a structural deviation threshold. The diet plan model is then simulated using the standard intake rate to output multiple individual deviation thresholds. The nutrient intake threshold is obtained by solving the intersection of the structural deviation threshold and multiple individual deviation thresholds. The maximum value of the nutrient intake threshold is used as a constraint to perform nutrient dynamics simulation of the diet plan model, and the maximum allowable deviation rate is obtained by calling the diet plan model based on the simulation results.

[0007] Preferably, the initial management control constraints are used as optimization conditions, and intake pacing optimization is performed based on the food intake record information and preset time constraints to determine the intake control speed. The method includes: Pre-constructed intake association prediction model; The reciprocal of the standard intake rate is used as the speed change scale to update the standard intake rate data, thereby obtaining the first backup intake rate; Input the first backup intake rate into the intake association prediction model to obtain the first deviation rate time limit, the first nutrient intake time limit, and the first predicted intake efficiency. The first predicted intake time is calculated using the food intake record information and the first predicted intake efficiency. If the first predicted intake time meets the preset time constraint, the first backup intake rate is retained, and the first difference between the first deviation rate time limit and the first nutrient intake time limit is used as the first control reliability coefficient of the first backup intake rate. Similarly, by continuously updating the standard intake rate and evaluating the update results using the intake association prediction model and preset time constraints, multiple alternative intake rates and multiple control reliability coefficients that meet the preset update frequency are obtained. The plurality of control reliability coefficients are serialized, and the intake control rate is located from the plurality of backup intake rates based on the sorting results.

[0008] Preferably, a pre-constructed intake association prediction model is used, the method comprising: Using the nutritional requirement selection and nutritional composition information as constraints for historical data retrieval, historical intake-related data is retrieved, wherein the historical intake-related data includes multiple sample intake rates, multiple sample intake efficiencies, multiple deviation rate change rates, and multiple nutrient intake change rates. Multiple nutrient intake change time limits are calculated based on the nutrient intake threshold and multiple nutrient intake change rates. Multiple deviation rate change time limits are calculated based on the maximum permissible deviation rate and the multiple deviation rate change rates; The intake rates of multiple samples, intake efficiencies of multiple samples, nutrient intake time limits, and deviation rate change time limits are used as training parameters to train the intake association prediction model based on a prediction neural network.

[0009] Preferably, the method involves performing a nutritional deviation trend analysis of the target diet plan based on the intake control rate, and performing dietary adjustment control optimization based on the analysis results to obtain dietary adjustment control parameters. The directional data of the intake control speed is retrieved to obtain the control deviation rate time limit and the control nutrient intake time limit; The smaller value is selected as the adjustment cycle by comparing the control deviation rate time limit and the control nutrient intake time limit; A fitness weight configuration is predefined, and a fitness evaluation function is constructed based on the fitness weight configuration, wherein the fitness weight configuration includes energy consumption weight, nutrient loss weight and time consumption weight; The initial management control constraints are used as data retrieval constraints to filter and retrieve historical diet adjustment records, thereby obtaining multiple sample diet adjustment records. Each of the multiple sample diet adjustment records includes sample control parameters and sample adjustment consumption. The fitness evaluation function is used to evaluate the fitness of the dietary adjustment records of the multiple samples, and the initial control parameters are located based on the evaluation results. The initial control parameters consist of food type flow rate, food type adjustment duration, nutrient supplementation flow rate, nutrient supplementation adjustment duration, and overall adjustment duration. The adjustment cycle is added to the initial control parameters to obtain the dietary adjustment control parameters.

[0010] Preferably, the sample control parameters consist of sample food type flow rate, sample food type adjustment time, sample nutrient supplementation flow rate, sample nutrient supplementation adjustment time, and sample comprehensive adjustment time.

[0011] Preferably, the sample adjustment consumption consists of sample energy consumption, sample nutrient loss, and sample time consumption.

[0012] Preferably, when managing food intake at the intake control rate during diet execution, the target diet plan is intermittently automatically adjusted using the diet adjustment control parameters, the method comprising: The dietary execution process is divided into multiple discrete time periods, and the food type adjustment duration in the dietary adjustment control parameters is evenly distributed among the multiple discrete time periods. After the previous discrete time period ends, determine whether the nutritional intake of the target diet plan during that discrete time period has met expectations; The timing of food type adjustments is adaptively modified for each subsequent discrete time period until the entire diet execution process is automatically adjusted.

[0013] Preferably, the method for evaluating the nutritional intake of the target diet plan includes: Obtain the changes in nutritional levels of the individual children before and after the implementation of the diet; The actual nutrient intake is obtained based on the changes in the nutrient level. The actual nutrient intake is analyzed in relation to the theoretical nutrient requirements to obtain the control weight of the dietary adjustment control parameters; The nutritional intake of the target diet plan is evaluated based on the control authority.

[0014] Compared with the prior art, the beneficial effects of the present invention are: The method for analyzing and intelligently managing nutrient intake in young children provided by this invention offers a comprehensive and precise solution for early childhood diet management through the scientific design and collaborative operation of multiple stages. In the stage of acquiring child information, this method collects individual physiological characteristic parameters and dietary history parameters of each child through interactive means. This includes physiological information closely related to nutritional needs, such as age, weight, and health status, as well as dietary information such as food intake records and nutritional composition. This comprehensive understanding of each child's individual situation provides rich and accurate basic information for subsequent nutritional needs analysis and diet planning, avoiding dietary management deviations caused by incomplete or inaccurate information. In the stage of matching nutritional needs and formulating dietary plans, the method matches nutritional needs based on the child's health status and nutritional composition information to ensure that the selected nutritional needs are in line with the child's current physical condition and nutritional requirements. Then, the dietary plan is initialized and configured based on the selected nutritional needs, so that the generated target dietary plan is highly targeted and can accurately meet the child's nutritional needs at a specific stage. Unlike the traditional method of formulating dietary plans based on experience, this method effectively avoids the problem of unreasonable diet caused by misjudgment of nutritional needs. In the initial stage of dietary management, by acquiring the target structural characteristics of the target dietary plan and combining them with nutritional needs to determine initial management and control constraints, the nutritional intake threshold and maximum allowable deviation rate are clarified, providing clear and quantifiable standards for dietary management. Parents or caregivers can intuitively judge whether the child's nutritional intake is within a reasonable range based on these standards, avoiding the previous problem of ineffective control of dietary intake due to the lack of clear standards, making dietary management more scientific and operable.

[0015] Regarding the optimization of intake pacing, the initial management and control constraints are used as optimization conditions. Combined with food intake record information and preset time constraints, the intake control speed is determined, which can reasonably adjust the rhythm of nutrient intake according to the child's actual intake and time requirements. This helps to avoid children consuming too much nutrition in a short period of time, which would increase the burden on their bodies, or that the intake speed is too slow, which would lead to untimely nutrient supply. This makes the child's nutrient intake process more stable and orderly, in line with the child's digestive and absorptive capacity and the body's metabolic rhythm.

[0016] In the stages of nutritional deviation trend analysis and dietary adjustment control optimization, the nutritional deviation trend of the target dietary plan is analyzed based on the intake control rate. Based on the analysis results, dietary adjustment control parameters are obtained. This allows for the timely detection of potential nutritional deviations during the implementation of the dietary plan, the prediction of deviation trends, and the development of scientific adjustment plans. This proactive deviation early warning and adjustment strategy effectively avoids the adverse effects of widening deviations on children's nutritional intake, ensuring that children continuously receive balanced nutrition. Automatic adjustments are made intermittently during the implementation of dietary plans. Dietary adjustment control parameters are used to dynamically adjust the target dietary plan, flexibly optimizing it based on the child's real-time intake and changes in physical condition. Whether there are slight fluctuations in the child's health or deviations between actual food intake and the plan, automatic adjustments can be made promptly without relying on frequent manual intervention. This reduces the management difficulty and workload for parents or caregivers, while ensuring that the dietary plan always aligns with the child's actual needs. This provides a stable and reliable nutritional guarantee for the child's healthy growth, helping them obtain sufficient and balanced nutritional support during critical growth and development stages. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the working principle of the method for analyzing and intelligently managing the nutritional intake of young children as described in this invention. Figure 2 Flowchart of methods for selecting and obtaining nutritional needs; Figure 3 Flowchart of the method for outputting initial management control constraints; Figure 4 A flowchart illustrating the method for pre-building an intake association prediction model. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides a method for analyzing and intelligently managing the nutritional intake of young children, the method comprising: The system collects key information about each child through an interactive interface. This information is divided into two main categories: first, physiological characteristics, including the child's age, current weight, and health status; and second, dietary history parameters, covering detailed records of food intake over a past period and nutritional composition information derived from analysis or input of these records. After obtaining this basic data, the system enters its core processing phase. It doesn't simply apply general nutritional standards, but rather matches targeted nutritional needs based on the child's specific health status and the nutritional composition characteristics of their historical diet. This process outputs a structured nutritional needs selection, defining the key nutritional goals and their quantitative requirements for the child during a specific period. Based on this selection, the system initializes a preliminary, personalized target diet plan, which includes suggested food types, portion sizes, and a preliminary meal schedule.

[0020] The system needs to set management rules for this target diet plan. It again obtains the target structural characteristics of the plan through interaction, and combines these structural characteristics with the aforementioned nutritional requirement selection. The system initializes specific dietary management conditions, with its core output being the initial management control constraints. These constraints include two key numerical indicators: nutrient intake threshold and maximum allowable deviation rate. These two constraints set boundaries for subsequent refined management. To better reflect the actual eating abilities and habits of young children, the system introduces the concept of intake pacing optimization. Using the children's food intake records and preset time constraints, it uses the aforementioned initial management control constraints as the optimization target to calculate the optimal intake control speed. This speed does not refer to physical eating speed, but rather to the progress of nutrient intake that should be achieved per unit of time, guiding the management of the eating rhythm.

[0021] The system doesn't remain static after the plan is set; it performs a nutritional deviation trend analysis on the target diet plan based on the calculated intake control rate. Based on this analysis, the system proactively optimizes dietary adjustments, calculating backup control parameters. These parameters define how to dynamically adjust subsequent food types, portions, or supplementation strategies, as well as the pace and magnitude of adjustments, when potential deviations are detected. During the actual execution of the diet plan, the system guides or manages the child's food intake according to the optimized intake control rate. Simultaneously, the system continuously monitors actual intake. Once monitoring data indicates that the nutritional intake trajectory begins to deviate from the planned trajectory, the system automatically triggers a preset adjustment mechanism, using the previously optimized dietary adjustment parameters to intermittently and automatically adjust the target diet plan. This adjustment is dynamic and timely, aiming to bring nutritional intake back on track and ensure that the child ultimately and consistently achieves their nutritional goals. The entire process forms a closed-loop intelligent management system of "planning-monitoring-prediction-adjustment."

[0022] Example 1: See Figure 2The physiological characteristics of each child are input into the system by the guardian using a dedicated application or connected health monitoring devices. Age information is accurate to the month, weight information is updated daily or weekly in kilograms, and health status information is obtained through structured questionnaires or by connecting to medical records, covering key indicators such as allergen lists, current disease diagnoses, recent medication history, and self-assessment levels of digestive function. Dietary history parameters rely on the guardian continuously entering details of each meal each day, including food names, estimated or measured weight, and start and end times of eating. The system's built-in or networked food nutrition database automatically analyzes the content of dozens of nutrients, such as protein, fat, carbohydrates, vitamin A, vitamin C, calcium, iron, and zinc, corresponding to each intake record, forming time-series nutritional composition information. When the system performs nutrient type breakdown, the above historical nutritional data is aggregated according to a preset nutrient classification tree. Macronutrients such as protein, fat, and carbohydrates are analyzed independently, while micronutrients are grouped by vitamin and mineral groups. Regarding iron, the calculated average daily intake was 7.8 mg, with a standard deviation of 2.1 mg, indicating unstable intake. Iron intake at dinner accounted for only 15%, significantly lower than the 45% at lunch. Compared to the recommended intake of 12 mg for anemia in children of the same age and sex, the difference was -35%. The average daily intake of vitamin C was 35 mg, with a standard deviation of 15 mg, a difference of -30% compared to the recommended intake of 50 mg. Furthermore, time-series analysis of vitamin C and iron intake showed a low co-occurrence rate of the two at the same meal. The "iron deficiency anemia" label in the health status information triggered a dynamic adjustment module for preset health standards. This module invoked the iron requirement correction coefficient for anemic children in clinical nutrition guidelines, increasing the recommended intake from the baseline of 12 mg to 15 mg, and simultaneously raising the recommended intake of vitamin C to 60 mg to promote iron absorption. Based on this, the system updated the distribution characteristics of iron nutritional content, with the difference from the revised standard changing from -35% to -48%, and the deficiency level upgraded from "moderate" to "severe"; the difference in vitamin C changed from -30% to -42%. Although the daily intake of other nutrients such as protein met the standard, the analysis found that the proportion of plant protein was too high, and the intake of animal protein, especially red meat from which heme iron is derived, was insufficient. This information was incorporated into the demand matching distribution characteristics.

[0023] The system employs a weighted algorithm based on nutritional physiology during the fusion phase. Iron, due to its severe deficiency and crucial role in anemia treatment, is given the highest weight. Vitamin C, due to its positive correlation with iron absorption, receives the second highest weight. Protein, while meeting the overall target, has a poor source structure, resulting in a third highest weight. Calcium, zinc, and other nutrients that meet the target have the lowest weight. The global nutrient distribution characteristics output by the fusion system highlight the core contradictions: severe iron deficiency with a poor source structure, moderate vitamin C deficiency with insufficient synergy with iron, and protein intake meeting the target but with room for improvement in bioavailability. Multi-level nutrient requirement matching identifies the core objectives: a daily iron intake of 15 mg and vitamin C of 60 mg are required for the next 7 days, and animal protein should account for more than 40% of total protein. The second level of matching plans nutrient allocation: breakfast focuses on providing highly bioavailable protein and vitamin C; lunch must include foods rich in heme iron; dinner supplements non-heme iron and is paired with vitamin C-rich fruits; and snacks include nuts and dairy products. The third level of food carrier matching: After excluding seafood based on allergy information, the preferred iron carrier for lunch is ground beef instead of pork liver, the preferred vitamin C carrier is strawberry instead of citrus, and the preferred protein carriers are eggs and ground chicken. The final nutritional needs selection consists of three layers: the first layer is the daily nutritional goal; the second layer is the distribution of nutrients across meals; and the third layer is the food implementation checklist. This selection clearly states that the diet for the coming week should focus on iron supplementation, achieving the goal by optimizing the distribution of nutrients across meals and food choices.

[0024] When the system breaks down nutrients into categories, its internal processing logic involves grouping and organizing the input nutrient composition information according to a scientifically recognized nutrient classification system. This grouping is not arbitrary but based on the physiological functions and chemical properties of nutrients in the human body, as well as the conventions of nutritional research. For each nutrient category, the system calculates multiple indicators reflecting its intake status. These indicators include the average intake of the nutrient over a period of time, which shows the usual intake level; the maximum and minimum intake values, which reflect the range of fluctuation in intake; the standard deviation of intake, which reflects the stability of intake; the percentage of the nutrient intake in different meals, such as what percentage of the total daily protein is consumed at breakfast; and the ratio of the actual intake to the recommended intake, which directly indicates whether there is a deficiency or excess.

[0025] These calculated indicators constitute the nutrient content distribution characteristics, collectively depicting a child's intake profile of various nutrients. When the system incorporates health status information, it adjusts the preset nutritional standards based on this information. These preset health standards are typically derived from authoritative dietary guidelines or clinical nutrition recommendations, but the system does not mechanically apply them; instead, it makes personalized adjustments based on the child's specific health issues. The system reassesses historical intake data using the adjusted standards, generating a demand-matching distribution characteristic. This characteristic more accurately reflects the child's actual nutrient needs being met under their current health condition. The system needs to integrate the needs of dozens of nutrients to form a global nutrient distribution characteristic. The fusion process considers the interactions and priorities between nutrients, assigning higher priority to nutrients directly related to the child's current health problems. The fusion result highlights the most pressing nutritional issues and key improvement goals, providing direction for subsequent dietary planning.

[0026] Multi-level nutritional needs matching is a process of refining goals in layers. The first level focuses on core nutritional gaps, determining the quantitative targets for key nutrients that must be achieved within the planned period. These target values ​​are derived from the most severe deficiencies identified in the overall nutritional distribution characteristics and their gaps with the revised standards. The second level addresses nutrient allocation, i.e., how to rationally distribute these overall goals across different meals. This requires considering the child's digestive patterns, eating habits, and nutrient absorption characteristics. The third level focuses on specific food choices, i.e., what foods to use to achieve the aforementioned nutritional goals.

[0027] This requires balancing nutritional needs, children's taste preferences, food allergies and contraindications, ingredient availability, and the rationality of food combinations. The matching results output multi-layered meal plan components, each containing specific nutritional goals, applicable time ranges, and recommended food types. The system sequentially assembles these components based on the priority and temporal logic of the nutritional goals. The core nutritional goal component is placed at the top level, meal allocation components are arranged in chronological order, and food execution components are associated with the corresponding meals. The assembly process ensures that the goals of each component are consistent and the logic is coherent, avoiding conflicts. The assembled nutritional needs selection is a clearly structured, well-defined, and actionable blueprint for nutritional improvement, laying the foundation for generating specific recipes.

[0028] Example 2: See Figure 3After the target diet plan is generated, the system presents the initial framework of the plan to the guardian through a graphical interface. This includes three meals a day (breakfast, lunch, and dinner) and two snacks in the morning and afternoon. The system also suggests the main food categories and approximate proportions for each meal, such as 40% staple food, 30% protein, and 30% vegetables for lunch. Based on the child's daily routine and eating habits, the guardian sets or confirms key parameters on the interface: desired breakfast duration of 20 minutes, lunch 30 minutes, dinner 25 minutes, and snacks 10 minutes each; and provides feedback that the child chews leafy vegetables slowly but accepts soft fruits quickly. The system captures these inputs to form target structural features, including the meal sequence, maximum duration for each meal, baseline proportions of main food categories within each meal, and eating speed markings for specific food categories. Based on these structural features and the child's weight of 14.5 kg and age of 36 months, the system retrieves historical eating speed data, combined with the maximum meal duration and food proportion constraints, to calculate the effective intake volume. Taking lunch as an example: the maximum duration is 30 minutes, and the food ratio is staple food: protein: vegetables = 4:3:3. According to historical data, a toddler can consume approximately 180 grams of staple food, 90 grams of protein, and 90 grams of vegetables within 30 minutes.

[0029] By summarizing similar calculations for each meal, the effective daily intake capacity is determined to be: no less than 130 grams of carbohydrates, no less than 35 grams of protein, and no less than 8 grams of dietary fiber. The system simultaneously analyzes the attributes of the multi-layered diet plan components in the nutritional needs selection, extracting key information: the first-layer component requires 15 mg of iron and 60 mg of vitamin C; the second-layer component specifies that lunch should provide 6 mg of iron and 15 mg of vitamin C; the third-layer component specifies that lunch should include 30 grams of ground beef and 50 grams of strawberries. These component attribute information, together with the calculated effective intake capacity, constitute the online query conditions. The system sends a request to the cloud-based nutritional behavior database, with query conditions including: age 3 years ± 6 months, weight 14 ± 2 kg, anemia marker, lunch containing red meat and berries, meal time 30 minutes, target iron intake 6 mg / meal, and vitamin C 15 mg / meal. The database returns three highly matching dietary simulation models: Model A is based on statistical regression analysis of similar children's eating records; Model B integrates food digestion dynamics parameters; and Model C applies machine learning to predict the eating process over time. The system employs a weighted fusion algorithm, assigning weights based on the matching degree between the model and the current characteristics of the children, to generate a fused diet plan model. This model can simulate the impact of different eating rhythms on the dynamic accumulation of nutrient intake under a specified meal structure.

[0030] The system retrieves a standard intake rate calculated based on the child's basal metabolic rate and activity level from local storage. This standard rate drives the diet plan model for nutrient kinetic simulation, simulating the nutrient accumulation curve over time under an ideal eating rhythm. The simulation revealed that due to the slight inhibition of iron absorption by oxalic acid in the lunch vegetables, the actual iron absorption rate was approximately 85%, resulting in a simulated iron intake of 5.1 mg at the end of 30 minutes, lower than the target value of 6 mg. The system outputs a structural deviation threshold, suggesting that the lower limit of iron intake be adjusted to 5.1 mg. Simultaneously, the same standard rate was used to drive individual models A, B, and C for simulation. Model A outputs the individual deviation threshold iron [4.8 mg, 6.5 mg], Model B outputs [5.0 mg, 6.0 mg], and Model C outputs [5.2 mg, 5.8 mg]. The system takes the upper limit of the intersection between the structural deviation threshold lower limit of 5.1 mg and the individual model threshold lower limit set {4.8, 5.0, 5.2}, i.e., 5.1 mg; and takes the lower limit of the intersection of the upper limit min(6.0, 5.8) = 5.8 mg. The final iron intake threshold for lunch was determined to be [5.1mg, 5.8mg]. To calculate the maximum allowable deviation rate, the system was run again to simulate the diet plan model with the upper limit of the intake threshold of 5.8mg as a constraint. The simulation simulated a slightly adverse scenario and observed the maximum negative deviation of the actual intake from the original target value of 6mg, provided that it does not exceed 5.8mg. Multiple simulation results showed that the maximum negative deviation was stable at -8.3%. Therefore, the maximum allowable deviation rate for iron in this meal was set at -8.3%. This value means that the system allows the actual intake to be 8.3% lower than the planned target value, and it is considered safe as long as it is not lower than 5.1mg.

[0031] Nutrient intake thresholds and the maximum permissible deviation rate together constitute the initial management control constraints, providing a quantitative boundary for subsequent speed optimization. The threshold defines the safe range of nutrient intake, while the deviation rate defines the permissible fluctuation range of actual intake relative to the planned target value. During calculations, the system relies on simulation models to predict nutrient absorption and utilization efficiency and the potential impact of various interfering factors. Nutrient kinetic simulation refers to using computer models to simulate the dynamic process of food breakdown in the digestive tract and nutrient absorption into the bloodstream. This simulation considers factors such as food component interactions and individual differences in digestive capacity. The deviation rate change time limit refers to the time required, predicted by the model, for the actual intake to deviate from the planned target value to reach the permissible deviation limit at a specific eating speed. This parameter is used to assess the resilience of the current eating rhythm to interference.

[0032] Calculating effective intake capacity essentially involves estimating the maximum reasonable amount of food a child can physiologically consume within a given timeframe and food structure, and converting this into quantifiable nutrient values. This process requires considering the child's chewing and swallowing abilities, stomach capacity comfort threshold, and historical eating behavior data. Dietary simulation models are mathematical tools extracted from numerous similar cases to predict the final nutrient intake under specific dietary plans and eating behaviors. Model fusion techniques are used to improve the robustness of predictions and avoid the limitations of single models. Determining nutrient intake thresholds is a multi-model consensus process. By comparing the prediction results of the fused model with the original individual models, the overlapping safe range is taken as the final management standard, enhancing the reliability of the thresholds. Calculating the maximum permissible deviation rate, while ensuring intake does not fall below the safe lower limit, explores the tolerable short-term fluctuation range during plan execution, providing necessary flexibility for management. The time frame for calculating deviation rate changes is as follows:

[0033] in: This represents the time required to reach the maximum permissible deviation rate. Represents the absolute value of the maximum permissible deviation rate. This represents the rate of change of the bias rate predicted by the model. This formula quantifies the length of the time window during which the system can tolerate the accumulation of bias at a specific feeding rate.

[0034] Example 3: See Figure 4 When the system pre-constructs the intake association prediction model, taking the lunch scenario of anemic children as an example, the nutritional requirement selection requires an iron intake of 6 mg, and the nutritional composition information shows that the historical average iron intake at lunch is 4.2 mg. The system calls historical intake association data of similar anemic children: including the intake speed of 20 groups of samples, the corresponding sample intake efficiency, the rate of change of deviation rate, and the rate of change of nutritional intake. Based on the lower limit of the iron intake threshold of 5.1 mg determined in Example 2 and the rate of change of nutritional intake, the time limit of nutritional intake change is calculated; based on the maximum allowable deviation rate of -8.3% and the rate of change of deviation rate, the time limit of deviation rate change is calculated. The four types of data (intake speed, intake efficiency, nutritional time limit, and deviation time limit) are input into a three-layer neural network. The activation function of the hidden layer is ReLU, and the output layer is linear regression. After 500 rounds of training, the prediction model is obtained. This model can take any speed value as input and output the predicted intake efficiency, nutritional intake time limit, and deviation rate change time limit.

[0035] The system performs intake pacing optimization, with an initial standard intake rate of 0.02 mg iron / min. Using the reciprocal of the rate over 50 minutes / mg as a scale of variation, and setting an adjustment coefficient K=1, a first backup intake rate of 0.0198 mg / min is generated. This rate is input into the prediction model, outputting a first deviation rate time limit of 18.2 minutes, a first nutrient intake time limit of 257 minutes, and a first predicted intake efficiency of 88%. Based on the historical average amount of food consumed at children's lunches (45 grams), the first predicted intake time is calculated as: planned iron intake 6 mg / (0.0198 mg / min * 88%) ≈ 34.4 minutes. Comparing this to the preset time constraint, 34.4 minutes > 30 minutes, therefore this rate is discarded. The coefficient K is updated to 0.5, generating a second backup rate of 0.0199 mg / min, but the predicted intake time of 31.7 minutes still exceeds the limit. K is further reduced to 0.2, resulting in a rate of 0.01996 mg / min, with a predicted time of 30.1 minutes, close to the threshold. At this point, using the K-increment direction: a third reserve rate of 0.02004 mg / min is generated. Inputting this into the model yields a third deviation rate time limit of 15.8 minutes, a third nutrient intake time limit of 252 minutes, and a third predicted intake efficiency of 92%. The prediction time is 6 / (0.02004*92%)≈29.8 minutes≤30 minutes, satisfying the constraints. The control reliability coefficient is calculated as |15.8-252|=236.2. Similarly, a fourth rate of 0.02008 mg / min is generated, with a prediction time of 29.5 minutes and a reliability coefficient of 230.5; the fifth rate is 0.02012 mg / min, with a prediction time of 29.2 minutes and a reliability coefficient of 225.1. The reliability coefficients are serialized: third rate 236.2, fourth rate 230.5, fifth rate 225.1. After sorting in descending order, the third rate of 0.02004 mg / min with the highest reliability coefficient is selected as the final intake control rate.

[0036] The essence of the intake association prediction model is to establish a quantitative relationship between eating speed and nutrient intake results through machine learning. The historical data required for training this model comes from a large number of actual eating records of similar young children, including completion status and the rate of accumulation of nutrient deviation under different eating rhythms. The rate of change in nutrient intake refers to the amount of a specific nutrient absorbed per unit time, while the rate of change in deviation rate reflects the rate at which actual intake deviates from the planned target. The nutrient intake change time limit represents the time required to reach the minimum safe intake at the current speed, and the deviation rate change time limit represents the time required to reach the maximum allowable deviation. These two time limit parameters together characterize the safe buffer space of the current eating rhythm. By learning these complex relationships, the predictive neural network can predict the nutrient absorption efficiency, the time to reach the target, and the risk of accumulated deviation at any given speed. In the pacing optimization process, the system starts with the standard speed and gradually generates candidate speeds according to mathematically defined step sizes. Each candidate speed needs to pass two tests: first, a time feasibility test, that is, whether the predicted completion time meets the upper limit of meal time set by the guardian; second, a reliability test, that is, assessing the degree of difference between the time to reach the nutrient target and the time of accumulated deviation at this speed. A greater difference indicates a stronger system fault tolerance; even if a child's brief distraction causes a speed fluctuation, it is less likely to trigger a nutrition alarm. The system employs a comprehensive scoring mechanism during iteration to quantify the applicability of candidate speeds, mathematically expressed as:

[0037] in: The overall score representing speed. It controls the reliability coefficient. It is the ingestion completion time predicted by the model. It is a preset time constraint value. This is the time deviation penalty coefficient. While retaining the core weight of the reliability coefficient, this scoring function applies an exponentially decaying penalty to speeds approaching the time constraint boundary, ensuring that the selected speed possesses both stability and timeliness.

[0038] The iterative optimization process must meet the preset update frequency requirement. The system explores bidirectionally around the standard speed: decreasing the speed may prolong the feeding time but increase reliability, while increasing the speed may shorten the time but decrease reliability. Each test speed requires a complete process: model prediction, time constraint verification, reliability coefficient calculation, and comprehensive scoring. For the case of anemic infant lunches, when the test speed increased from 0.02004 mg / min to 0.02008 mg / min, the predicted completion time decreased from 29.8 minutes to 29.5 minutes, and the reliability coefficient decreased from 236.2 to 230.5. Substituting into the scoring formula (assuming...)... , Third speed rating Fourth speed rating Although the fourth speed is shorter, its score is lower than the third speed, so maintaining 0.02004 mg / min is the optimal choice. This speed means that the system recommends absorbing 0.02004 mg of iron per minute during lunch, equivalent to completing 1% of the planned iron intake every 2.5 minutes. If the child's actual eating speed consistently falls below this value, the model predicts that the accumulated deviation may exceed the limit after 15.8 minutes, or that iron intake may reach the safe lower limit of 5.1 mg after 252 minutes. Since 252 minutes is far beyond the upper limit of meal time, the deviation limit of 15.8 minutes becomes a critical constraint, prompting the system to detect insufficient speed and initiate intervention before this time point. The essence of positioning the intake control speed is to find the optimal balance between time feasibility and nutritional stability; this speed will serve as the baseline for subsequent real-time management.

[0039] Example 4: The system performs nutritional deviation trend analysis based on the location-based intake control rate of 0.02004 mg iron / min. After inputting this rate into a pre-constructed intake association prediction model, the model outputs a control deviation rate time limit of 15.8 minutes and a control nutrient intake time limit of 252 minutes. By comparing these two time limit values, the system selects the smaller value of 15.8 minutes as the adjustment period. This value indicates that, while maintaining the current eating rhythm, the actual iron intake deviation rate is expected to accumulate to the maximum allowable value of -8.3% at the 15.8-minute mark. The system predefines fitness weight configurations, where energy consumption is weighted at 0.4, nutrient loss at 0.5, and time consumption at 0.1, reflecting that nutrient preservation is prioritized over time efficiency. Based on this configuration, a fitness evaluation function is constructed: the score for each adjustment plan equals energy consumption multiplied by 0.4, nutrient loss multiplied by 0.5, and time consumption multiplied by 0.1. Using the initial management control constraints as screening conditions, the system retrieves five sample dietary adjustment records from the historical database that meet these constraints. Each set of records contains two parts of data: sample control parameters and sample adjustment consumption. See Table 1 for details.

[0040] Table 1: Sample dataset of historical dietary adjustment records.

[0041]

[0042] The system uses a comprehensive scoring model to quantify the applicability of each sample, and its expression is:

[0043] in: This indicates the final overall score. To control the reliability coefficient, It predicts the time it will take to finish ingesting the food. It is a preset time constraint. It is the reliability enhancement index. This is the time deviation penalty index. The model amplifies the core role of the reliability coefficient through exponential calculation, while simultaneously imposing a nonlinear penalty on time deviation. Sample 1 scored 60.3, Sample 3 scored 58.5, and Sample 5 scored 68.1. After comparison, Sample 4 scored the highest, therefore its control parameters were selected as the initial control parameters.

[0044] The percentage adjustment of food type flow rate reflects the extent to which the eating speed of a specific food is changed; a positive value indicates accelerated eating. The nutrient supplementation flow rate refers to the intake of nutrients per unit time through additional supplements. The energy expenditure index is a quantitative indicator of the caregiver's workload; a higher index value indicates greater effort required. Nutrient loss is the reduction in the intake of core nutrients due to the adjustment strategy. A higher control reliability coefficient indicates stronger resistance to interference with the current eating rhythm. The comprehensive scoring model... The index controls the degree of amplification of the reliability coefficient; the larger the value, the more emphasis is placed on reliability. The index controls the penalty intensity for time deviations; a higher value indicates greater sensitivity to timeouts. Operational parameters in historical sample data are derived from real intervention records. The system uses quantitative evaluation to select the template with the lowest overall cost. The final adjustment cycle of 15.8 minutes is integrated into the initial control parameters to form a dietary adjustment control protocol: within 7 minutes of deviation triggering, the food type flow rate is increased by 6%, and iron is supplemented simultaneously at a rate of 0.015 mg / min. The first 4 minutes implement both measures, while the subsequent 3 minutes maintain only the increased flow rate. This protocol is converted into specific operational instructions through a control interface. The final output dietary adjustment control parameters integrate temporal control logic and physical operational specifications, providing an executable framework for automatic adjustment.

[0045] Example 5: The system divides the 30-minute lunch execution process for anemic children into six discrete time slots, each 5 minutes long, labeled TS1 to TS6. Based on the total food type adjustment time defined in the dietary adjustment control parameters, a 7-minute adjustment period is allocated evenly across the six time slots, with each slot receiving approximately 1.167 minutes of adjustable time. The 4-minute nutritional supplement adjustment period is allocated to the first four time slots, with 1 minute allocated to each slot. After time slot TS1 ends, the system obtains the actual food intake data through the weighing sensor built into the smart tableware: 45 grams of staple food, 25 grams of ground beef, and 15 grams of vegetables. Based on the food nutrition database, the actual iron intake is 3.0 mg, while the planned iron intake for this time slot is 3.6 mg. The system determines that the actual value is lower than the planned lower limit of 3.24 mg, triggering the adjustment mechanism. According to the preset rules, the duration of food type adjustment for the subsequent TS2 time period was adaptively modified: the originally allocated 1.167 minutes was increased to 1.5 minutes, and the flow rate of vegetable foods was increased from the parameter setting of 6% to 8%. During the execution of the TS2 time period, the guardian guided the child to eat spinach faster according to the system prompts, and the actual vegetable intake reached 18 grams, with an iron intake of 1.1 mg. At the end of TS2, the cumulative iron intake was 4.1 mg, still lower than the planned cumulative value of 4.68 mg. The system then modified the adjustment parameters for TS3 again: the food type adjustment duration was increased to 1.7 minutes, the vegetable flow rate was increased to 10%, and the nutritional supplementation mechanism was activated: during the 1-minute nutritional supplementation period allocated in TS3, iron supplements were fed at a rate of 0.015 mg per minute. After the end of TS3, monitoring showed an iron supplementation of 0.015 mg, vegetable iron intake of 1.25 mg, and a cumulative iron intake of 5.35 mg, which is close to 92.8% of the planned value of 5.76 mg. The system determined that the progress had been restored, and TS4 was executed with the original parameters. For TS5-TS6, since the cumulative intake has approached the target value, the system shuts down active adjustment and only performs monitoring. The entire execution process achieves gradual deviation correction through time-segmented evaluation and parameter modification.

[0046] The acquisition of nutritional level changes requires analysis of medical test data and dietary records. 24 hours before lunch, the system retrieved the child's latest hemoglobin level from the associated health record, which was 105 g / L. 24 hours after lunch, the guardian submitted a new rapid hemoglobin test result of 107 g / L. The system calculated the hemoglobin change as +2 g / L. Combining the lunch iron intake record and daily dietary log, the estimated increase in hemoglobin contributed by this meal was approximately 0.8 g / L. Actual nutritional intake was calculated by summing the weight of food recorded by sensors at various time points and converting it using a nutritional database: total iron intake 6.2 mg, vitamin C intake 32 mg. The theoretical nutritional requirements are derived from the lunch targets defined in the nutritional requirements selection: iron 6.0 mg, vitamin C 15 mg. The system performed a structured comparison between actual intake and theoretical requirements: iron achievement rate 103.3%, vitamin C achievement rate 213.3%. The control weight was calculated using a multi-dimensional weighted algorithm: time dimension weight 0.3, nutritional deviation dimension weight 0.6, and operational intervention dimension weight 0.1. The calculated scores were 0.97 for the time dimension, 1.05 for the nutritional dimension, and 0.9 for the operational dimension. The final control weight was calculated as 0.3 × 0.97 + 0.6 × 1.05 + 0.1 × 0.9 = 1.021. Based on this weight value, the system generated a nutritional intake evaluation report: iron intake reached an excellent level, vitamin C intake posed a risk of excessive intake, and the overall control effect was rated B+. The report recommended reducing the proportion of high-vitamin C fruits in future plans while maintaining the current iron supplementation strategy.

[0047] The division of discrete time periods essentially transforms the continuous eating process into manageable monitoring units. Each time period is equivalent to a monitoring window, and the system collects data and adjusts strategies when the window is closed. The percentage increase in food type flow rate indicates the degree of enhancement in the eating speed of a specific food category. The physical realization of nutrient supplementation flow rate relies on standardized equipment. The clinical significance of hemoglobin changes lies in reflecting the efficiency of iron conversion into hematopoietic function; a change of +2 g / L indicates that nutritional intervention produces a physiological response. Control weight is an indicator of the execution accuracy of the management system; a value greater than 1 indicates that the actual control effect exceeds expectations, while a value less than 1 indicates inadequate management. The basis for determining the risk of vitamin C overdose is the maximum tolerable intake (UL). The UL for 4-year-old children is 400 mg / day. 32 mg per meal is still within the safe range, but long-term trends need to be monitored. The adaptive modification rules adopt a step-by-step enhancement logic: the first deviation triggers the basic adjustment intensity, and subsequent deviations gradually increase the intervention intensity until the planned trajectory is restored. The rating logic in the evaluation report uses a piecewise function mapping. The B+ rating corresponds to a weight range of 1.0-1.05, which is clinically interpreted as "the overall nutritional goals have been achieved, but there is room for improvement in minor nutrients".

[0048] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0049] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for infant nutrition intake analysis and diet intelligence management, characterized in that, The method comprises: interactively obtaining physiological characteristic parameters and dietary history parameters of a young child individual, wherein the physiological characteristic parameters comprise age information, weight information and health status information, and the dietary history parameters comprise food intake record information and nutritional composition information; performing nutritional demand matching according to the health status information and the nutritional composition information, obtaining a nutritional demand selection, and then performing initialization configuration of a dietary plan based on the nutritional demand selection to obtain a target dietary plan; interactively obtaining target structural characteristics of the target dietary plan, and performing dietary management condition initialization according to the target structural characteristics and the nutritional demand selection to output initial management control constraints, wherein the initial management control constraints comprise a nutritional intake threshold and a maximum allowable deviation rate; taking the initial management control constraints as optimization conditions, performing intake speed optimization according to the food intake record information and a preset time constraint to locate an intake control speed; performing nutritional deviation trend analysis of the target dietary plan according to the intake control speed, and performing dietary adjustment control optimization based on an analysis result to obtain dietary adjustment control parameters; when food intake management is performed at the intake control speed in a dietary execution process, the dietary adjustment control parameters are used for intermittent automatic adjustment of the target dietary plan.

2. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 1, wherein, According to the health status information and the nutritional composition information, nutritional demand matching is performed to obtain a nutritional demand selection, and the method comprises: based on nutritional types, the nutritional composition information is resolved to obtain multiple nutritional content distribution characteristics of multiple nutritional types; according to a preset health standard, the multiple nutritional content distribution characteristics are updated to obtain multiple demand matching distribution characteristics; the multiple demand matching distribution characteristics are fused to obtain a global nutritional distribution characteristic; based on the global nutritional distribution characteristic, multi-level nutritional demand matching is performed to obtain multiple layers of dietary plan components, wherein the multiple layers of dietary plan components have multiple nutritional scales, plan lengths and food type identifiers; based on the multiple nutritional scales, the multiple layers of dietary plan components are serialized and assembled to obtain the nutritional demand selection.

3. The method of claim 2, wherein the method further comprises: Interactive obtaining of target structural characteristics of the target dietary plan, and according to the target structural characteristics and the nutritional demand selection, dietary management condition initialization is performed to output initial management control constraints, and the method comprises: interactively obtaining multiple component attribute information of the multiple layers of dietary plan components; according to the target structural characteristics, effective intake capacity is calculated to obtain effective intake capacity; after multiple dietary simulation models are obtained through network data calling according to the multiple component attribute information and the effective intake capacity, the multiple dietary simulation models are fused to obtain a dietary plan model; after a standard intake speed is obtained through local data calling of the dietary plan, nutritional kinetics simulation of the dietary plan model is performed using the standard intake speed to output a structural deviation threshold, and nutritional kinetics simulation of the multiple dietary simulation models is performed using the standard intake speed to output multiple single-body deviation thresholds; intersection solving is performed on the structural deviation threshold and the multiple single-body deviation thresholds to obtain the nutritional intake threshold; Conducting a nutrition dynamics simulation of the diet plan model with a maximum value of the nutrition intake threshold as a constraint, and calling the maximum allowed deviation rate from the diet plan model based on a simulation result.

4. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 3, wherein, The method comprises the following steps: Pre-building an intake-related prediction model; Updating data of the standard intake speed by taking the reciprocal of the standard intake speed as a speed change scale to obtain a first backup intake speed; Inputting the first backup intake speed into the intake-related prediction model to obtain a first deviation rate time limit, a first nutrition intake time limit, and a first predicted intake efficiency; Calculating a first predicted intake time by using the food intake record information and the first predicted intake efficiency, and if the first predicted intake time meets the preset time constraint, retaining the first backup intake speed and taking a first difference between the first deviation rate time limit and the first nutrition intake time limit as a first control reliability coefficient of the first backup intake speed; By continuously updating the standard intake speed and evaluating the update result by using the intake-related prediction model and the preset time constraint, a plurality of backup intake speeds and a plurality of control reliability coefficients meeting a preset update frequency are obtained; Serializing the plurality of control reliability coefficients and positioning the intake control speed from the plurality of backup intake speeds based on a sorting result.

5. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 4, wherein, Pre-building an intake-related prediction model, the method comprising: Using the nutrition demand selection and nutrition composition information as historical data calling constraints to call historical intake-related data, wherein the historical intake-related data comprises a plurality of sample intake speeds, a plurality of sample intake efficiencies, a plurality of deviation rate change speeds, and a plurality of nutrition intake change speeds; Calculating a plurality of nutrition intake change time limits based on the nutrition intake threshold and the plurality of nutrition intake change speeds; Calculating a plurality of deviation rate change time limits based on the maximum allowed deviation rate and the plurality of deviation rate change speeds; Using the plurality of sample intake speeds, the plurality of sample intake efficiencies, the plurality of nutrition intake time limits, and the plurality of deviation rate change time limits as training construction to train the intake-related prediction model based on a prediction neural network.

6. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 4, wherein, Conducting nutrition deviation trend analysis of the target diet plan based on the intake control speed, and performing diet adjustment control optimization based on an analysis result to obtain a diet adjustment control parameter, the method comprising: Calling direction data of the intake control speed to obtain a control deviation rate time limit and a control nutrition intake time limit; Calling a smaller value by comparing the control deviation rate time limit and the control nutrition intake time limit as an adjustment period; Predefining an adaptability weight configuration and constructing an adaptability evaluation function based on the adaptability weight configuration, wherein the adaptability weight configuration comprises an energy consumption weight, a nutrition loss weight, and a time consumption weight; The initial management control constraint is called as a data calling constraint for historical dietary adjustment record screening, and a plurality of sample dietary adjustment records are obtained, wherein each sample dietary adjustment record in the plurality of sample dietary adjustment records comprises a sample control parameter and a sample adjustment consumption; The fitness evaluation function is used to evaluate the fitness of the plurality of sample dietary adjustment records, and the initial control parameter is located based on the evaluation result, wherein the initial control parameter is composed of a food type flow rate, a food type adjustment time length, a nutritional supplement flow rate, a nutritional supplement adjustment time length, and a comprehensive adjustment time length; The adjustment period is added to the initial control parameter to obtain the dietary adjustment control parameter.

7. The method of claim 6, wherein the method further comprises: determining the nutritional intake of the infant based on the food intake data; and providing a recommendation for the infant based on the nutritional intake of the infant. The sample control parameter is composed of a sample food type flow rate, a sample food type adjustment time length, a sample nutritional supplement flow rate, a sample nutritional supplement adjustment time length, and a sample comprehensive adjustment time length.

8. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 7, wherein, The sample adjustment consumption is composed of a sample energy consumption, a sample nutrient loss, and a sample time consumption.

9. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 1, wherein, During the food intake management at the intake control speed in the dietary execution process, the dietary adjustment control parameter is used for intermittent automatic adjustment of the target dietary plan, and the method comprises: The dietary execution process is divided into a plurality of discrete time periods, and the food type adjustment time length in the dietary adjustment control parameter is evenly distributed to the plurality of discrete time periods; After the end of a previous discrete time period, it is judged whether the nutrient intake of the target dietary plan in the discrete time period reaches the expectation; The food type adjustment time length in a subsequent discrete time period is adaptively modified until the automatic adjustment of the entire dietary execution process is completed.

10. The infant nutritional intake analysis and diet intelligence management method as claimed in claim 9, wherein, The method for evaluating the nutrient intake of the target dietary plan comprises: Obtaining the change value of the nutrient level of the infant individual before and after the dietary execution; Based on the change value of the nutrient level, the actual nutrient intake amount is obtained; The actual nutrient intake amount and the theoretical nutrient requirement are analyzed to obtain the control degree of the dietary adjustment control parameter; Based on the control degree, the nutrient intake of the target dietary plan is evaluated.

Images