Intelligent ice making amount control method of ice maker

Abstract

The invention provides an intelligent ice-making amount control method for an ice maker, which comprises the following steps of: acquiring operation data including environment temperature, humidity, residual ice amount of an ice storage bin and energy consumption of a refrigerating system in real time through an environment temperature sensor, a humidity sensor, an ice storage bin sensor and an energy consumption detection module, and combining with user ice using historical data recorded by a storage module. And based on the data, establishing an ice consumption rule model for predicting the next ice consumption time point and the required ice amount, and calculating the amount of ice to be supplemented by combining the current remaining ice amount of the ice storage bin. And calculating the required ice-making time by using the amount of ice to be supplemented and the environmental parameters, and determining the optimal ice-making starting time point according to the required ice-making time so as to dynamically optimize the ice-making time arrangement. In the ice making process, the ice making power is adjusted by monitoring the return water temperature change rate, and when the return water temperature change rate is larger than a first threshold value, the ice making power is maintained. According to the method, the ice making amount can be intelligently adjusted, the ice making efficiency is improved, the energy consumption is reduced, and diversified ice using requirements of users are met.

Description

Technical Field

[0001] The invention relates to the technical field of ice maker control, and in particular to an intelligent ice making quantity control method for an ice maker. Background Art

[0002] Ice machines are widely used in home, commercial and industrial scenarios. Conventional ice machines usually make ice through a simple refrigeration cycle, with fixed ice-making time and power settings, and do not dynamically adjust operating parameters. In addition, some advanced ice machines can achieve basic temperature monitoring or ice storage detection by installing environmental sensors and ice storage sensors, but lack intelligent analysis and control of ice demand.

[0003] However, the existing technology has the following major problems: the ice-making process lacks effective prediction of users' ice usage patterns, and cannot dynamically adjust the ice-making plan according to real-time environmental parameters and ice storage needs, resulting in excessive or insufficient ice-making, which causes energy waste or fails to meet user needs. In addition, during the ice-making process, the energy consumption optimization and power adjustment mechanisms are insufficient, and it is impossible to effectively balance ice-making efficiency and energy conservation.

[0004] Therefore, the present invention proposes an intelligent ice making amount control method for an ice maker. Summary of the invention

[0005] The present application provides a first embodiment of the present application to provide an intelligent ice making amount control method for an ice maker, so as to improve ice making efficiency and reduce energy consumption.

[0006] The present application provides an intelligent ice making quantity control method for an ice maker, comprising:

[0007] Collecting operation-related data of the ice maker, specifically including obtaining operation data through installed ambient temperature sensors, humidity sensors, ice storage bin sensors and energy consumption detection modules, the operation data including ambient temperature, ambient humidity, the amount of ice remaining in the current ice storage bin and energy consumption data of the refrigeration system, and recording the time period distribution of users taking ice, the amount of ice taken at a single time, the total amount of ice used per day and the difference in ice used between working days and weekends through the storage module;

[0008] Based on the operation-related data, an ice usage law model is established, and the ice usage law model is used to predict the next ice usage time point and the estimated ice quantity required; and the amount of ice to be replenished is calculated according to the remaining ice quantity in the current ice storage bin;

[0009] Calculate the required ice-making time by combining the amount of ice to be replenished with the ambient temperature and humidity;

[0010] Determine the optimal ice-making start time based on the expected ice-use time and required ice-making time;

[0011] During the ice-making process, the ice-making power is controlled by monitoring the return water temperature change rate, specifically including maintaining the current ice-making power when the return water temperature change rate is greater than a first threshold, reducing the ice-making power to save energy when the return water temperature change rate is less than the first threshold, and stopping the ice-making process when the return water temperature change rate drops to a second threshold.

[0012] The beneficial effects of the technical solution provided by this application include:

[0013] (1) By collecting real-time data on ambient temperature, humidity, remaining ice in the ice storage bin, and energy consumption, and combining it with the user's ice usage model, it can accurately predict ice demand, dynamically adjust ice-making time and power, avoid excessive or insufficient ice-making, and improve ice-making efficiency. (2) By monitoring the return water temperature change rate, it can dynamically adjust the ice-making power, reduce the power when the ice-making efficiency decreases, and stop the ice-making process in time after ice-making is completed, significantly reducing unnecessary energy consumption and achieving energy-saving effects. (3) Based on the prediction function of the ice usage model, it can reasonably arrange the ice-making start time to ensure sufficient ice storage during the peak ice usage period, while reducing ice-making time during the trough period, optimizing ice supply, and improving user experience. (4) By comprehensively analyzing real-time data and historical data, an intelligent control mechanism is established to realize the automated operation of the entire ice-making process without the need for manual intervention by the user. At the same time, it has strong adaptability and flexibility to meet the needs of different ice usage scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The first embodiment of the present application provides a flowchart of a method for controlling the amount of ice made by an ice maker. DETAILED DESCRIPTION

[0015] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0016] The first embodiment of the present application provides an intelligent ice making amount control method for an ice maker. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 A first embodiment of the present application provides an intelligent ice making quantity control method for an ice maker for detailed description.

[0017] Step S101: Collecting operation-related data of the ice maker, specifically including obtaining operation data through installed ambient temperature sensors, humidity sensors, ice storage bin sensors and energy consumption detection modules, wherein the operation data include ambient temperature, ambient humidity, the current amount of ice remaining in the ice storage bin and energy consumption data of the refrigeration system, and recording the time period distribution of users taking ice, the amount of ice taken at a single time, the total amount of ice used per day and the difference in ice used between weekdays and weekends through the storage module.

[0018] Step S101 aims to collect the operation-related data of the ice maker, laying the foundation for the subsequent construction of the ice usage law model and intelligent ice making control. Specifically, this step completes the comprehensive collection and recording of real-time data and historical data through the collaborative operation of hardware equipment and software.

[0019] The ambient temperature sensor and humidity sensor are installed outside the ice maker or in the ice making environment to monitor the ambient temperature and humidity in real time. These sensors can be devices based on digital measurement technology, such as using thermal resistors or semiconductor materials as temperature and humidity sensing elements. The collected temperature and humidity data can quantify the impact of the environment on ice making efficiency. For example, under high temperature and high humidity conditions, ice making time may be extended.

[0020] The ice storage bin sensor is used to monitor the remaining ice in the ice storage bin. It usually uses a weight sensing module or an infrared distance sensor to determine the ice quantity data. The weight sensing module is based on the piezoelectric effect and calculates the current ice quantity in the ice storage bin by measuring the pressure of the ice cubes on the sensor, while the infrared sensor measures the stacking height of the ice cubes through the reflection distance. Accurate collection of ice quantity data is important to avoid excessive ice making or insufficient ice.

[0021] The energy consumption detection module obtains energy consumption data during the refrigeration process by measuring the current, voltage and power of the refrigeration system in real time. This module is usually composed of a Hall sensor or a shunt, which can record the energy consumption characteristics of the ice maker under different operating conditions and provide basic data support for energy-saving optimization.

[0022] The storage module records the user's ice-taking history data, including the distribution of ice-taking time periods, the amount of ice taken at a time, the total amount of ice used per day, and the difference in ice usage between weekdays and weekends. The user's ice-taking data is usually automatically recorded by inputting a log into the control panel of the ice maker, or by collecting the trigger time and quantity information of each ice-taking event through a sensor (such as a photoelectric sensor) installed at the ice-taking port. For the time period distribution data, regular features can be extracted by counting the user's ice-using behavior in multiple time periods of the day; the amount of ice taken at a time is obtained by accumulating the actual amount of ice cubes in each ice-taking event; and the total amount of ice used per day is the cumulative result of all ice-taking events in a day. In addition, by analyzing the ice usage data within a week, the difference in ice usage between weekdays and weekends can be extracted.

[0023] The real-time collected temperature, humidity, ice volume and energy consumption data are transmitted to the main controller through the communication module (such as I2C, SPI or wireless communication protocols such as Wi-Fi). The main controller processes these data, including data filtering (removing outliers), normalization (ensuring that the scale of different data is consistent), and storage updates. The storage of historical data uses a database system, such as an embedded SQL database, to achieve efficient data query and analysis. The processed data is stored in categories for subsequent steps.

[0024] This embodiment realizes redundant data collection through a dual sensor architecture (for example, two temperature sensors or humidity sensors) to improve the robustness of the system and prevent the failure of a single sensor from affecting the accuracy of the data. In the data redundancy design, by adopting a dual sensor architecture, such as installing two temperature sensors and two humidity sensors at different locations of the ice maker, the robustness of the system can be significantly improved. The design goal of this configuration is to avoid interruption or deviation of data collection due to failure or reduced accuracy of a single sensor. The specific implementation method is to collect data from two sensors at the same time and calculate their consistency by comparing the two sets of data. If it is detected that the reading deviation of the two exceeds the preset threshold, a warning is issued and the backup sensor is enabled, and the operating status of the faulty sensor is recorded for subsequent maintenance.

[0025] This embodiment adopts dynamic calibration. During operation, dynamic calibration of the sensor is performed regularly. A known standard environment (such as constant temperature and humidity conditions) is used to detect the accuracy of the sensor and perform compensation to ensure data reliability. In this embodiment, the sensor of the ice maker is regularly exposed to known standard environmental conditions for calibration. For example, the ice maker can be operated in a constant temperature and humidity chamber, and the deviation value of the sensor can be identified and the calibration parameters can be automatically generated by comparing with the reference temperature and humidity data of the environment. These parameters will be loaded into the compensation algorithm for sensor data processing through the main control module to ensure the reliability and accuracy of the data during operation. In addition, a self-test function can also be designed, such as triggering a calibration mode when the ice maker is started, simulating a variety of temperature and humidity conditions, and checking the response time and stability of the sensor.

[0026] This embodiment uses real-time data visualization to provide a dynamic display of real-time data on the user interface, such as displaying the ambient temperature, humidity and current ice volume, so that the user can intuitively understand the operating status of the ice maker. In specific implementation, the user interface can be designed as an embedded LCD display or linked with an application on a mobile device to transmit ambient temperature, humidity, remaining ice volume in the ice storage bin and energy consumption data in real time through wireless communication protocols (such as Wi-Fi or Bluetooth). These data can be presented in the form of charts or numbers, such as using a line chart to show the trend of temperature and humidity changes, or using a dynamic progress bar to display the current capacity of the ice storage bin. At the same time, an alarm prompt function can be designed on the interface. For example, when the ice storage volume is lower than the set minimum value, the interface can pop up a low ice storage volume warning and suggest that the user check the operating status of the ice maker or start ice making in time.

[0027] Step S101 realizes the comprehensive collection and storage of ice maker operation-related data, provides complete and high-quality data support for subsequent intelligent model construction and optimization, and enhances the system reliability and user experience.

[0028] Furthermore, the collection of ice-making machine operation-related data includes:

[0029] The ambient temperature and humidity data at different locations around the ice maker are collected through multi-point arranged ambient temperature sensors and humidity sensors, and the collected data are transmitted to the main control module for multi-point data fusion, and the spatial average values ​​of ambient temperature and humidity are calculated, and the areas with large fluctuations are marked for further analysis;

[0030] In the ice storage bin, the infrared ranging sensor is used to dynamically monitor the distribution of the remaining ice in the ice storage bin. By sensing the ice accumulation height and coverage area, the current remaining ice volume in the ice storage bin is estimated. At the same time, the weight data of the piezoelectric sensor at the bottom of the ice storage bin is measured to further verify the accuracy of the remaining ice calculation.

[0031] The energy consumption detection module collects the current, voltage and power data of the refrigeration system in real time, records the collected instantaneous data in time series and performs filtering processing to remove interference signals caused by power grid fluctuations and generate a stable energy consumption curve;

[0032] Combined with the data from the ice storage bin sensor, the energy efficiency of each ice-making cycle of the ice maker is analyzed.

[0033] The collection of ambient temperature and humidity data is achieved using sensors arranged at multiple points. Multiple ambient temperature sensors and humidity sensors are installed at different locations of the ice maker, such as the top, side, and key areas of the nearby space. This arrangement can capture temperature and humidity gradients and non-uniformities that may exist in the environment. Each sensor collects data regularly at set time intervals and transmits the data to the main control module via wireless or wired communication. The main control module fuses the data from multiple sensors and calculates the spatial average using a weighted average or interpolation algorithm to obtain a comprehensive reflection of the ambient temperature and humidity around the ice maker. At the same time, for areas with large fluctuations, such as locations with significant temperature or humidity differences, marking and recording can be used for subsequent optimization analysis, such as determining the local impact of environmental conditions on ice-making efficiency.

[0034] The monitoring of the remaining ice in the ice storage bin combines dual verification of sensors based on infrared ranging and piezoelectric sensors. The infrared ranging sensor measures the height and coverage area of ​​the ice cubes in the ice storage bin in a non-contact manner, and calculates the remaining volume of ice cubes in the current ice storage bin based on the known geometric dimensions of the ice storage bin. In addition, a piezoelectric sensor is installed at the bottom of the ice storage bin to measure the weight data of the ice cubes in the ice storage bin in real time. By comparing the volume data calculated by the infrared ranging sensor with the weight data measured by the piezoelectric sensor, the calculation result of the remaining ice amount can be further verified and optimized to ensure the accuracy of the monitoring results. Especially when the shape and density of the ice cubes may change, this dual verification mechanism can significantly improve the reliability of the data.

[0035] The energy consumption detection module is designed to monitor the operating status of the ice-making machine refrigeration system in real time. By collecting the current, voltage and power data of the refrigeration system, time series data is generated to characterize the instantaneous energy consumption of the refrigeration system. In order to improve the stability and availability of the data, the main control module filters the collected data to filter out transient signals caused by power grid fluctuations or external interference, ensuring that the generated energy consumption curve has a high signal-to-noise ratio. The filtering algorithm can choose low-pass filtering or Kalman filtering, and optimize the filtering parameters according to the dynamic characteristics of the refrigeration system.

[0036] Combined with the remaining ice data from the ice storage bin sensor, the output of the energy consumption detection module can also be used to analyze the energy efficiency of each ice-making cycle of the ice-making machine. By matching the energy consumption curve with the time period of the ice-making cycle and calculating the total energy consumption and unit ice consumption of each ice-making cycle, a quantitative basis can be provided for the optimization of ice-making machine operation. This correlation analysis can be further extended to trend analysis over a long period of time, such as evaluating the long-term impact of ambient temperature and humidity on energy efficiency or comparing energy consumption differences under different ice-making modes.

[0037] Through the above detailed implementation steps, high-quality operation-related data can be collected by utilizing technologies such as multi-point sensor arrangement, data fusion processing and multi-mode verification, which provides reliable technical support for the subsequent steps of ice usage law model construction and intelligent control.

[0038] Furthermore, the collecting of the operation-related data of the ice-making machine also includes:

[0039] In the data record of user ice taking behavior, the photoelectric sensor installed at the ice taking port of the ice storage bin and the ice taking port activity switch are used to sense the triggering time of each user taking ice and the opening time of the ice taking port, so as to calculate the single ice taking amount and accurately record the ice taking time by date and time;

[0040] Set the user authority grouping mode to distinguish the difference in ice usage of different users on weekdays and weekends, and store the grouping data in the historical database of the main control module for subsequent analysis.

[0041] First, the photoelectric sensor installed at the ice outlet of the ice storage bin and the active switch of the ice outlet work together to sense the user's ice taking behavior in real time. The photoelectric sensor can detect the state change of the ice outlet by emitting and receiving infrared light beams. For example, when the ice outlet is opened, the infrared light beam is interrupted, triggering the sensor to record the start time of the event; and when the ice outlet is closed, the light beam is restored and the end time of the event is recorded. At the same time, the active switch of the ice outlet can further confirm the validity of the ice taking event, for example, through the mechanical switch to sense whether the ice outlet is completely closed or not fully opened, to avoid data errors caused by abnormal operations. The opening time of the ice outlet is calculated from the start time and end time of the event. Combined with the ice release rate of the ice storage bin, the amount of ice taken at a single time can be calculated.

[0042] To ensure the accuracy of data recording, each triggering moment of ice taking is accurately recorded by date and time. This precise time recording uses the high-precision clock of the main control module as the time reference to ensure the accuracy and consistency of the data. For example, when the user takes ice at 9:15 am, the system binds the timestamp with the estimated amount of ice taken and stores it in the historical database. This recording method provides detailed time and quantitative data support for subsequent analysis of ice usage patterns.

[0043] In addition, in order to further explore the characteristics of users' ice usage behavior, the system supports user permission grouping mode. For example, for multiple users of the same ice machine, user grouping can be achieved by entering user identification (such as personal password or smart card authentication). The grouping mode can distinguish the differences in ice usage behavior of different users on weekdays and weekends. For example, the analysis of a user's ice usage on weekdays may be concentrated in the morning and afternoon, while on weekends it may be more dispersed. Through the grouped recorded data, the main control module can further construct an ice usage law model based on user classification, thereby providing personalized ice-making optimization solutions for different users.

[0044] All recorded data is stored in the historical database of the main control module for subsequent analysis. This data can be used for multiple purposes, such as counting the total amount of ice used by a user in a specific time period, evaluating the difference in ice use between weekdays and weekends, or predicting ice demand in the future. This grouping and recording method not only improves the sophistication of data collection, but also provides accurate basic information for the intelligent control of the ice machine.

[0045] Step S102: Based on the operation-related data, an ice usage law model is established, and the ice usage law model is used to predict the next ice usage time point and the estimated required ice volume; and the amount of ice to be replenished is calculated based on the remaining ice volume in the current ice storage bin.

[0046] The core of step S102 is to establish an ice usage law model based on the collected operation-related data. The main purpose of this model is to predict the next ice usage time point and the estimated amount of ice required, and at the same time calculate the amount of ice to be replenished based on the current remaining ice in the ice storage bin, thereby providing accurate data support and decision-making basis for subsequent steps.

[0047] When establishing the ice usage pattern model, the data obtained in step S101 must first be preprocessed. This includes cleaning, normalizing, and feature extraction of real-time and historical data. For example, for the distribution data of users' ice collection time periods, the peak and low periods of ice usage in a day can be extracted through statistical analysis; for the daily total ice usage data, the seasonal variation trend of ice usage can be extracted based on historical records of different seasons; for the difference in ice usage between working days and weekends, classification processing can be used to establish ice usage patterns for different date types.

[0048] Next, we use the preprocessed data to build a model of ice usage patterns using data modeling techniques. This model can use statistical analysis-based methods, such as time series analysis models or regression models, to capture the periodic characteristics and trend changes of users' ice usage behaviors. For more complex ice usage patterns, we can use machine learning-based algorithms, such as long short-term memory networks (LSTM) or random forest models, which can effectively process multidimensional data and nonlinear relationships, further improving prediction accuracy.

[0049] During the model training process, the input data includes multiple variables such as ambient temperature, humidity, remaining ice in the ice storage bin, historical ice usage, and the time distribution of users taking ice. Through iterative training, the model can learn the correlation between different variables. For example, an increase in ambient temperature may lead to an increase in ice demand, or there is a significant difference in ice usage between weekdays and weekends. The model output results include the next time to use ice and the estimated amount of ice required, which are used to guide the operation strategy of the ice maker.

[0050] After obtaining the model output, the amount of ice to be replenished is calculated based on the remaining ice in the current ice storage bin. This process requires the introduction of the maximum capacity parameter of the ice storage bin and the safety reserve threshold. For example, if the estimated amount of ice required is 10 kg and the remaining ice in the current ice storage bin is 6 kg, the amount of ice to be replenished is 4 kg, and it is necessary to ensure that the final ice storage does not exceed the maximum capacity of the ice storage bin. The calculation result of the amount of ice to be replenished will be directly used as the key input for calculating the ice making time in the next step (S103).

[0051] In order to improve the adaptability and accuracy of the model, a dynamic update mechanism can be designed. The data after each user's actual ice usage is compared with the model prediction results, and the model parameters are adjusted according to the deviation. For example, if the model underestimates the ice usage multiple times, the prediction weight during peak hours is increased; if the ice usage is overestimated multiple times, the influence weight of seasonal characteristics is reduced. In addition, the dynamic update mechanism can also introduce ice usage data from the most recent month and retrain the model to adapt to seasonal changes or long-term adjustments in user behavior.

[0052] In order to enhance the robustness of the model, outlier processing and model validation mechanisms can be added. Outlier processing can be achieved through statistical methods or machine learning algorithms, such as using the box plot method to eliminate excessive single ice extraction data. Model validation can use the cross-validation method to evaluate the generalization ability of the model by dividing the training set and the test set to ensure reliability in actual use.

[0053] Through the above detailed design, the ice usage law model can make full use of the collected data, accurately predict the ice demand and calculate the amount of ice to be replenished, ensuring that the ice making process is efficient, intelligent and energy-saving.

[0054] Furthermore, the ice usage regularity model includes an environment analysis module, a user behavior analysis module and a comprehensive prediction module;

[0055] The environmental analysis module is used to analyze the impact of environmental conditions on ice demand. The input of the environmental analysis module includes the real-time collected environmental temperature and humidity; the output is the influencing factor of the environmental conditions; the environmental analysis module is implemented by associating the environmental temperature and humidity in the historical records with the trend of ice consumption, and using a model building method based on statistical regression, so as to obtain the dynamic impact of the current environmental conditions on ice demand;

[0056] The user behavior analysis module is used to analyze the user's ice usage habits and ice usage patterns; the input of the user behavior analysis module includes the time period distribution of users taking ice, the amount of ice taken at a time, the total amount of ice used per day, and the difference in ice usage between weekdays and weekends; the output is the peak time period for users to use ice and the expected trend of ice usage; the user behavior analysis module uses a classification algorithm and a cluster analysis model based on user behavior characteristics to extract users' ice usage habits from historical data and predict users' ice demand characteristics in future time periods;

[0057] The comprehensive prediction module is used to combine the output of the environmental analysis module and the output of the user behavior analysis module to generate the next ice usage time point and the predicted ice usage amount required; the input of the comprehensive prediction module is the influencing factors of environmental conditions, the peak time period of user ice usage and the predicted ice usage trend; the output is the predicted next ice usage time point and the predicted ice usage amount required; the comprehensive prediction module is implemented based on a multivariate weighted regression model, comprehensively considering the environment and user behavior characteristics, and optimizing the prediction accuracy by dynamically adjusting the weights.

[0058] The function of the environmental analysis module is to quantify the dynamic impact of environmental conditions on ice demand. First, the module receives the ambient temperature and humidity data collected in real time by the temperature sensor and humidity sensor, and smoothes the data through the weighted moving average algorithm to remove the noise signal of short-term fluctuations while retaining the overall trend of environmental changes. To ensure the flexibility of data processing, the weighting factor is dynamically adjusted according to the length of the time window, for example, a shorter time window is used in an environment with large short-term fluctuations, and a longer time window is used in a stable environment, so as to balance real-time performance and data stability. The smoothed data is associated with the trend of ambient temperature, humidity and ice consumption in the historical records, and a nonlinear mapping relationship is extracted using a statistical regression-based method. This mapping relationship can reveal the specific impact of temperature and humidity on ice demand, for example, high temperature may increase ice demand, while high humidity may reduce ice making efficiency. Finally, the environmental analysis module calculates the influencing factors of environmental conditions through the support vector regression model, including the positive correction coefficient of ambient temperature on ice demand and the negative correction coefficient of ambient humidity on ice making efficiency, providing dynamic input for subsequent modules.

[0059] The user behavior analysis module is used to extract the user's ice usage habits and patterns and generate prediction parameters for ice demand. The input data includes the time period distribution of users' ice collection, the amount of ice collected in a single trip, the total amount of ice used per day, and the difference in ice usage between weekdays and weekends. These data are processed through classification algorithms and cluster analysis models. First, the typical behavior patterns of users are identified, such as ice demand during peak time periods and random demand during off-peak periods; then, these patterns are used to build a user behavior feature model. The model further analyzes the ice usage patterns in historical data, combines the ice usage characteristics of specific date types (such as holidays or weekdays), and outputs the user's peak ice usage time period and the expected ice usage trend. These outputs provide a quantitative description of user behavior for the subsequent comprehensive prediction module.

[0060] The task of the comprehensive prediction module is to combine the outputs of the environmental analysis module and the user behavior analysis module to generate the next ice usage time point and the predicted ice usage amount. The inputs received by the comprehensive prediction module include the influencing factors of environmental conditions, the peak time period of ice usage for users, and the expected ice usage trend. Based on the multivariate weighted regression model, this module optimizes the prediction accuracy by dynamically adjusting the weights. For example, when the changes in ambient temperature and humidity have a greater impact on the ice usage demand, the weight of environmental factors increases; and when the regularity of user behavior dominates, the weight of user behavior data is increased. Through this dynamic adjustment, the comprehensive prediction module can provide high-precision prediction results under complex environmental and user behavior conditions. In addition, to ensure the reliability of the prediction, the comprehensive prediction module performs anomaly detection and correction on the generated next ice usage time point and the predicted ice usage amount, and identifies and corrects possible anomalies by comparing the distribution range of the prediction results with the historical data.

[0061] Through the close collaboration of the environmental analysis module, user behavior analysis module and comprehensive prediction module, the ice usage pattern model can achieve comprehensive and accurate prediction of ice usage demand.

[0062] Furthermore, the environmental analysis module is specifically used for:

[0063] Receive the real-time collected ambient temperature and ambient humidity data, and smooth the data through a weighted sliding average algorithm to obtain stable environmental input data as the basis for subsequent steps; the weighting factor of the weighted sliding average algorithm is dynamically adjusted according to the length of the time window to ensure that the smoothed data can reflect the overall trend of environmental changes while filtering out short-term fluctuation noise;

[0064] Based on the obtained smoothed data, the changing trend of ambient temperature and humidity and ice consumption in historical records is correlated, and the nonlinear mapping relationship between ambient temperature and humidity and ice consumption demand is extracted using the time series segmented regression method.

[0065] According to the real-time collected ambient temperature and ambient humidity data, and in combination with the extracted nonlinear mapping relationship, the influencing factors of the environmental conditions are calculated through a support vector regression model based on a kernel function, wherein the influencing factors include a positive correction coefficient of the ambient temperature on the ice demand and a negative correction coefficient of the ambient humidity on the ice making efficiency, and the specific effects of the environmental conditions on the ice demand and the ice making time under different scenarios are quantified respectively;

[0066] After each ice-making cycle is completed, the output influencing factors are compared with the measured values ​​of actual ice-making time and ice usage through multivariate cross-validation, the segmented weight mechanism and support vector regression parameters are dynamically adjusted, and the nonlinear mapping relationship and kernel function parameters of the model are updated to make it more accurately adapt to environmental changes and long-term operation status.

[0067] The environmental analysis module first receives the real-time environmental temperature and humidity data collected by the temperature sensor and humidity sensor. These raw data may contain short-term fluctuations and noise signals, so they need to be smoothed. To this end, the module uses a weighted moving average algorithm to preprocess the data. The weighted moving average algorithm dynamically adjusts the weighting factor according to the length of the time window to adapt to different environmental change characteristics. For example, when the environment changes drastically in a short period of time, the algorithm uses a smaller time window to respond to changes more quickly; when the environment is stable, the time window is extended to improve the data smoothing effect. Through this dynamic adjustment mechanism, the smoothed data can more accurately reflect the overall trend of the environment, while filtering out high-frequency noise, providing stable input data as the basis for subsequent analysis.

[0068] Based on the smoothed environmental data, the module associates it with the changing trends of ambient temperature and humidity and ice usage in historical records. To reveal the nonlinear relationship between the two, the module uses the time series segmented regression method to analyze the data. Time series segmented regression divides historical data into different time periods, such as seasons, day and night, or peak ice usage, and builds a regression model for each segment of data. This segmented processing method can capture the specific impact of ambient temperature and humidity on ice demand in different time periods, thereby generating a time-specific nonlinear mapping relationship. This mapping relationship is used as a core feature in subsequent steps to dynamically calculate the influencing factors of environmental conditions.

[0069] After combining the real-time collected environmental data and the extracted nonlinear mapping relationship, the module further calculates the influencing factors of environmental conditions through the support vector regression (SVR) model. The kernel function of the support vector regression model is specially designed as a complex function that combines the characteristics of environmental temperature and humidity, which can capture the dynamic impact of environmental factors on ice demand. The output influencing factors include the positive correction coefficient of ambient temperature on ice demand and the negative correction coefficient of ambient humidity on ice-making efficiency. The positive correction coefficient indicates that high temperature conditions may increase ice demand, while the negative correction coefficient reflects the decrease in ice-making efficiency due to increased humidity. These influencing factors are used to quantitatively describe the specific impact of environmental conditions on the operation of the ice maker and further optimize the ice-use law model.

[0070] To ensure the accuracy and adaptability of the model in long-term operation, the module performs multivariate cross-validation after each ice-making cycle, comparing the influencing factors output by the support vector regression model with the actual ice-making time and ice usage measurements. This validation mechanism allows the identification of deviations in the prediction results and the dynamic adjustment of the piecewise regression weights and support vector regression parameters. In particular, the kernel function parameters are optimized based on the latest environmental and operating data to improve the model's responsiveness to environmental changes. This closed-loop optimization design enables the environmental analysis module to continuously improve prediction accuracy over time, ensuring accurate calculation of influencing factors under different operating conditions.

[0071] In summary, the environmental analysis module accurately quantifies the impact of environmental conditions on ice demand and ice-making efficiency through dynamic smoothing of real-time data, segmented regression analysis of historical data, and kernel function calculation of the support vector regression model. Combined with multivariate cross-validation and dynamic parameter adjustment, the module implements a highly adaptable and precise intelligent control foundation, providing key technical support for the efficient operation of the ice machine.

[0072] Furthermore, the user behavior analysis module is specifically used to:

[0073] By performing cluster analysis on the time period distribution of users' ice-picking behaviors, the users' ice-picking behaviors are divided into peak hours, off-peak hours and low-peak hours based on the ice-picking times extracted from historical records. By adopting a dynamic time window adjustment mechanism, the clustering results are optimized in real time according to the changes in ice-picking behaviors in different seasons or on specific dates, and the characteristic ice usage in each time period is output.

[0074] Combining the single ice taking amount and the total daily ice usage, a classification algorithm based on Gaussian mixture model is used to generate the probability distribution of users' ice usage patterns, and analyze the randomness and regularity of users' ice usage. By distinguishing the ice usage behavior characteristics on weekdays and weekends, a mapping of users' ice usage patterns under different date types is established, and ice usage prediction parameters containing date characteristics are output.

[0075] By combining the clustering results of ice usage period distribution and the ice usage parameters generated by probability distribution, the impact of peak and random behaviors on ice usage demand is comprehensively considered in a weighted manner, and the user's ice usage peak time period and the expected ice usage trend are output.

[0076] The user behavior analysis module first performs cluster analysis on the distribution of time periods when users take ice, and divides the users' ice use behavior into peak time periods, off-peak time periods, and off-peak time periods. Based on the ice-taking time extracted from historical records, the module uses a clustering algorithm to automatically identify time periods with higher ice use frequency as peak time periods, and time periods with lower frequency as off-peak and off-peak time periods. This clustering result is optimized in real time through a dynamic time window adjustment mechanism to adapt to changes in ice use behavior in different seasons or on specific dates (such as holidays). For example, in summer, users may use ice more frequently in the afternoon, while the peak of ice use in winter may be concentrated in the morning. The time window adjustment mechanism dynamically expands or contracts the analysis time period according to seasonality and user behavior change trends, thereby ensuring that the results of the cluster analysis always meet current actual needs. Finally, the module outputs the characteristic ice usage for each time period, providing a basis for further user behavior modeling.

[0077] Based on the analysis of the single ice-taking amount and the total daily ice usage, the module uses the Gaussian mixture model (GMM) to classify and model the user's ice usage pattern. The model can generate a probability distribution of the user's ice usage pattern and quantitatively analyze the randomness and regularity of ice usage behavior. For example, through the Gaussian mixture model, the module can identify whether the user tends to take ice in a fixed amount, or whether the ice usage behavior is significantly affected by certain external factors (such as weather). At the same time, by classifying the ice usage data on weekdays and weekends, the module can establish a user's ice usage pattern mapping under different date types. For example, a user may start using ice earlier on a weekday, and the total ice usage is lower than on weekends, while the ice usage behavior on weekends may be more dispersed and the usage is larger. These pattern mappings provide highly distinguishing date features for generating ice usage prediction parameters with date information.

[0078] The user behavior analysis module finally analyzes the user's ice usage behavior in a weighted manner by combining the clustering results of the ice usage time distribution and the probability distribution data generated by the Gaussian mixture model. The weighted method dynamically assigns weights according to the importance of different factors. For example, peak hours may have higher weights, while the weight of random behavior increases as the random characteristics of ice usage increase. Through this weighted processing, the module can more accurately predict the user's peak ice usage time period and ice usage trend. The output results not only include peak time periods and ice usage trends, but can also dynamically adapt to short-term changes and long-term patterns of user behavior.

[0079] The design of this module provides high flexibility and accuracy for modeling user ice behavior by combining cluster analysis, Gaussian mixture model and dynamic weighting method. The module can capture user behavior characteristics in a variety of complex scenarios and generate parameters that play a key supporting role in subsequent predictions.

[0080] Furthermore, the comprehensive prediction module is specifically used for:

[0081] Combining the environmental condition influencing factors output by the environmental analysis module with the peak time period of ice usage and the expected ice usage trend output by the user behavior analysis module, a multivariate regression model based on time series is used to comprehensively calculate the next ice usage time point and the predicted required ice volume. The regression model uses the dynamic characteristics of environmental factors and the regularity of user behavior as input variables of the prediction factor.

[0082] The prediction accuracy of the regression model is optimized by dynamically adjusting the weighting factors, where the weight of environmental conditions is dynamically adjusted according to the range of changes in real-time temperature and humidity; the weight of user behavior is corrected by the historical ice usage accuracy;

[0083] The output results are detected for anomalies, and the historical statistical distribution of environmental and user behavior data is used to identify anomalies in the prediction results and make corrections based on the set tolerance range to ensure that the generated next ice usage time point and the predicted amount of ice required are consistent with the actual operating conditions.

[0084] The comprehensive prediction module first receives the environmental condition influencing factors output by the environmental analysis module and the user's peak ice usage time period and expected ice usage trend output by the user behavior analysis module. These input data represent the dynamic impact of the environment and user behavior on ice demand, covering the factors affecting ice-making efficiency due to real-time changes in ambient temperature and humidity, as well as the predictive characteristics of user ice usage patterns in different time periods. The module uses a multivariate regression model based on time series, taking these multidimensional data as input variables of the predictor, and comprehensively calculating the next ice usage time point and the predicted amount of ice required. The time series model can capture the time dependency in the data, such as the continuous impact of changes in ambient temperature and humidity on ice demand, as well as the short-term patterns and long-term trends of user ice usage behavior. By analyzing these characteristics, the module can accurately predict the user's next ice demand.

[0085] To improve the prediction accuracy of the regression model, the module optimizes the model by dynamically adjusting the weighting factors. The weight of the environmental conditions is dynamically adjusted according to the range of changes in real-time temperature and humidity. For example, when the ambient temperature and humidity fluctuate greatly, the weight of the environmental conditions is increased to reflect its significant impact on the demand for ice; when the temperature and humidity are relatively stable, the weight is appropriately reduced to reduce unnecessary interference with the prediction results. At the same time, the weight of user behavior is corrected by the historical ice usage accuracy. For example, if the user's ice usage behavior in a certain period of time deviates little from the prediction result, the weight of the user's behavior pattern is increased to enhance the model's reliance on behavioral features with high historical accuracy. This dynamic weighting mechanism ensures that the model can be adaptively adjusted under different environmental and user behavior conditions, improving the credibility of the prediction results.

[0086] The comprehensive prediction module also performs anomaly detection on the generated prediction results to ensure that the output results are consistent with the actual operating conditions. The module uses the historical statistical distribution of environmental data and user behavior data to analyze whether the prediction results are within a reasonable range. For example, by comparing the predicted time point of the next ice usage with the distribution of historical peak time periods, it is determined whether there is an anomaly at the predicted time point; at the same time, the predicted required ice volume is compared with the average ice usage and maximum deviation in the historical data to identify possible outliers. Once an anomaly is found, the module corrects the result according to the set tolerance range to ensure that the final prediction result is statistically reasonable and meets the actual operating needs.

[0087] Through the above steps, the comprehensive prediction module can effectively combine the dynamic characteristics of environmental conditions and the regularity of user behavior to generate an accurate next ice usage time and the predicted amount of ice required. At the same time, the module's dynamic weighting mechanism and anomaly detection function significantly improve the adaptability and robustness of the prediction model.

[0088] Furthermore, the kernel function-based support vector regression model adopts the kernel function provided by the following formula 5:

[0089]

[0090] Among them, K(x, y) is the kernel function value, which is used to measure the similarity of input vectors x and y in the high-dimensional feature space; the higher the similarity value, the closer the two input data are in the feature space.

[0091] The input vectors x and y represent the ambient temperature and humidity data and their corresponding ice demand data respectively; these data are collected by temperature and humidity sensors and matched with the historical database.

[0092] α is the Gaussian kernel strength adjustment parameter, which is used to control the rate at which similarity decreases. The recommended value is 1.5.

[0093] ||xy|| is the Euclidean distance, which represents the square distance between two input vectors and is used to measure the similarity of data. The larger the distance, the lower the similarity between the input data.

[0094] T env The current ambient temperature is collected in real time through the temperature sensor;

[0095] H is the current ambient humidity, which is collected in real time by the humidity sensor;

[0096] β is the ambient humidity correction coefficient, and the recommended value is 0.05.

[0097] γ is the temperature compensation constant, and the recommended value is 1.0.

[0098] δ is the cosine term amplitude adjustment parameter, with a recommended value of 0.3, which is used to adjust the impact of ambient temperature and humidity differences on similarity calculation; ∈ is a small positive constant to avoid the denominator being zero.

[0099] The following is the reference implementation code using the ice law model:

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] Step S103: Calculate the required ice-making time by combining the amount of ice to be replenished with the ambient temperature and humidity.

[0106] The core task of step S103 is to use the amount of ice to be added in combination with the ambient temperature and humidity to accurately calculate the required ice making time, thereby providing a reliable basis for the operation of the ice maker. This process not only requires the comprehensive influence of multiple variables, but also ensures that the calculation method has a high degree of accuracy and applicability to meet different environmental conditions and user needs.

[0107] First, according to the amount of ice to be replenished calculated in step S102, the amount of ice that the ice maker needs to produce is determined. This data is the basis for calculating the ice making time. On this basis, the data of ambient temperature and humidity need to be input into the refrigeration performance model. The refrigeration performance model can be predetermined by experiments or simulations based on the refrigeration capacity curve of the ice maker. For example, under conditions of low ambient temperature and low humidity, the ice making efficiency is high; while in a high temperature and high humidity environment, the refrigeration efficiency will be significantly reduced. Therefore, the refrigeration performance model needs to consider the specific effects of temperature and humidity on refrigeration efficiency, including changes in evaporation temperature, condensation temperature, and heat transfer performance.

[0108] In order to achieve accurate ice making time calculation, a correction factor can be introduced to dynamically adjust the basic refrigeration capacity of the ice maker. For example, the basic refrigeration capacity can be set as a standard value through the rated parameters of the refrigerator, and the temperature and humidity correction factor can be calculated by the following formula:

[0109] Correction coefficient = 1 / (1+α×ambient temperature+β×ambient humidity), where α and β are the temperature and humidity influencing factors determined by experiments. The corrected refrigeration capacity can be used to calculate the actual time required, and the formula is: ice making time = amount of ice to be replenished / corrected refrigeration capacity.

[0110] In actual implementation, the heat load characteristics of the ice storage bin also need to be considered. The heat load of the ice storage bin includes heat transfer loss and ambient heat recovery, which may cause the actual ice making time to be longer than the theoretical value. Therefore, a heat load compensation term can be added to the above formula, for example, the additional time required can be calculated based on the thermal insulation performance of the ice storage bin and the external ambient temperature difference, to further improve the accuracy of the calculation.

[0111] In order to improve the applicability of the method, a real-time feedback mechanism can be incorporated into the process of calculating the required ice-making time. For example, during the ice-making process, the changing trends of the evaporation temperature and the return water temperature are monitored in real time, and the parameters of the refrigeration capacity model are adjusted dynamically. If it is found that the actual refrigeration efficiency is lower than the expected value, it can be compensated by increasing the power or extending the ice-making time to avoid insufficient ice. In addition, the correction coefficient and the heat load compensation item can be optimized by learning from historical data. For example, after multiple ice-making cycles, by comparing the deviation between the predicted ice-making time and the actual ice-making time, the model parameters can be updated using linear regression or machine learning methods to improve long-term accuracy.

[0112] In addition, different ice-making strategies can be used for peak ice-use periods and off-peak ice-use periods. During peak periods, priority is given to ensuring ice-making speed, which may increase refrigeration power appropriately and sacrifice some energy efficiency; during off-peak periods, priority is given to energy efficiency optimization, reducing power to reduce energy consumption. In addition, environmental prediction data can be introduced, such as temperature and humidity trends in weather forecasts, to adjust the ice-making time plan in advance to adapt to possible future environmental changes.

[0113] Step S103 can dynamically adjust the calculation method of ice making time under different conditions, which not only ensures sufficient supply of ice cubes, but also achieves the goal of efficient and energy-saving operation.

[0114] Furthermore, the method of calculating the required ice making time by combining the amount of ice to be replenished with the ambient temperature and humidity includes:

[0115] Use the following formula 1 to calculate the required ice making time:

[0116]

[0117] Where M is the amount of ice to be replenished, which is calculated by the difference between the remaining ice in the ice storage bin and the estimated amount of ice required;

[0118] Q is the rated refrigeration capacity of the refrigerator, which is provided by the rated parameters of the equipment, usually provided by the manufacturer through test data, in kilograms per second, reflecting the maximum refrigeration efficiency of the ice maker under ideal conditions.

[0119] C loss It is the heat loss coefficient of the ice storage bin, which reflects the thermal insulation performance of the ice storage bin. The unit is joule / second·degree Celsius, which depends on the physical properties of the ice storage bin, such as material, thickness, surface area, etc. This coefficient is calibrated through experiments. The worse the thermal insulation performance, the higher the coefficient.

[0120] V storage The volume of ice to be added in the ice storage bin;

[0121] k eff It is the refrigeration efficiency constant, which reflects the cooling efficiency per unit energy consumption. It is obtained through experiments, and the recommended value is 100 to 500 joules / second, depending on the performance of the refrigeration system of the equipment.

[0122] ΔT system It is the difference between the refrigerant evaporation temperature and the internal temperature of the ice storage bin, expressed in degrees Celsius, and is measured by the operating parameters of the refrigeration system. Usually, the evaporation temperature is determined by the refrigerant type and system configuration, and the internal temperature of the ice storage bin is monitored in real time by sensors.

[0123] η(T env , H) is the environmental condition correction factor, calculated using the following formula 2:

[0124] η(T env , H)=1-k T ·(T env -T rated )-k H ·H(2)

[0125] Among them, T envis the current ambient temperature, which is collected in real time by the temperature sensor; H is the current ambient humidity, which is collected in real time by the humidity sensor; T rated is the rated operating temperature of the ice machine;

[0126] k T is the ambient temperature correction coefficient, the recommended value is 0.01, the unit is (℃ -1 ), determined experimentally, is used to quantify the effect of temperature deviation on refrigeration efficiency.

[0127] k H is the ambient humidity correction factor, the recommended value is 0.1, k T and k H They are used to quantify the effects of temperature and humidity on refrigeration efficiency, respectively, and are obtained through experimental calibration.

[0128] The first term in Formula 1 represents the time required for the ice maker to complete the ice-making task under standard operating conditions. This time is proportional to the amount of ice to be replenished and inversely proportional to the corrected refrigeration capacity. The environmental condition correction factor reflects the impact of ambient temperature and humidity on ice-making efficiency through dynamic adjustment. The second term supplements the heat loss compensation time of the ice storage bin, taking into account the additional impact of the ice storage bin's insulation performance and internal temperature difference on refrigeration demand. This sub-item calculation method ensures that the formula can accurately reflect the operating time requirements of the ice maker under different environmental and operating conditions.

[0129] Step S104: determining the optimal ice-making start time point according to the estimated ice-usage time point and the required ice-making time.

[0130] The core of step S104 is to reasonably determine the optimal ice-making start time based on the expected ice-use time and the required ice-making time, so as to ensure that the ice-making task can be completed in time before the ice-use demand occurs, and at the same time achieve the goal of efficient and energy-saving operation. To achieve this goal, it is necessary to comprehensively consider multiple factors such as the user's ice-use pattern, ice-making time, ice storage bin capacity, and environmental conditions, and design a scientific time scheduling strategy.

[0131] First, the predicted ice usage time point is provided by the ice usage regularity model in step S102, and this time point can be a high probability period of the user's historical ice usage behavior. For example, based on statistical results, if the user usually uses ice at 10 am every day, this period can be used as the predicted ice usage time point. On this basis, combined with the ice making time calculated in step S103, the start time of the ice making task can be preliminarily determined. However, this preliminary result needs to be further optimized to adapt to the actual situation of dynamic changes.

[0132] In order to accurately determine the optimal start time point, a time buffer strategy needs to be introduced. The purpose of the time buffer is to provide additional time margin for possible delays in the ice-making process to avoid insufficient ice supply due to sudden changes in environmental conditions or reduced ice-making efficiency. For example, in the hot and humid summer, the ice-making efficiency usually decreases. At this time, the time buffer can be dynamically adjusted according to the changing trends of ambient temperature and humidity to make the start time more adaptable. In addition, the time buffer can also be adjusted by analyzing the user's sensitivity to the amount of ice in the ice storage bin. For example, for users with stable ice demand, the time buffer can be shorter; while for users with large fluctuations in ice demand, the time buffer should be appropriately increased.

[0133] In order to further optimize the startup time point, the strategy can be adjusted based on the user's peak and off-peak periods of ice usage. During peak periods, priority is given to ensuring sufficient supply of ice cubes, such as starting the ice-making task in advance to avoid insufficient ice in the ice storage bin when ice is used. During off-peak periods, the startup time can be appropriately delayed to reduce energy waste caused by long ice storage time. For example, if the peak ice usage is expected to be at 10 a.m. and the off-peak ice usage is expected to be at 3 p.m., and the current ice storage bin has a certain amount of surplus, the startup time can be delayed as much as possible during the off-peak period, and ice storage can be completed before the peak period.

[0134] In order to optimize the time point, the control system of the ice maker needs to run a dynamic time scheduling algorithm. The algorithm can adjust the start time in real time based on real-time monitoring data input (such as ambient temperature, humidity and the current amount of ice in the ice storage bin), combined with the ice usage law model and ice making time. For example, if the ambient temperature suddenly rises and the ice making efficiency decreases, the algorithm will automatically advance the ice making start time to ensure that the ice making task is completed before the scheduled ice usage time point.

[0135] In addition to determining the optimal start time based on a single ice demand, system efficiency can also be improved through periodic optimization. For example, in cases where users' ice use behavior is highly periodic (such as a basically fixed daily ice use time), a long-term optimization model can be established by analyzing the ice use patterns for multiple consecutive days. The model can predict the peak and trough of ice use throughout the cycle based on historical data, and on this basis, plan the start time of the ice-making task to minimize energy consumption.

[0136] In addition, the determination of the optimal start time can also be optimized in combination with external data. For example, by connecting to the weather forecast system, the changing trend of ambient temperature and humidity can be known in advance, so as to make forward-looking adjustments to the ice-making time and start time. If the forecast shows that the temperature will rise significantly in a certain period of time in the future, the system can complete the ice-making task in advance to reduce the difficulty of ice-making in the future.

[0137] Step S104 can dynamically adjust the ice-making start time point under different ice demand and environmental conditions to ensure the timeliness of ice supply and the efficiency of the ice-making process.

[0138] Furthermore, the step of determining the optimal ice-making start time point according to the expected ice-use time point and the required ice-making time includes:

[0139] Use the following formula 3 to determine the optimal ice making start time t start :

[0140]

[0141] Among them, t pred The estimated time point of ice usage is output by the ice usage law model, indicating the time point when the user will use ice next time;

[0142] T is the required ice making time, which can be calculated according to formula 1;

[0143] T env is the current ambient temperature, which is collected in real time by the ambient temperature sensor; H is the current ambient humidity, which is collected in real time by the humidity sensor;

[0144] α 1 It is the temperature correction coefficient in seconds / degree Celsius. The recommended value is 10. It is determined through experiments and is used to indicate the advance amount of high temperature on refrigeration start.

[0145] β 1 It is the environmental correction factor, dimensionless, with a recommended value of 0.1, which reflects the relative impact of humidity on ice-making efficiency.

[0146] Δt buffer is the time buffer, calculated by the following formula 4:

[0147]

[0148] Among them, E sys It is the real-time energy consumption of the ice maker in watts. It is collected in real time by the energy consumption detection module and reflects the actual operating status of the current refrigeration system.

[0149] T sys The heat exchange response time of the ice maker is measured in seconds and is determined by the equipment performance parameters. The heat exchange response time of the ice maker refers to the time required from the ice maker starting the refrigeration process to the refrigerant reaching a stable heat exchange state.

[0150] k sys It is the system adjustment coefficient, which reflects the influence of the ice machine system status on the start-up time. The unit is second·watt -1, quantify the impact of system state on the time buffer, through experimental calibration, the recommended value is 0.01.

[0151] Step S105: During the ice-making process, the ice-making power is controlled by monitoring the return water temperature change rate, specifically including maintaining the current ice-making power when the return water temperature change rate is greater than a first threshold value, reducing the ice-making power to save energy when the return water temperature change rate is less than the first threshold value, and stopping the ice-making process when the return water temperature change rate drops to a second threshold value.

[0152] The core of step S105 is to dynamically adjust the ice-making power of the ice-making machine by monitoring the return water temperature change rate to optimize ice-making efficiency, save energy consumption and ensure the smooth completion of ice-making tasks. This process requires accurate data collection and real-time feedback mechanisms, combined with scientific control strategies to achieve dynamic adjustment.

[0153] During the ice-making process, the return water temperature is a key parameter of the ice-making machine's operating status, which can reflect the heat exchange efficiency of the refrigeration system and the current progress of ice-making. The temperature sensor installed in the return water pipeline monitors the change of the return water temperature in real time, and the temperature change rate is calculated by the main controller. The temperature change rate can be calculated by the differential method, by comparing the return water temperature values ​​of two consecutive samples and dividing them by the sampling interval time to obtain the change rate value. For example, if the current return water temperature is 5°C, the return water temperature was 8°C at the last sampling, and the sampling interval is 10 seconds, the change rate is -0.3°C / second.

[0154] When the return water temperature change rate is greater than the first threshold, it means that the refrigeration efficiency is high and the heat exchange process is still in the rapid cooling stage. At this time, the current ice-making power is maintained to maintain the ice-making efficiency. When the return water temperature change rate drops below the first threshold, it indicates that the ice-making efficiency begins to decline and enters a slow cooling stage, which may be due to the increase in ice thickness or the gradual saturation of heat transfer. In this case, energy consumption is reduced by reducing the ice-making power, while the ice-making time is extended to ensure the adequacy of the ice-making process. This adjustment strategy can be achieved by reducing the operating frequency of the compressor or reducing the refrigerant flow.

[0155] When the return water temperature change rate further drops to the second threshold, it can be determined that the ice making process is nearly completed, because the return water temperature change rate is close to zero at this time, indicating that the heat exchange capacity of the ice maker has reached the minimum temperature required for the ice layer. At this time, the refrigeration cycle of the ice maker should be stopped immediately to avoid unnecessary energy consumption and potential damage to the ice maker components due to over-refrigeration. After stopping refrigeration, a drainage or ice cube separation process can be triggered to ensure that there is no residual effect when the ice maker is started next time.

[0156] In this process, the setting of thresholds is crucial. The specific values ​​of the first threshold and the second threshold can be determined through experiments and dynamically adjusted according to the ambient temperature, humidity and the specific model of the ice maker. For example, in a high temperature and high humidity environment, the rate of change of the return water temperature may decrease more slowly, so the absolute values ​​of the first threshold and the second threshold need to be appropriately increased to adapt to different environmental conditions.

[0157] In order to improve the intelligence and adaptability of the system, a machine learning algorithm can be introduced to dynamically optimize the threshold. For example, based on historical ice-making data, a prediction model is trained using the relationship between the return water temperature change rate and the actual ice-making completion time to adjust the numerical range of the first threshold and the second threshold in real time. In addition, an adaptive sampling frequency can be designed to increase the sampling frequency when the change rate is large to capture more details; and reduce the sampling frequency when the change rate tends to be stable to save processing resources.

[0158] In addition, when the return water temperature change rate remains at an abnormal value (such as too low or too high) for a long time, the abnormal detection mechanism can be triggered to determine whether there is a sensor failure, refrigerant leakage or other refrigeration system problems, and send an alarm signal to the user. In addition, when the ice layer is too thick or the heat transfer performance decreases, the system can appropriately extend the power reduction stage to ensure the uniformity of ice quality.

[0159] Step S105 can dynamically adapt to different environmental conditions and operating states while optimizing the ice-making power adjustment, thereby significantly improving the efficiency and energy-saving effect of the ice-making process.

[0160] A second embodiment of the present application provides an electronic device, the electronic device comprising:

[0161] processor;

[0162] The memory is used to store a program. When the program is read and executed by the processor, the program executes an intelligent ice making amount control method for an ice maker provided in the first embodiment of the present application.

[0163] The third embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, an intelligent ice making amount control method for an ice maker provided in the first embodiment of the present application is executed.

[0164] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. An intelligent ice making quantity control method for an ice maker, characterized in that: include: Collecting operation-related data of the ice maker, specifically including obtaining operation data through installed ambient temperature sensors, humidity sensors, ice storage bin sensors and energy consumption detection modules, the operation data including ambient temperature, ambient humidity, the amount of ice remaining in the current ice storage bin and energy consumption data of the refrigeration system, and recording the time period distribution of users taking ice, the amount of ice taken at a single time, the total amount of ice used per day and the difference in ice used between working days and weekends through the storage module; Based on the operation-related data, an ice usage law model is established, and the ice usage law model is used to predict the next ice usage time point and the estimated ice quantity required; and the amount of ice to be replenished is calculated according to the remaining ice quantity in the current ice storage bin; Calculate the required ice-making time by combining the amount of ice to be replenished with the ambient temperature and humidity; Determine the optimal ice-making start time based on the expected ice-use time and required ice-making time; During the ice-making process, the ice-making power is controlled by monitoring the return water temperature change rate, specifically including maintaining the current ice-making power when the return water temperature change rate is greater than a first threshold, reducing the ice-making power to save energy when the return water temperature change rate is less than the first threshold, and stopping the ice-making process when the return water temperature change rate drops to a second threshold.

2. The intelligent ice making quantity control method of an ice maker according to claim 1, characterized in that: The collecting of the operation-related data of the ice-making machine includes: The ambient temperature and humidity data at different locations around the ice maker are collected through multi-point arranged ambient temperature sensors and humidity sensors, and the collected data are transmitted to the main control module for multi-point data fusion, and the spatial average values ​​of ambient temperature and humidity are calculated, and the areas with large fluctuations are marked for further analysis; In the ice storage bin, the infrared ranging sensor is used to dynamically monitor the distribution of the remaining ice in the ice storage bin. By sensing the ice accumulation height and coverage area, the current remaining ice volume in the ice storage bin is estimated. At the same time, the weight data of the piezoelectric sensor at the bottom of the ice storage bin is measured to further verify the accuracy of the remaining ice calculation. The energy consumption detection module collects the current, voltage and power data of the refrigeration system in real time, records the collected instantaneous data in time series and performs filtering processing to remove interference signals caused by power grid fluctuations and generate a stable energy consumption curve; Combined with the data from the ice storage bin sensor, the energy efficiency of each ice-making cycle of the ice maker is analyzed.

3. The intelligent ice making amount control method of an ice maker according to claim 2, characterized in that: The collecting of the operation-related data of the ice-making machine also includes: In the data record of user ice taking behavior, the photoelectric sensor installed at the ice taking port of the ice storage bin and the ice taking port activity switch are used to sense the triggering time of each user taking ice and the opening time of the ice taking port, so as to calculate the single ice taking amount and accurately record the ice taking time by date and time; Set the user authority grouping mode to distinguish the difference in ice usage of different users on weekdays and weekends, and store the grouping data in the historical database of the main control module for subsequent analysis.

4. The intelligent ice making amount control method of an ice maker according to claim 1, characterized in that: The ice usage regularity model includes an environment analysis module, a user behavior analysis module and a comprehensive prediction module; The environmental analysis module is used to analyze the impact of environmental conditions on ice demand. The input of the environmental analysis module includes the real-time collected environmental temperature and humidity; the output is the influencing factor of the environmental conditions; the environmental analysis module is implemented by associating the environmental temperature and humidity in the historical records with the trend of ice consumption, and using a model building method based on statistical regression, so as to obtain the dynamic impact of the current environmental conditions on ice demand; The user behavior analysis module is used to analyze the user's ice usage habits and ice usage patterns; the input of the user behavior analysis module includes the time period distribution of users taking ice, the amount of ice taken at a time, the total amount of ice used per day, and the difference in ice usage between weekdays and weekends; the output is the peak time period for users to use ice and the expected trend of ice usage; the user behavior analysis module uses a classification algorithm and a cluster analysis model based on user behavior characteristics to extract users' ice usage habits from historical data and predict users' ice demand characteristics in future time periods; The comprehensive prediction module is used to combine the output of the environmental analysis module and the output of the user behavior analysis module to generate the next ice usage time point and the predicted ice usage amount required; the input of the comprehensive prediction module is the influencing factors of environmental conditions, the peak time period of user ice usage and the predicted ice usage trend; the output is the predicted next ice usage time point and the predicted ice usage amount required; the comprehensive prediction module is implemented based on a multivariate weighted regression model, comprehensively considering the environment and user behavior characteristics, and optimizing the prediction accuracy by dynamically adjusting the weights.

5. The intelligent ice making quantity control method of an ice maker according to claim 4, characterized in that: The environmental analysis module is specifically used for: Receive the real-time collected ambient temperature and ambient humidity data, and smooth the data through a weighted sliding average algorithm to obtain stable environmental input data as the basis for subsequent steps; the weighting factor of the weighted sliding average algorithm is dynamically adjusted according to the length of the time window to ensure that the smoothed data can reflect the overall trend of environmental changes while filtering out short-term fluctuation noise; Based on the obtained smoothed data, the changing trend of ambient temperature and humidity and ice consumption in historical records is correlated, and the nonlinear mapping relationship between ambient temperature and humidity and ice consumption demand is extracted using the time series segmented regression method. According to the real-time collected ambient temperature and ambient humidity data, and in combination with the extracted nonlinear mapping relationship, the influencing factors of the environmental conditions are calculated through a support vector regression model based on a kernel function, wherein the influencing factors include a positive correction coefficient of the ambient temperature on the ice demand and a negative correction coefficient of the ambient humidity on the ice making efficiency, and the specific effects of the environmental conditions on the ice demand and the ice making time under different scenarios are quantified respectively; After each ice-making cycle is completed, the output influencing factors are compared with the measured values ​​of actual ice-making time and ice usage through multivariate cross-validation, the segmented weight mechanism and support vector regression parameters are dynamically adjusted, and the nonlinear mapping relationship and kernel function parameters of the model are updated to make it more accurately adapt to environmental changes and long-term operation status.

6. The intelligent ice making quantity control method of an ice maker according to claim 1, characterized in that: The user behavior analysis module is specifically used for: By performing cluster analysis on the time period distribution of users' ice-picking behaviors, the users' ice-picking behaviors are divided into peak hours, off-peak hours and low-peak hours based on the ice-picking times extracted from historical records. By adopting a dynamic time window adjustment mechanism, the clustering results are optimized in real time according to the changes in ice-picking behaviors in different seasons or on specific dates, and the characteristic ice usage in each time period is output. Combining the single ice taking amount and the total daily ice usage, a classification algorithm based on Gaussian mixture model is used to generate the probability distribution of users' ice usage patterns, and analyze the randomness and regularity of users' ice usage. By distinguishing the ice usage behavior characteristics on weekdays and weekends, a mapping of users' ice usage patterns under different date types is established, and ice usage prediction parameters containing date characteristics are output. By combining the clustering results of ice usage period distribution and the ice usage parameters generated by probability distribution, the impact of peak and random behaviors on ice usage demand is comprehensively considered in a weighted manner, and the user's ice usage peak time period and the expected ice usage trend are output.

7. The intelligent ice making quantity control method of an ice maker according to claim 1, characterized in that: The comprehensive prediction module is specifically used for: Combining the environmental condition influencing factors output by the environmental analysis module with the peak time period of ice usage and the expected ice usage trend output by the user behavior analysis module, a multivariate regression model based on time series is used to comprehensively calculate the next ice usage time point and the predicted required ice volume. The regression model uses the dynamic characteristics of environmental factors and the regularity of user behavior as input variables of the prediction factor. The prediction accuracy of the regression model is optimized by dynamically adjusting the weighting factors, where the weight of environmental conditions is dynamically adjusted according to the range of changes in real-time temperature and humidity; the weight of user behavior is corrected by the historical ice usage accuracy; The output results are detected for anomalies, and the historical statistical distribution of environmental and user behavior data is used to identify anomalies in the prediction results and make corrections based on the set tolerance range to ensure that the generated next ice usage time point and the predicted amount of ice required are consistent with the actual operating conditions.

8. The intelligent ice making quantity control method of an ice maker according to claim 1, characterized in that: The method of calculating the required ice making time by combining the amount of ice to be replenished with the ambient temperature and humidity includes: Use the following formula 1 to calculate the required ice making time: Where M is the amount of ice to be replenished, which is calculated by the difference between the remaining ice in the ice storage bin and the estimated amount of ice required; Q is the rated refrigeration capacity of the refrigerator, which is provided by the rated parameters of the equipment; C loss is the heat loss coefficient of the ice storage bin, which reflects the thermal insulation performance of the ice storage bin and is determined by the material and structural parameters of the ice storage bin; V storage k is the volume of ice to be added in the ice storage bin; eff is the refrigeration efficiency constant, which reflects the cooling efficiency per unit energy consumption and is obtained through experiments; ΔT system is the difference between the refrigerant evaporation temperature and the internal temperature of the ice storage bin, which is determined by the refrigeration cycle parameters of the ice maker; η(T env , H) is the environmental condition correction factor, calculated using the following formula 2: η(T env ,H)=1-k T ·(T env -T rated )-k H ·H(2) Among them, T env is the current ambient temperature, which is collected in real time by the temperature sensor; H is the current ambient humidity, which is collected in real time by the humidity sensor; T rated is the rated operating temperature of the ice machine; k T is the ambient temperature correction factor, k H is the ambient humidity correction factor, which is used to quantify the effects of temperature and humidity on refrigeration efficiency and is obtained through experimental calibration.

9. The intelligent ice making quantity control method of an ice maker according to claim 1, characterized in that: The step of determining the optimal ice-making start time point according to the expected ice-use time point and the required ice-making time includes: Use the following formula 3 to determine the optimal ice making start time t start : Among them, t pred is the estimated time point of ice usage, which is output by the ice usage law model and indicates the time point when the user will use ice next time; T is the required ice making time; T env is the current ambient temperature, which is collected in real time by the ambient temperature sensor; H is the current ambient humidity, which is collected in real time by the humidity sensor; α1 and β1 are environmental correction coefficients; Δt buffer is the time buffer, calculated by the following formula 4: Among them, E sys is the real-time energy consumption of the ice maker; T sys k is the heat exchange response time of the ice maker, which is determined by the equipment performance parameters; sys It is the system adjustment coefficient, which reflects the influence of the ice machine system status on the start-up time.

10. The intelligent ice making amount control method of an ice maker according to claim 5, characterized in that: The kernel function-based support vector regression model adopts the kernel function provided by the following formula 5: Among them, K(x, y) is the kernel function value, which is used to measure the similarity of input vectors x and y in the high-dimensional feature space; input vectors x and y represent the environmental temperature and humidity data and their corresponding ice demand data respectively; α is the Gaussian kernel strength adjustment parameter, which is used to control the rate of similarity decrease; T env is the current ambient temperature, which is collected in real time by the temperature sensor; H is the current ambient humidity, which is collected in real time by the humidity sensor; β is the ambient humidity correction coefficient; γ is the temperature compensation constant; δ is the cosine term amplitude adjustment parameter, which is used to adjust the influence of the ambient temperature and humidity difference on the similarity calculation; ∈ is a small positive constant to avoid the denominator being zero.

Images