Feed production method and system based on Internet of Things

Abstract

The invention provides a feed production method and system based on the Internet of Things. The feed production method comprises the following steps: collecting preset real-time dielectric constant change data of feed raw materials in real time; transmitting the real-time dielectric constant change data to a computing node in communication connection with the low-power wide area network and analyzing the real-time dielectric constant change data to generate a moisture characteristic value; matching the moisture characteristic value with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result; when the raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, generating a drying equipment power adjusting instruction; and the drying equipment power adjusting instruction is sent to a preset drying equipment executing mechanism so as to adjust the feed drying working power. According to the method, the hysteresis of traditional manual sampling is broken through, second-level response of moisture abnormity is achieved, feed drying power adjustment is precisely matched with the actual water content state of raw materials, the mildew risk caused by excessive drying or excessive moisture is avoided, energy consumption is remarkably reduced, and the quality stability of finished products is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of feed production, and in particular to a feed production method and system based on the Internet of Things. BACKGROUND

[0002] With the deepening application of the Internet of Things technology in the field of agriculture and animal husbandry, the feed production process is developing towards intelligence and refinement. In the feed raw material storage and processing process, the sudden change of environmental humidity will directly affect the moisture content of the raw material, and then affect the quality stability of the finished feed. Therefore, how to realize real-time sensing of the moisture change of the raw material and dynamically adjust the production process parameters according to the moisture change, especially the energy consumption control of the drying link, has become one of the key problems that the current feed production enterprises need to solve.

[0003] The current mainstream scheme is an online moisture monitoring system based on near-infrared spectrum analysis. The near-infrared sensor installed above the conveying belt performs non-contact scanning on the flowing feed raw material to obtain the moisture distribution data of the surface of the raw material. After the moisture distribution data is processed by the local industrial control computer, it is combined with the preset moisture threshold to determine whether the working state of the subsequent drying equipment needs to be adjusted. The existing scheme has some inherent defects, including that the near-infrared technology mainly reflects the moisture state of the surface layer of the raw material, which makes it difficult to accurately represent the moisture content of the whole material, and misjudgment is easy to occur; the centralized data processing mode limits the reaction speed of the system to the abnormal situation on site, and real-time regulation and control cannot be realized; the communication mode of the existing system relies on fixed wiring network, which limits the flexibility of equipment layout and increases the maintenance complexity, and it is difficult to meet the efficient operation demand of the feed production in the variable scene. SUMMARY

[0004] The present application provides a feed production method and system based on the Internet of Things, which solves the problems in the prior art that the near-infrared technology mainly reflects the moisture state of the surface layer of the raw material, which makes it difficult to accurately represent the moisture content of the whole material, and misjudgment is easy to occur; the centralized data processing mode limits the reaction speed of the system to the abnormal situation on site, and real-time regulation and control cannot be realized; the communication mode of the existing system relies on fixed wiring network, which limits the flexibility of equipment layout and increases the maintenance complexity, and it is difficult to meet the efficient operation demand of the feed production in the variable scene.

[0005] In a first aspect, the present application provides a feed production method based on the Internet of Things, comprising: real-time collecting real-time dielectric constant change data of preset feed raw material by using a preset microwave resonant cavity sensor; transmitting the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide-area network, analyzing the real-time dielectric constant change data to generate a moisture characteristic value; match the moisture characteristic value with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result; generate a drying equipment power adjustment instruction when a raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range; send the drying equipment power adjustment instruction to a preset drying equipment actuator to adjust the feed drying work power.

[0006] Optionally, real-time dielectric constant change data of the preset feed raw material is collected in real time by using a preset microwave resonant cavity sensor, including: The working frequency band of the preset microwave resonant cavity sensor is set as a raw material detection frequency band; In the raw material detection frequency band, a microwave emission source in the microwave resonant cavity sensor emits a scanning frequency signal to the preset feed raw material to generate a resonance echo signal; Calculate the phase difference value between the resonance echo signal corresponding to each discrete frequency point in the scanning frequency signal and the scanning frequency signal, combine each phase difference value to generate phase difference spectrum data; Input the phase difference spectrum data into a preset dielectric constant mapping table to generate real-time dielectric constant change data.

[0007] Optionally, the real-time dielectric constant change data is transmitted to a computing node connected with a preset low-power wide-area network, and the real-time dielectric constant change data is analyzed to generate a moisture characteristic value, including: Determine the initial moisture value corresponding to the real-time dielectric constant change data by using a preset dielectric constant-moisture mapping relationship; According to a preset time window, the initial moisture value is segmented to generate continuous data units; Input the preset timestamp data and the preset sensor location identification data into each data unit to generate a transmission data packet body; Use a reserved channel of a preset low-power wide-area network to transmit the transmission data packet body to a computing node connected with the preset low-power wide-area network to generate an initial moisture sequence; Differential operation is performed on the initial moisture sequence to generate a moisture change rate as a moisture characteristic value.

[0008] Optionally, the moisture characteristic value is matched with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result, including: Extract a sub-historical moisture sequence associated with a preset current environmental temperature from a preset raw material moisture historical fluctuation model; Taking a detection time point corresponding to the moisture characteristic value as a reference, cached moisture value data of a preset time length is extracted from a preset moisture value cache area, the cached moisture value data is arranged in a preset timestamp order to generate a current moisture sequence; A similarity value between the current moisture sequence and each historical sub-sequence in the sub-historical moisture sequence is calculated, and the historical sub-sequences with the similarity value less than a preset similarity threshold value are screened to generate a candidate matching result set; A historical sub-sequence with the latest timestamp in the candidate matching result set is extracted to generate a raw material moisture matching result.

[0009] Optionally, calculating a similarity value between the current moisture sequence and each historical sub-sequence in the sub-historical moisture sequence, and screening the historical sub-sequences with the similarity value less than a preset similarity threshold value to generate a candidate matching result set, includes: The current moisture sequence is divided into three continuous and equal-length current sub-segments, a moisture value change slope of each current sub-segment is calculated, and a current slope feature vector is generated; Each historical sub-sequence in each sub-historical moisture sequence is divided into three continuous and equal-length historical sub-segments, a moisture value change slope of each historical sub-segment is calculated, and a historical slope feature vector is generated; An absolute value difference value between each component in the current slope feature vector and a same-order component in the historical slope feature vector is calculated, and a sequence similarity value is generated by using each absolute value difference value and a temperature compensation factor corresponding to a preset current ambient temperature; The sequence similarity value is screened to obtain a plurality of target sequence similarity values less than a preset similarity threshold value; The historical sub-sequences corresponding to each target sequence similarity value are taken as matching items, and a candidate matching result set is integrated from each matching item.

[0010] Optionally, when a raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, a drying equipment power adjustment instruction is generated, including: A moisture change trend parameter is extracted from the raw material moisture matching result; A basic power adjustment amount of a preset drying equipment is determined according to the moisture change trend parameter; A current working power value is generated by analyzing a preset drying equipment working power state message; An arithmetic superposition operation is performed on the basic power adjustment amount and the current working power value to generate a target power value; A difference absolute value between the target power value and the current working power value is calculated, and the difference absolute value is taken as a power change step; Combining the target power value and the power change step length generates a drying equipment power adjustment instruction.

[0011] Optionally, the drying equipment power adjustment instruction is sent to a preset drying equipment execution mechanism to adjust the feed drying work power, including: The drying equipment power adjustment instruction is encapsulated by using a reverse control channel of the preset low-power wide-area network to generate a power regulation data packet. The power regulation data packet is parsed in the preset drying equipment execution mechanism to obtain a target power value and a power change step length. According to the power change step length, the output power value of the preset drying equipment execution mechanism is adjusted to be consistent with the target power value to generate a power adjustment completion signal.

[0012] In a second aspect, the present application provides a feed production system based on the Internet of Things, including: The acquisition module is configured to acquire real-time dielectric constant change data of the preset feed raw material in real time by using a preset microwave resonant cavity sensor. The analysis module is configured to transmit the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide-area network, analyze the real-time dielectric constant change data, and generate a moisture characteristic value. The matching module is configured to match the moisture characteristic value with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result. The generation module is configured to generate a drying equipment power adjustment instruction when a raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range. The sending module is configured to send the drying equipment power adjustment instruction to a preset drying equipment execution mechanism to adjust the feed drying work power.

[0013] In a third aspect, the present application provides a computing device including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the feed production method based on the Internet of Things according to any one of the first aspect.

[0014] In a fourth aspect, the present application provides a computer storage medium storing computer program instructions, wherein the computer program instructions are executed by a processor to implement the feed production method based on the Internet of Things according to any one of the first aspect.

[0015] The application collects the dielectric constant change data of raw materials in real time through a microwave resonant cavity sensor, analyzes the data into moisture characteristic values combined with a low-power wide area network, and matches the historical fluctuation model to generate a moisture matching result, dynamically triggers power adjustment of the drying equipment in the raw material storage humidity mutation scene, breaks through the hysteresis of traditional manual sampling, realizes second-level response of moisture anomaly by using the Internet of Things architecture, accurately adapts the feed drying power adjustment to the actual moisture content of the raw materials, avoids the risk of mold caused by excessive drying or moisture exceeding the standard, significantly reduces energy consumption and ensures the stability of the finished product quality.

[0016] Further, by setting a raw material exclusive detection frequency band to emit a scanning frequency signal, the phase difference between the resonant echo and the emitted signal is captured to generate frequency spectrum data, which is mapped into dielectric constant change values, eliminating the influence of environmental electromagnetic interference on moisture detection, and using the physical correlation between dielectric constant and moisture content, deep layer moisture penetration monitoring of raw materials is realized in the humidity mutation scene, solving the response delay problem of traditional contact sensors, and providing a high-precision data basis for dynamic power regulation.

[0017] These and other aspects of the present application will become more fully understood from the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 A flowchart of a feed production method based on the Internet of Things provided by the embodiment of the present application is shown in the figure. Figure 2 A structural schematic diagram of a feed production system based on the Internet of Things provided by the embodiment of the present application is shown in the figure. Figure 3 A structural schematic diagram of a computing device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0020] In order to make the person skilled in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0021] In some of the flowcharts described in the description and claims of the present application and in the above figures, there are included more operations than those expressly shown in a flowchart. However, it will be understood by those skilled in the art that the operations of the flowcharts can be executed in sequence or in parallel, and the serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, the flowcharts can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this paper are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do "first" and "second" represent different types.

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0023] Figure 1 A flowchart of a feed production method based on the Internet of Things is provided for the embodiments of the present application, as shown in Figure 1 , which comprises: The existing feed production faces serious technical bottlenecks in the raw material storage environment humidity mutation scene: the traditional manual sampling method has a detection lag of several hours, which cannot capture sudden humidity fluctuations; the fixed threshold sensor is easily disturbed by the environment temperature, leading to water misjudgment; the independent control of the drying equipment lacks dynamic linkage with the moisture content of the raw materials, often causing excessive drying or mold risk. To solve these problems, the research and development idea of the present application is to build a closed-loop control system driven by the Internet of Things, to penetrate the real-time changes of the dielectric constant of the raw materials through non-contact microwave resonant cavity sensing technology, to overcome the transmission delay in the complex warehouse environment by using a low-power wide-area network, to combine the historical model of the moisture absorption characteristics of the raw materials after converting the physical quantity into a moisture characteristic value at the computing node, to make a trend prediction, and finally to dynamically generate a drying power instruction according to the moisture state deviation, to realize a second-level response closed loop from raw material moisture abnormality sensing to accurate adjustment of the drying power, and to completely solve the industry pain points of quality loss of control and energy waste in the humidity mutation scene. Based on this, the present application provides a feed production method based on the Internet of Things, as shown in Figure 1 , comprising: Step 101: Real-time acquisition of real-time dielectric constant change data of the preset feed raw materials by using a preset microwave resonant cavity sensor.

[0024] In this step, the real-time dielectric constant change data refers to the physical quantity data generated by detecting the dynamic change of the dielectric properties of the feed raw material through the microwave resonant cavity sensor, reflecting the electromagnetic wave phase shift caused by the moisture absorption of the raw material.

[0025] In the embodiment of the present application, first, the preset microwave resonant cavity sensor sets its working frequency band to the raw material detection frequency band, second, controls the microwave emission source to emit a stepwise increasing scanning frequency signal to the feed raw material in this frequency band, then receives the resonant echo signal generated after penetrating the raw material, then calculates the phase difference value of the resonant echo signal corresponding to each discrete frequency point in the scanning frequency signal and the original emission signal, and finally combines the phase difference values of all discrete frequency points to form phase difference spectrum data and input the preset dielectric constant mapping table, and output the real-time dielectric constant change data.

[0026] Step 102: transmit the real-time dielectric constant change data to the computing node connected with the preset low-power wide-area network, analyze the real-time dielectric constant change data, and generate a moisture characteristic value.

[0027] In this step, the analysis operation refers to the process of moisture feature extraction of the dielectric constant change data at the computing node, including data unit segmentation, space-time identifier encapsulation, sequence reorganization and difference operation; the moisture characteristic value refers to a quantitative index representing the change trend of the moisture content of the raw material, which is obtained based on the change rate value of the moisture sequence difference operation.

[0028] In the embodiment of the present application, first, the preset dielectric constant-moisture mapping relationship is used to convert the real-time dielectric constant change data into an initial moisture value, second, the initial moisture value is segmented into continuous data units according to the preset time window, then a data packet body is generated by adding header information containing a timestamp and a sensor location identifier to each data unit, then the data packet body is transmitted to the computing node through the reserved channel of the low-power wide-area network, and finally the header is removed at the computing node and the data units are reorganized according to the timestamp, and the difference operation is performed on the reorganized initial moisture sequence to output the moisture characteristic value.

[0029] Step 103: match the moisture characteristic value with the preset raw material moisture historical fluctuation model to generate a raw material moisture matching result.

[0030] In this step, the preset raw material moisture historical fluctuation model refers to a time series moisture data set stored in a temperature sub-database constructed according to long-term monitoring data, containing the complete fluctuation law of the moisture absorption rate of the raw material under different temperature and humidity conditions; the matching operation refers to the process of similarity comparison between the real-time moisture feature and the historical model sub-sequence, which is realized through slope feature vectorization and temperature compensation distance calculation; the raw material moisture matching result refers to the historical sub-sequence closest to the current moisture change selected from the historical model, containing the moisture value and the change trend parameter.

[0031] In the embodiment of the present application, firstly, a sub-historical moisture sequence associated with the current ambient temperature is extracted from the raw material moisture history fluctuation model, secondly, a current moisture sequence is generated by extracting a preset length of historical moisture values from the pre-stored cache area based on the detection time point corresponding to the moisture characteristic value, then the current moisture sequence is divided into three continuous equal-length sub-sequences and the moisture value change slope of each sub-sequence is calculated to generate a current slope characteristic vector, next, the same operation is synchronously performed on each historical sub-sequence in the sub-historical moisture sequence to generate a historical slope characteristic vector, finally, the Manhattan distance with temperature compensation weight of the current and historical slope characteristic vectors is calculated as a similarity value, the historical sub-sequences with similarity values less than a threshold value are screened to generate a candidate matching result set, and the historical sub-sequence with the latest time stamp is extracted as the raw material moisture matching result.

[0032] Step 104: When the raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, a drying equipment power adjustment instruction is generated.

[0033] In this step, the raw material moisture value refers to the measured value of the moisture content at the end of the historical sub-sequence in the matching result, which serves as a reference benchmark for the current raw material moisture content; the preset moisture value range refers to a safe threshold interval of the moisture content set according to the safe storage requirements of the feed raw material; the drying equipment power adjustment instruction refers to a control command containing a target power value and a change step, which is used to guide the actuator to adjust the output power in stages.

[0034] In the embodiment of the present application, firstly, the average value of the moisture change rate of the last five data points in the raw material moisture matching result is extracted as a moisture change trend parameter, secondly, the basis power adjustment amount is determined from the pre-set mapping table according to the positive and negative nature of the trend parameter, then the working power state message periodically sent by the drying equipment is received to parse the current working power value, next, the basis power adjustment amount and the current working power value are arithmetically superimposed to generate a target power value, finally, the absolute value of the difference between the target power value and the current working power value is calculated as a power change step, and the target power value and the power change step are combined to generate a power adjustment instruction.

[0035] Step 105: The drying equipment power adjustment instruction is sent to a preset drying equipment actuator to adjust the feed drying working power.

[0036] In this step, the adjustment operation refers to the action process of the drying equipment actuator gradually changing the output power by the power change step.

[0037] In the embodiment of the present application, first, the power adjustment instruction is packaged into a power regulation data packet through the reverse control channel of the low-power wide-area network at the computing node, second, the data packet is sent to the drying equipment actuator, then the target power value and the power change step are extracted by parsing the data packet in the actuator, next, the output power of the actuator is gradually adjusted in a stepwise increasing or decreasing manner according to the power change step, and finally, the power adjustment completion signal is triggered when the output power stabilizes at the target power value.

[0038] For example, first, the microwave resonance cavity sensor deployed in the feed raw material warehouse transmits a scanning frequency signal in a preset detection frequency band, collects raw material dielectric constant change data and transmits it to the cloud computing node; second, the computing node parses the data into moisture characteristic values and extracts the current temperature-related sub-historical moisture sequence from the historical model library for similarity matching to generate a raw material moisture matching result; third, when the moisture value in the matching result exceeds the safety threshold, the target power and change step instructions of the drying equipment are generated according to the moisture change trend; fourth, the instructions are sent to the drying equipment actuator through the low-power wide-area network reverse channel; and finally, the actuator adjusts the output power in a stepwise manner according to the power change step until it meets the standard, completing the dynamic regulation and control loop of the feed drying process.

[0039] The embodiment of the present application captures the change of raw material dielectric properties in real time through the microwave resonance cavity sensor, transmits it to the computing node for analysis into moisture characteristic values, and generates a moisture state prediction result by matching with the historical fluctuation model. When the moisture value is detected to exceed the safety range, a power adjustment instruction is dynamically generated, and finally the drying power is adjusted by the actuator in stages. This process realizes a second-level response of raw material moisture abnormality in the scenario of sudden change of warehouse humidity, breaks through the traditional lagging nature of manual detection, avoids the risk of over-drying or mold, significantly reduces energy consumption, and guarantees the stability of feed quality.

[0040] In order to solve the problem of insufficient accuracy of raw material moisture detection due to environmental electromagnetic interference, this step improves the noise immunity of dielectric constant data by customizing frequency band scanning and phase difference spectrum mapping. The present application provides a specific embodiment, step 101, which uses a preset microwave resonance cavity sensor to collect real-time dielectric constant change data of a preset feed raw material in real time, specifically including the following steps: Step 111: Set the working frequency band of the preset microwave resonance cavity sensor to the raw material detection frequency band.

[0041] In this step, the working frequency band refers to the electromagnetic wave frequency range enabled by the microwave resonance cavity sensor when performing detection tasks, which is set according to the resonance characteristics of the feed raw material; the raw material detection frequency band refers to the electromagnetic wave frequency band optimized for a specific feed raw material, covering the core frequency band of the moisture sensitive resonance peak of the raw material.

[0042] In the embodiment of the present application, first, the working frequency band of the microwave resonant cavity sensor is configured as a special detection frequency band suitable for the characteristics of the feed raw material, and the range of the frequency band is determined according to the preset response curve of the dielectric properties of the raw material.

[0043] Step 112: In the raw material detection frequency band, a scanning frequency signal is emitted to the preset feed raw material by using a microwave emission source in the microwave resonant cavity sensor, and a resonant echo signal is generated.

[0044] In this step, the microwave emission source refers to an electromagnetic wave generating device integrated in the sensor, which is used to generate a microwave signal with a controllable frequency; the scanning frequency signal refers to a sequence of electromagnetic waves with a frequency that changes continuously according to a preset step, which is used to excite the dielectric response of the raw material; and the resonant echo signal refers to a reflected electromagnetic wave carrying information about the dielectric properties of the raw material after penetrating the feed raw material, and the amplitude and phase of the resonant echo signal are modulated by the water content of the raw material.

[0045] In the embodiment of the present application, first, the microwave emission source is started in the raw material detection frequency band, then the output frequency of the emission source is controlled to continuously increase the scanning signal that penetrates the feed raw material, and then the resonant echo signal formed by the reflection and transmission inside the raw material is received.

[0046] Step 113: Calculate the phase difference between the resonant echo signal and the scanning frequency signal corresponding to each discrete frequency point in the scanning frequency signal, combine each phase difference value to generate phase difference spectrum data.

[0047] In this step, the discrete frequency point refers to an independent frequency value selected at a fixed interval in the scanning frequency signal, which is used as a reference sampling point for phase difference calculation; the phase difference value refers to the phase angle offset between the emitted signal and the echo signal at the same discrete frequency point, which reflects the propagation delay of the electromagnetic wave in the raw material; the combination operation refers to the process of integrating the phase difference values of each discrete frequency point in frequency order to form a continuous frequency spectrum, which constructs a frequency-phase relationship map; and the phase difference spectrum data refers to a two-dimensional data set representing the phase shift rule of electromagnetic waves at different frequencies in the raw material, which includes the frequency dimension and the phase difference value dimension.

[0048] In the embodiment of the present application, first, the discrete frequency points in the scanning frequency signal are extracted in a stepwise increasing manner, then the phase offset between the resonant echo signal and the original emitted signal at each discrete frequency point is calculated, and then the phase offset values of all discrete frequency points are arranged and combined in frequency order to form the phase difference spectrum data.

[0049] Step 114: Input the phase difference spectrum data into a preset dielectric constant mapping table to generate real-time dielectric constant change data.

[0050] In this step, the preset dielectric constant mapping table refers to a phase difference spectrum and dielectric constant corresponding relationship database calibrated through experiments, and is used for converting the spectrum features into physical parameters.

[0051] In the embodiment of the present application, the phase difference spectrum data is first input into the preset dielectric constant mapping table for matching query, and then the dielectric constant change data reflecting the real-time water content state of the raw material is output according to the corresponding relationship between the spectrum features and the dielectric constant recorded in the mapping table.

[0052] The embodiment of the present application excites the dielectric response of the raw material by customizing the frequency band scanning, captures the phase offset of multiple frequency points to form spectrum features, converts the spectrum features into dielectric constant change data by combining the pre-calibrated mapping relationship, realizes the penetration type moisture monitoring, overcomes the environmental temperature and humidity interference, and provides a high-precision physical quantity basis for subsequent moisture control.

[0053] In order to improve the space-time correlation and analysis efficiency of the moisture characteristic value in the transmission process, the data unit encapsulation and differential operation are used to realize the real-time extraction of the change rate. The present application provides a specific embodiment, step 102, transmitting the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide area network, analyzing the real-time dielectric constant change data, and generating a moisture characteristic value, specifically including the following steps: Step 201: determining the initial moisture value corresponding to the real-time dielectric constant change data by using the preset dielectric constant-moisture mapping relationship.

[0054] In this step, the preset dielectric constant-moisture mapping relationship refers to a dielectric constant and moisture content corresponding rule database calibrated through experiments, and is used for converting physical quantities into moisture values; the initial moisture value refers to the basic water content numerical value of the raw material based on the dielectric constant mapping relationship, reflecting the instantaneous moisture state.

[0055] In the embodiment of the present application, the preset dielectric constant-moisture mapping relationship database is first called, the real-time dielectric constant change data is then input into the database for matching query, and subsequently the initial moisture value is output according to the dielectric constant and moisture content corresponding rule in the mapping relationship.

[0056] Step 202: dividing the initial moisture value according to a preset time window to generate continuous data units.

[0057] In this step, the preset time window refers to a time unit for dividing data according to a fixed time length, and is used for segmenting the continuous moisture values; the division operation refers to the action of cutting the continuous moisture value sequence according to the time window boundary to generate data segments with equal time length; the data unit refers to an independent data block containing the moisture values in a single time window, which is used as the minimum transmission unit.

[0058] In the embodiment of the present application, first, a fixed time length time window is set, second, the continuously input initial moisture values are cut into equal time length data segments according to the time window boundaries, and then each data segment is encapsulated into an independent data unit.

[0059] Step 203: inputting preset timestamp data and preset sensor position identification data into each data unit to generate a transmission data packet body; In this step, the preset timestamp data refers to the accurate time mark data generated by the system clock, indicating the time when the data is generated; the preset sensor position identification data refers to the unique position code burned in the sensor chip, used for spatial positioning; and the transmission data packet body refers to the network transmission entity composed of the data unit, timestamp, position identification and check code.

[0060] In the embodiment of the present application, first, the timestamp data generated by the system unified time service and the position identification data burned in the sensor are acquired, second, the timestamp data and the position identification data are injected into the header area of each data unit, and then a check code is added to generate a complete transmission data packet body.

[0061] Step 204: transmitting the transmission data packet body to a computing node in communication connection with the preset low-power wide-area network by using a reserved channel of the preset low-power wide-area network, to generate an initial moisture sequence.

[0062] In this step, the reserved channel refers to the exclusive communication channel allocated for key data in the low-power wide-area network, guaranteeing the transmission priority; and the initial moisture sequence refers to the set of continuous moisture values sorted by time after reorganization in the computing node.

[0063] In the embodiment of the present application, first, the exclusive transmission channel reserved for moisture data in the low-power wide-area network is activated, second, the transmission data packet body is sent to the computing node through the channel, and then the header is stripped off in the computing node and the data units are reorganized in the order of timestamp to form a continuous initial moisture sequence.

[0064] Step 205: performing a difference operation on the initial moisture sequence to generate a moisture change rate as a moisture feature value.

[0065] In this step, the difference operation refers to the mathematical operation of calculating the change amount of the moisture value of adjacent data points in the sequence, revealing the change trend; and the moisture change rate refers to the increment or decrement of the moisture value per unit time, representing the dynamic change speed of the water content.

[0066] In the embodiment of the present application, first, each data point in the initial moisture sequence is traversed in time order, second, the difference value of the moisture values of adjacent data points is calculated, and then the difference value is divided by the time interval between the previous and subsequent data points, and the moisture change rate per unit time is output as the moisture feature value.

[0067] The embodiment of the present application realizes the conversion of physical quantity to moisture value through dielectric constant mapping, guarantees data transmission integrity through time window segmentation and space-time identification packaging, reorganizes into a continuous sequence through exclusive channel transmission to a computing node, and finally extracts moisture change rate characteristics through differential operation, thereby providing high timeliness trend analysis basis for humidity mutation scenarios.

[0068] In order to solve the defects of historical model matching lag and high misjudgment rate in the humidity mutation scenario, the matching accuracy is improved through the temperature correlation sub-sequence and timeliness screening mechanism in this step. The present application provides a specific embodiment, step 103, matching the moisture characteristic value with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result, specifically including the following steps: Step 301: Extracting a sub-historical moisture sequence associated with the preset current environmental temperature from the preset raw material moisture historical fluctuation model.

[0069] In this step, the extraction operation refers to the action of retrieving the target data set from the structured data storage, including index query and partition positioning; the sub-historical moisture sequence refers to the moisture time series data set in the historical fluctuation model strictly associated with the current temperature, reflecting the moisture absorption law under specific temperature and humidity conditions.

[0070] In the embodiment of the present application, first, the temperature index library of the raw material moisture historical fluctuation model is queried, second, the associated temperature partition is located according to the current environmental temperature value, and then the complete historical moisture time series data set is extracted from the partition as the sub-historical moisture sequence.

[0071] Step 302: Taking the detection time point corresponding to the moisture characteristic value as the reference, extracting the cache moisture value data of the preset time length from the preset moisture value cache area, arranging the cache moisture value data in the preset timestamp order to generate the current moisture sequence.

[0072] In this step, the preset moisture value cache area refers to a ring-shaped data buffer area for temporarily storing recent moisture characteristic values, used for quickly building time series context; the cache moisture value data refers to the set of moisture characteristic values recorded in the cache area in time sequence, serving as a short-term historical reference for real-time analysis; the preset timestamp order refers to the rule of strictly arranging data according to time sequence, ensuring the time continuity of the sequence; the current moisture sequence refers to the ordered set of moisture values containing historical context built based on the detection time point.

[0073] In the embodiment of the present application, first, the detection time point of the moisture characteristic value is taken as the time coordinate origin, second, the historical interval of the preset time length is traced back, then all the cache moisture value data in the interval is extracted from the moisture value cache area, and finally the current moisture sequence is generated by arranging the timestamp in the order from early to late.

[0074] Step 303: Calculate the similarity values between the current moisture sequence and each historical sub-sequence in the sub-historical moisture sequence, and screen out the historical sub-sequences with similarity values less than a preset similarity threshold to generate a candidate matching result set.

[0075] In this step, the similarity value refers to an index quantifying the consistency of the change trend of the current sequence and the historical sub-sequence, and the smaller the value is, the higher the similarity is; the preset similarity threshold refers to a maximum similarity critical value for determining the availability of the historical sub-sequence, which is used to filter abnormal fluctuation data; and the candidate matching result set refers to a qualified historical sub-sequence set screened by the similarity, which is used as a candidate pool for the final matching.

[0076] In the embodiment of the present application, the current moisture sequence is first divided into three equal-length sub-segments to calculate the slopes of each segment and generate a feature vector, then the same operation is performed on each historical sub-sequence in the sub-historical moisture sequence to generate a historical feature vector, then the absolute value difference of each component of the current and historical feature vectors is calculated and summed, and finally the sum is multiplied by a temperature compensation factor to output the similarity value, and the historical sub-sequences with similarity values less than the preset threshold are screened to generate the candidate matching result set.

[0077] Step 304: Extract the historical sub-sequence with the latest timestamp from the candidate matching result set to generate the raw material moisture matching result.

[0078] In this step, the latest timestamp refers to the historical sub-sequence in the candidate set with the time marker closest to the current time, which represents the most similar working condition.

[0079] In the embodiment of the present application, the timestamps of all historical sub-sequences in the candidate matching result set are first traversed, then the order of each timestamp is compared, and finally the historical sub-sequence closest to the current time is selected as the raw material moisture matching result.

[0080] The embodiment of the present application extracts historical sub-sequences through a temperature correlation mechanism, constructs a moisture sequence containing context by combining real-time cache data, screens a candidate set through feature vectorization and temperature-compensated similarity calculation, finally selects the most optimal matching result in terms of timeliness, and realizes the accurate mapping of historical experience and real-time state in the humidity mutation scenario, thereby providing a reliable decision basis for drying regulation.

[0081] In order to break through the bottleneck of high calculation complexity and ignoring temperature micro-changes in traditional sequence matching, this step optimizes the matching efficiency through three-segment slope features and temperature-compensated similarity calculation. The present application provides a specific embodiment, step 303, which calculates the similarity values between the current moisture sequence and each historical sub-sequence in the sub-historical moisture sequence, screens out the historical sub-sequences with similarity values less than a preset similarity threshold, and generates a candidate matching result set, which specifically includes the following steps: Step 331: dividing the current moisture sequence into three continuous equal-length current sub-segments, calculating the moisture value change slope of each current sub-segment, and generating a current slope feature vector.

[0082] In this step, the current sub-segment refers to the continuous time period after the current moisture sequence is evenly divided, which is used for segment calculation of moisture change trend; the moisture value change slope refers to the increment or decrement of moisture value per unit time, which is obtained by dividing the difference between the end point moisture value and the start point moisture value by the time length; and the current slope feature vector refers to an array composed of the slope values of the three current sub-segments in time sequence, which represents the three-stage characteristics of the current moisture change.

[0083] In the embodiment of the present application, firstly, the current moisture sequence is evenly divided into three continuous time periods as current sub-segments, secondly, the difference between the start point and the end point of the moisture value in each current sub-segment is divided by the time length to obtain the moisture value change slope, and finally, the slope values of the three sub-segments are combined in time sequence into a current slope feature vector.

[0084] Step 332: dividing each historical sub-sequence in each of the sub-historical moisture sequences into three continuous equal-length historical sub-segments, calculating the moisture value change slope of each historical sub-segment, and generating a historical slope feature vector.

[0085] In this step, the historical sub-segment refers to the continuous time period after the historical sub-sequence is evenly divided, which is used for extracting historical change characteristics; and the historical slope feature vector refers to an array composed of the slope values of the three sub-segments of a single historical sub-sequence in sequence, which records the historical moisture change mode.

[0086] In the embodiment of the present application, firstly, each historical sub-sequence in the sub-historical moisture sequence is evenly divided into three continuous time periods as historical sub-segments, secondly, the moisture value change slope of each historical sub-segment is calculated, and finally, the three slope values of each historical sub-sequence are combined in time sequence into a corresponding historical slope feature vector.

[0087] Step 333: calculating the absolute value difference between each component in the current slope feature vector and the same sequence component in the historical slope feature vector, and generating a sequence similarity value by using each absolute value difference and a temperature compensation factor corresponding to the current ambient temperature.

[0088] In this step, the same sequence component refers to the slope value in the same time sequence in the feature vector, such as the first sub-segment slope being the first component; the absolute value difference refers to the absolute value of the difference between the same sequence component values of two feature vectors, which reflects the local trend difference; the temperature compensation factor refers to a weight coefficient set according to the current ambient temperature, which is used to correct the influence of temperature difference on similarity calculation; and the sequence similarity value refers to a comprehensive similarity index obtained by weighted summation of the three component difference values, and the smaller the value is, the higher the similarity is.

[0089] In the embodiment of the present application, first, the absolute value difference of the first component of the current slope feature vector and the historical slope feature vector is calculated, second, the absolute value difference of the second component and the third component is also calculated, then the three absolute value differences are added and multiplied by the temperature compensation factor corresponding to the current ambient temperature, and the sequence similarity value of the historical subsequence is output.

[0090] Step 334: screening the sequence similarity values to obtain a plurality of target sequence similarity values less than a preset similarity threshold.

[0091] In this step, the target sequence similarity value refers to the similarity value set that meets the matching condition after threshold screening.

[0092] In the embodiment of the present application, first, a preset similarity threshold is set, second, all sequence similarity values generated in step 333 are compared with the threshold, and finally a plurality of target sequence similarity values less than the threshold are screened out.

[0093] Step 335: taking the historical subsequence corresponding to each target sequence similarity value as a matching item, and integrating each matching item to obtain a candidate matching result set.

[0094] In this step, the matching item refers to the qualified historical subsequence corresponding to the target sequence similarity value as the candidate matching element; the integration operation refers to the process of combining the dispersed matching items into a unified data set to generate a structured candidate pool.

[0095] In the embodiment of the present application, first, the corresponding historical subsequence is located according to the target sequence similarity value, second, these historical subsequences are marked as matching items, and finally all matching items are aggregated into a candidate matching result set.

[0096] The embodiment of the present application quantifies the moisture change rule by three-stage slope feature extraction, calculates the historical sequence similarity in combination with the temperature compensation mechanism, screens the qualified matching items through the threshold, and integrates them into a candidate set, realizes the accurate comparison between real-time data and historical patterns in the humidity mutation scene, and provides a strong timeliness trend reference for drying decision.

[0097] In order to solve the problem of energy waste caused by the disconnection between drying power regulation and raw material moisture change trend, this step realizes dynamic power adaptation through trend parameter mapping and step grading mechanism. The present application provides a specific embodiment, step 104, when the raw material moisture value in the raw material moisture matching result exceeds the preset moisture value range, a drying equipment power regulation instruction is generated, which specifically includes the following steps: Step 401: extracting the moisture change trend parameter from the raw material moisture matching result.

[0098] In this step, the moisture change trend parameter refers to the average value of the moisture change amount of the last data points in the raw material moisture matching result, and reflects a quantitative index of the moisture change trend.

[0099] In the embodiment of the present application, first, the moisture value change amount of the last several continuous data points in the raw material moisture matching result is analyzed, second, the average value of the change amount is calculated, and finally, the moisture change trend parameter is output.

[0100] Step 402: determining the basic power adjustment amount of the preset drying equipment according to the moisture change trend parameter.

[0101] In this step, the basic power adjustment amount refers to the power reference adjustment value obtained from the preset mapping table according to the moisture change trend, and the positive and negative signs represent the power increase direction and the power decrease direction.

[0102] In the embodiment of the present application, first, the preset trend parameter-adjustment mapping table is queried, second, the corresponding interval is located according to the positive and negative nature and the absolute value size of the moisture change trend parameter, and finally, the basic power adjustment amount is extracted from the mapping table.

[0103] Step 403: analyzing the working power state message of the preset drying equipment to generate the current working power value.

[0104] In this step, the working power state message refers to the network transmission message containing real-time power data sent by the drying equipment at regular intervals, and the format conforms to the Internet of Things communication protocol; the analysis operation refers to the action of extracting the effective power data from the specific field of the message, including message decoding and value conversion; the current working power value refers to the real-time output power value of the device obtained by analyzing the working power state message.

[0105] In the embodiment of the present application, first, the working power state message sent periodically by the drying equipment is received, second, the power value field in the message payload area is extracted, and finally, it is converted into the current working power value.

[0106] Step 404: performing arithmetic superposition operation on the basic power adjustment amount and the current working power value to generate the target power value.

[0107] In this step, the arithmetic superposition operation refers to the algebraic addition of the basic power adjustment amount and the current working power value; the target power value refers to the expected output power final value of the drying equipment after the superposition operation.

[0108] In the embodiment of the present application, first, the numerical attribute of the basic power adjustment amount is read, second, it is judged whether the adjustment amount is positive or negative, then the algebraic addition operation is performed on the current working power value, and finally, the target power value is output.

[0109] Step 405: calculating the absolute value of the difference between the target power value and the current working power value as a power change step.

[0110] In this step, the absolute value of the difference between the target power value and the current working power value is a non-negative value representing the power adjustment amplitude; the power change step is a single power adjustment amount quantized by the absolute value, used to control the power ramping rate.

[0111] In the embodiment of the present application, first, the algebraic difference between the target power value and the current working power value is obtained, second, the absolute value of the algebraic difference is calculated, and finally the absolute value result is defined as the power change step.

[0112] Step 406: combining the target power value and the power change step to generate a drying equipment power adjustment instruction.

[0113] In this step, the combination operation refers to the action of integrating the target power value and the power change step into a structured control instruction.

[0114] In the embodiment of the present application, first, an empty instruction structure containing the target power value and the power change step is created, second, the target power value is filled into the first field of the structure, then the power change step is filled into the second field, and finally the drying equipment power adjustment instruction is generated.

[0115] The embodiment of the present application extracts the moisture change trend to map the basic power adjustment amount, combines the real-time power state of the equipment to calculate the target value and the change step, and finally generates a structured control instruction, which realizes the dynamic adaptation of the drying power and the moisture content of the raw materials, effectively avoids the power step impact in the humidity mutation scene, and ensures the safe operation of the equipment.

[0116] In order to overcome the unreliable transmission of wide area network control instructions and the risk of power step impact of the equipment, this step ensures smooth power transition through reverse channel encapsulation and step-by-step adjustment. The present application provides a specific embodiment, step 105, sending the drying equipment power adjustment instruction to a preset drying equipment actuator to adjust the feed drying working power, which specifically includes the following steps: Step 501: encapsulating the drying equipment power adjustment instruction using the reverse control channel of the preset low-power wide area network to generate a power control data packet.

[0117] In this step, the reverse control channel refers to the physical communication channel in the low-power wide area network that is specifically used for downlink transmission of control instructions, which is isolated from the uplink data channel to ensure control priority; the encapsulation processing refers to the action of packing the control instruction into a data frame according to the communication protocol format, including adding a frame header, payload data and a check code.

[0118] In the embodiment of the present application, firstly, the low-power wide-area network is called for a physical channel dedicated to downlink control, namely, a reverse control channel; secondly, the drying equipment power adjustment instruction is encapsulated into a data frame according to a preset Internet of Things communication protocol format; and finally, a check code is added to generate a complete power regulation data packet.

[0119] Step 502: The power regulation data packet is parsed in a preset drying equipment execution mechanism to obtain a target power value and a power change step.

[0120] In this step, the power regulation data packet refers to a structured network transmission unit containing control parameters such as a target power value and a power change step, which conforms to the Internet of Things communication protocol specification; the parsing operation refers to the process of separating and converting control parameters from the data packet, including protocol unpacking and data format conversion.

[0121] In the embodiment of the present application, firstly, the power regulation data packet is received through the network interface of the drying equipment execution mechanism; secondly, the protocol header of the data packet is stripped to obtain the payload; and finally, the target power value field and the power change step field in the payload are parsed and converted into control parameters.

[0122] Step 503: The output power value of the preset drying equipment execution mechanism is adjusted to be consistent with the target power value according to the power change step, to generate a power adjustment completion signal.

[0123] In this step, the output power value refers to the amount of heat energy output by the drying equipment execution mechanism to the feed raw materials, measured in power units; the adjustment operation refers to the action process of gradually changing the output power value by the power change step, realizing smooth transition of power; and the power adjustment completion signal refers to a state confirmation signal triggered when the output power value reaches the target power value, marking the end of the regulation process.

[0124] In the embodiment of the present application, firstly, the value of the power change step is read; secondly, the output power value of the execution mechanism is adjusted in stages with the step as the unit; then, the change state of the output power value is monitored in real time; and finally, the power adjustment completion signal is triggered when the output power value is completely consistent with the target power value.

[0125] The embodiment of the present application guarantees reliable transmission of regulation instructions through a special reverse channel, adjusts the output power in stages according to the control parameters accurately parsed at the execution mechanism end, and finally triggers a completion signal to form a closed-loop control, realizing safe, smooth, and verifiable adjustment of drying power in a humidity mutation scenario.

[0126] Figure 2 A structural schematic diagram of a feed production system based on the Internet of Things is provided for the embodiment of the present application, as shown in Figure 2 The system comprises: The collection module 21 is used for collecting real-time dielectric constant change data of the preset feed raw material in real time by using a preset microwave resonant cavity sensor. The analysis module 22 is used for transmitting the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide-area network, analyzing the real-time dielectric constant change data, and generating a moisture characteristic value. The matching module 23 is used for matching the moisture characteristic value with a preset raw material moisture historical fluctuation model, and generating a raw material moisture matching result. The generation module 24 is used for generating a drying equipment power adjustment instruction when a raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range. The sending module 25 is used for sending the drying equipment power adjustment instruction to a preset drying equipment execution mechanism to adjust a feed drying working power.

[0127] Figure 2 The feed production system based on the Internet of Things can perform the following functions Figure 1 The feed production method based on the Internet of Things has the implementation principle and technical effects which will not be repeated. The specific operation manner of each module and unit of the feed production system based on the Internet of Things in the above embodiment has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0128] In one possible design, Figure 2 The feed production system based on the Internet of Things can be implemented as a computing device, such as a server. Figure 3 As shown in the figure, the computing device can include a storage component 31 and a processing component 32. The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0129] The processing component 32 is configured to collect real-time dielectric constant change data of the preset feed raw material in real time by using a preset microwave resonant cavity sensor, transmit the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide-area network, analyze the real-time dielectric constant change data, generate a moisture characteristic value, match the moisture characteristic value with a preset raw material moisture historical fluctuation model, generate a raw material moisture matching result, generate a drying equipment power adjustment instruction when a raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, and send the drying equipment power adjustment instruction to a preset drying equipment execution mechanism to adjust a feed drying working power.

[0130] The processing component 32 can include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components, for executing the above method.

[0131] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage devices or their combinations, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0132] Of course, the computing device can also include other components, such as input / output interfaces, display components, communication components, etc.

[0133] The input / output interface provides an interface between the processing component and peripheral interface modules, which can be output devices, input devices, etc.

[0134] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0135] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform, and the computing device can be a cloud server, and the processing component, the storage component, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0136] The embodiment of the application also provides a computer storage medium, which stores a computer program, and the computer program can implement the above-mentioned Figure 1 The embodiment shown in the figure is a feed production method based on the Internet of Things.

[0137] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be described here.

[0138] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0139] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0140] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An IoT-based feed production method, characterized by, The application relates to a microwave resonant cavity sensor-based real-time feed moisture content monitoring method. Real-time dielectric constant change data of preset feed raw materials are collected in real time by using a preset microwave resonant cavity sensor; The real-time dielectric constant change data are transmitted to a computing node connected with a preset low-power wide-area network, the real-time dielectric constant change data are analyzed, and a moisture characteristic value is generated; The moisture characteristic value is matched with a preset raw material moisture historical fluctuation model, and a raw material moisture matching result is generated; When the raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, a drying equipment power adjustment instruction is generated; The drying equipment power adjustment instruction is sent to a preset drying equipment actuator to adjust the feed drying working power.

2. The method of claim 1, wherein, Real-time dielectric constant change data of preset feed raw materials are collected in real time by using a preset microwave resonant cavity sensor, which comprises the following steps: The working frequency band of the preset microwave resonant cavity sensor is set as a raw material detection frequency band; In the raw material detection frequency band, a microwave emission source in the microwave resonant cavity sensor emits a scanning frequency signal to the preset feed raw materials, and a resonant echo signal is generated; The phase difference values between the resonant echo signals corresponding to each discrete frequency point in the scanning frequency signal and the scanning frequency signal are calculated, each phase difference value is combined, and phase difference spectrum data are generated; The phase difference spectrum data are input into a preset dielectric constant mapping table to generate real-time dielectric constant change data.

3. The method of claim 1, wherein, The real-time dielectric constant change data are transmitted to a computing node connected with a preset low-power wide-area network, the real-time dielectric constant change data are analyzed, and a moisture characteristic value is generated, which comprises the following steps: The initial moisture value corresponding to the real-time dielectric constant change data is determined by using a preset dielectric constant-moisture mapping relationship; The initial moisture value is segmented according to a preset time window to generate continuous data units; Preset timestamp data and preset sensor position identification data are input into each data unit to generate a transmission data packet body; The transmission data packet body is transmitted to a computing node connected with the preset low-power wide-area network by using a reserved channel of the preset low-power wide-area network to generate an initial moisture sequence; Difference operation is performed on the initial moisture sequence to generate a moisture change rate as a moisture characteristic value.

4. The method of claim 1, wherein, The moisture characteristic value is matched with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result, which comprises the following steps: A sub-historical moisture sequence associated with a preset current environmental temperature is extracted from the preset raw material moisture historical fluctuation model; A detection time point corresponding to the moisture characteristic value is taken as a reference, preset time length cache moisture value data are extracted from a preset moisture value cache area, the cache moisture value data are arranged in a preset timestamp order, and a current moisture sequence is generated; The similarity values between the current moisture sequence and each historical subsequence in the sub-historical moisture sequence are calculated, and the historical subsequences with similarity values smaller than a preset similarity threshold value are screened out to generate a candidate matching result set; The historical subsequence with the latest timestamp in the candidate matching result set is extracted to generate a raw material moisture matching result.

5. The method of claim 4, wherein, calculating similarity values between the current moisture sequence and each historical sub-sequence in the sub-historical moisture sequence, screening the historical sub-sequences with similarity values less than a preset similarity threshold to generate a candidate matching result set, comprising: dividing the current moisture sequence into three continuous and equal-length current sub-segments, calculating the moisture value change slope of each current sub-segment to generate a current slope feature vector; dividing each historical sub-sequence in each sub-historical moisture sequence into three continuous and equal-length historical sub-segments, calculating the moisture value change slope of each historical sub-sequence to generate a historical slope feature vector; calculating the absolute value difference between each component in the current slope feature vector and the same sequence component in the historical slope feature vector, and using each absolute value difference and a temperature compensation factor corresponding to a preset current ambient temperature to generate a sequence similarity value; screening the sequence similarity values to obtain a plurality of target sequence similarity values less than a preset similarity threshold; taking the historical sub-sequences corresponding to each target sequence similarity value as matching items, and integrating each matching item to obtain a candidate matching result set.

6. The method of claim 1, wherein, when the raw material moisture value in the raw material moisture matching result exceeds a preset moisture value range, generating a drying equipment power adjustment instruction, comprising: extracting a moisture change trend parameter from the raw material moisture matching result; determining a preset basic power adjustment amount of the drying equipment according to the moisture change trend parameter; analyzing the preset working power state message of the drying equipment to generate a current working power value; performing arithmetic superposition operation on the basic power adjustment amount and the current working power value to generate a target power value; calculating the absolute value of the difference between the target power value and the current working power value, and taking the absolute value of the difference as a power change step; combining the target power value and the power change step to generate a drying equipment power adjustment instruction.

7. The method of claim 1, wherein, sending the drying equipment power adjustment instruction to a preset drying equipment execution mechanism to adjust the feed drying working power, comprising: performing encapsulation processing on the drying equipment power adjustment instruction using the reverse control channel of the preset low-power wide area network to generate a power control data packet; analyzing the power control data packet in the preset drying equipment execution mechanism to obtain a target power value and a power change step; adjusting the output power value of the preset drying equipment execution mechanism to be consistent with the target power value according to the power change step to generate a power adjustment completion signal.

8. A feed production system based on the Internet of Things, characterized by, comprising: a collection module for collecting real-time dielectric constant change data of a preset feed raw material in real time using a preset microwave resonant cavity sensor; an analysis module for transmitting the real-time dielectric constant change data to a computing node in communication connection with a preset low-power wide area network, analyzing the real-time dielectric constant change data to generate a moisture feature value; a matching module for matching the moisture feature value with a preset raw material moisture historical fluctuation model to generate a raw material moisture matching result; The generating module is configured to generate a drying equipment power adjustment instruction when the raw material moisture value in the raw material moisture matching result is out of the preset moisture value range; The sending module is configured to send the drying equipment power adjustment instruction to a preset drying equipment execution mechanism to adjust the feed drying work power.

9. A computing device, comprising: The storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to realize the feed production method based on the Internet of Things as any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer program is stored in the computer and is executed to realize the feed production method based on the Internet of Things as any one of claims 1-7.

Images