Soilless culture nutrient solution multi-element detection method and system based on automatic digestion

Abstract

The application relates to the technical field of element detection, and discloses a soilless culture nutrient solution multi-element detection method and system based on automatic digestion, which automatically determines a first detection strategy containing a segmented digestion program and a colorimetric analysis parameter by collecting sample intrinsic parameters, realizes adaptive detection process configuration for different sample matrix characteristics, and effectively improves the automation and intelligent level of multi-element synchronous detection. In the detection process, strategy execution process data of each channel are collected in real time and execution exception judgment is carried out, corresponding detection adjustment measures are taken once an abnormal signal is identified, and the reliability and fault tolerance are significantly enhanced. Not only is the manual intervention greatly reduced, the operation complexity and safety risk are reduced, but also the consistency and accuracy of the detection results are guaranteed through dynamic strategy matching and real-time exception response, and the application is especially suitable for the actual needs of continuous, dynamic and reliable monitoring of nutrient solution components in a soilless culture production site.

Description

Technical Field

[0001] This invention relates to the field of element detection technology, and more specifically to a method and system for multi-element detection in hydroponics nutrient solution based on automatic digestion. Background Technology

[0002] As an important form of modern agriculture, soilless cultivation relies on the composition of its nutrient solution to directly affect crop yield and quality. Therefore, rapid and accurate dynamic monitoring of the concentration of various elements in the nutrient solution is crucial.

[0003] While commonly available water quality testing instruments are easy to operate, their detection capabilities are limited, typically covering only a few conventional parameters. This makes it difficult to meet the comprehensive testing needs for essential elements such as nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, and iron in hydroponics nutrient solutions. Specialized testing instruments for soil and fertilizer analysis, although offering more comprehensive detection capabilities, generally rely on complex manual pretreatment procedures, including digestion, dilution, and reagent preparation. These procedures are not only cumbersome and require highly skilled personnel, but also pose safety hazards due to the use of strong acids and other hazardous reagents. They are ill-suited to the practical needs of online, real-time, and dynamic monitoring of nutrient solutions in hydroponics production sites. Furthermore, existing equipment lacks the ability to adaptively adjust the testing process when faced with nutrient solution samples from different sources and with varying matrix characteristics, severely limiting the reliability and intelligence of the test results. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method and system for multi-element detection in hydroponics nutrient solution based on automatic digestion, so as to solve the technical problem that the element index detection is difficult to meet the requirements of comprehensiveness, reliability and intelligent detection of multi-element nutrient solution in hydroponics.

[0005] The first aspect of this invention discloses a method for multi-element detection in hydroponics nutrient solution based on automatic digestion, the method comprising the following steps:

[0006] S1. Obtain the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters.

[0007] S2. The nutrient solution sample to be tested is assigned to the first channel group in the multi-channel detection unit, and the nutrient solution sample to be tested is detected based on the first detection strategy. The strategy execution process data of the first channel group is collected in real time during the detection process.

[0008] S3. Based on the strategy execution process data, perform anomaly judgment. When an anomaly signal is detected, adjust the detection according to the type of anomaly signal. If no anomaly signal is detected in the strategy execution process data throughout the detection process, generate the first concentration detection result of each target element in the nutrient solution sample to be tested based on the colorimetric analysis data.

[0009] Furthermore, the determination of the first detection strategy based on intrinsic parameters includes:

[0010] The intrinsic parameters are input into a pre-trained policy generation model, which then outputs a first detection policy; whereby...

[0011] The training process of the policy generation model includes:

[0012] Obtain intrinsic parameter samples of historical nutrient solution samples and corresponding optimal detection strategy labels that have been manually verified;

[0013] An initial random forest model is constructed, with the intrinsic parameter samples as input and the optimal detection policy label as output. Supervised learning training is performed on the initial random forest model to obtain the policy generation model.

[0014] Furthermore, the intrinsic parameters include conductivity, pH, turbidity, color, ambient temperature data, and the identification of the planting area from which the sample originated; wherein,

[0015] The strategy generation model input layer simultaneously receives the ambient temperature data and the sample source planting area identifier, which are used as auxiliary features.

[0016] Furthermore, the anomaly detection includes anomaly detection in the digestion stage and the colorimetric reaction stage; wherein, the anomaly detection process in the digestion stage includes:

[0017] When the real-time temperature feedback value of the heating subunit of the multi-channel detection unit is within the preset heating stage, and the deviation from the set temperature value exceeds the preset threshold and continues for a preset duration, a first abnormal signal is generated.

[0018] When the pressure inside the digestion tube of the multi-channel detection unit exceeds the preset safety threshold during the atmospheric pressure digestion stage, a second abnormal signal is generated.

[0019] Furthermore, the abnormality detection process during the colorimetric reaction stage includes:

[0020] When the rate curve of absorbance change over time in the colorimetric reaction stage fails to reach the preset stable plateau value within the preset time window, a third abnormal signal is generated.

[0021] When the blank absorbance of the nutrient solution sample to be tested exceeds the preset cleanliness threshold, a fourth abnormal signal is generated.

[0022] Furthermore, the process of detecting and adjusting according to the type of abnormal signal when an abnormal signal is detected specifically includes:

[0023] When abnormal signals occur during the digestion stage, a joint analysis is performed on the generation of abnormal signals during the digestion stage to obtain the first joint analysis result, and a detection adjustment strategy is determined based on the first joint analysis result.

[0024] When no abnormal signal appears during the digestion stage but an abnormal signal appears during the colorimetric reaction stage, a joint analysis is performed on the generation of abnormal signals during the colorimetric reaction stage to obtain a second joint analysis result, and a detection adjustment strategy is determined based on the second joint analysis result.

[0025] Furthermore, after outputting the first concentration detection results of each target element in the nutrient solution sample to be tested, the method also includes performing a rationality check on the first concentration detection results based on agronomic logic.

[0026] If the rationality verification result fails, a second detection strategy is determined based on the rationality verification result and the first detection strategy, and the same nutrient solution sample to be tested is tested based on the second detection strategy to obtain the second concentration detection results of each target element.

[0027] Furthermore, the process of determining the second detection strategy includes:

[0028] Based on the items that fail the rationality check, determine the type of logical contradiction;

[0029] Based on the aforementioned logical contradiction type, suspicious target elements involved in the first concentration detection result are identified;

[0030] Based on the type of the suspicious target element and the original segmented resolution procedure corresponding to the suspicious target element in the first detection strategy, a second detection strategy is generated for the suspicious target element.

[0031] Furthermore, the method also includes:

[0032] During colorimetric analysis, the temperature inside the colorimetric cell is collected in real time, and the measured absorbance is corrected to the equivalent absorbance at the standard temperature according to the preset temperature compensation model.

[0033] The second aspect of this invention discloses a multi-element detection system for hydroponics nutrient solution based on automatic digestion, the system comprising:

[0034] The detection strategy determination module is used to acquire the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters.

[0035] The detection strategy execution module is used to allocate the nutrient solution sample to be tested to the first channel group in the multi-channel detection unit, detect the nutrient solution sample to be tested based on the first detection strategy, and collect the strategy execution process data of the first channel group in real time during the detection process.

[0036] The anomaly detection module is used to detect anomalies based on data from the strategy execution process. When an anomaly signal is detected, the module performs detection and adjustment according to the type of the anomaly signal.

[0037] The detection result generation module is used to generate the first concentration detection results of each target element in the nutrient solution sample based on colorimetric analysis data when no abnormal signals appear in the data during the entire strategy execution process of the detection process.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] This invention automatically determines a first detection strategy, including a segmented digestion procedure and colorimetric analysis parameters, by collecting intrinsic parameters of the sample. This enables adaptive detection process configuration for different sample matrix characteristics, effectively improving the automation and intelligence level of multi-element simultaneous detection. During the detection process, the strategy execution data of each channel is collected in real time, and execution anomalies are judged. Once an anomaly signal is identified, corresponding detection adjustment measures are taken, significantly enhancing reliability and fault tolerance. This not only greatly reduces manual intervention, lowers operational complexity and safety risks, but also ensures the consistency and accuracy of detection results through dynamic strategy matching and real-time anomaly response. It is particularly suitable for the practical needs of continuous, dynamic, and reliable monitoring of nutrient solution components in hydroponics production sites. Attached Figure Description

[0040] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0041] Figure 1 This is a flowchart illustrating a method for multi-element detection of hydroponic nutrient solution based on automatic digestion, as disclosed in an embodiment of the present invention. Detailed Implementation

[0042] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0043] Example 1

[0044] The first aspect of this invention discloses a method for multi-element detection in hydroponics nutrient solution based on automated digestion, which belongs to the field of element detection technology. Please refer to... Figure 1 , Figure 1 This is a flowchart illustrating a method for multi-element detection in hydroponics nutrient solution based on automatic digestion, as disclosed in an embodiment of the present invention. The method includes the following steps:

[0045] S1. Obtain the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters.

[0046] S2. The nutrient solution sample to be tested is assigned to the first channel group in the multi-channel detection unit, and the nutrient solution sample to be tested is detected based on the first detection strategy. The strategy execution process data of the first channel group is collected in real time during the detection process.

[0047] S3. Based on the strategy execution process data, perform anomaly judgment. When an anomaly signal is detected, adjust the detection according to the type of anomaly signal. If no anomaly signal is detected in the strategy execution process data throughout the detection process, generate the first concentration detection result of each target element in the nutrient solution sample to be tested based on the colorimetric analysis data.

[0048] Furthermore, the determination of the first detection strategy based on intrinsic parameters includes:

[0049] The intrinsic parameters are input into a pre-trained policy generation model, which then outputs a first detection policy. These intrinsic parameters include, but are not limited to, conductivity, pH, turbidity, color, ambient temperature data, and the identifier of the sample's originating planting area. The input layer of the policy generation model simultaneously receives the ambient temperature data and the identifier of the sample's originating planting area, using them as auxiliary features.

[0050] In this embodiment of the invention, after acquiring the hydroponic nutrient solution sample to be tested, multiple intrinsic parameters of the sample are collected in real time by a micro-sensor module integrated in the sample injection path. The collected intrinsic parameters include, but are not limited to, conductivity, pH, turbidity, and color. Simultaneously, the current ambient temperature data is acquired through a configured temperature sensor, and the identification of the sample's originating planting area is obtained through user input or device location information. These intrinsic parameters are used to determine a first detection strategy suitable for the current sample.

[0051] Understandably, the first detection strategy is a comprehensive detection scheme, the core of which mainly includes segmented digestion procedures independently set for different target elements and corresponding colorimetric analysis operating parameters. The segmented digestion procedures specifically define a series of operational steps and process conditions required for each element detection channel during the pretreatment stage, such as including at least two stages with different target temperatures and different reagent environments. For example, for the detection of phosphorus, the segmented digestion procedure can be set as follows: In the first stage, the reaction tube is heated to 50 degrees Celsius and held at that temperature for 5 minutes. No digestion acid reagent is added in this stage, or only a small amount of dilute acid is added, to gently decompose easily released organic phosphorus components in the sample. In the second stage, a quantitative amount of ammonium molybdate-ascorbic acid mixed colorimetric reagent is automatically added, and the temperature is raised to 90 degrees Celsius and held at that temperature for 20 minutes to complete the acidification and reduction environment required for the phosphomolybdate blue colorimetric reaction. For example, in the detection of potassium, the segmented digestion procedure can include a pretreatment stage with the addition of a masking agent and a direct colorimetric stage that does not require high-temperature digestion. Turbidity determination is achieved by adding sodium tetraphenylborate reagent at room temperature and controlling the reaction time. The segmented digestion procedures for each element channel are independently adjustable in terms of the number of temperature steps, target temperature values ​​for each stage, isothermal duration, heating rate, reagent type and addition order, and dosage, thus providing an optimal pretreatment chemical environment for different elements.

[0052] The colorimetric analysis operating parameters specify the detailed settings for the optical detection process, including but not limited to the characteristic wavelengths selected for each target element, the optical path length of the cuvette, the setting of the reading delay time and integration time, and whether to enable the standard addition calibration procedure or the dual-wavelength correction algorithm. These colorimetric parameters, together with the segmented digestion procedure, constitute the primary detection strategy for a specific sample. The overall goal is to balance detection speed and result reliability, making it suitable for rapid screening of large batches of samples in daily operations.

[0053] After the strategy generation model is trained, the intrinsic parameters collected in step S1 are used as input vectors to feed into the model. After inference, the model outputs the first detection strategy code applicable to the current sample. Based on the code, the corresponding segmented digestion program and colorimetric analysis parameters are retrieved and sent to each detection channel for execution.

[0054] Furthermore, the training process of the policy generation model includes:

[0055] Obtain intrinsic parameter samples of historical nutrient solution samples and corresponding optimal detection strategy labels that have been manually verified;

[0056] An initial random forest model is constructed, with the intrinsic parameter samples as input and the optimal detection policy label as output. Supervised learning training is performed on the initial random forest model to obtain the policy generation model.

[0057] Specifically, in this embodiment of the invention, a random forest model is preferably used as the core algorithm for the strategy generation model. During the model training process, at least 500 historical nutrient solution samples from different growing seasons, different crop types, and different nutrient solution formulations are first collected to ensure that the samples have sufficient representativeness in terms of intrinsic parameters such as conductivity, pH, turbidity, and color. For each historical sample, professional analysts determine a set of detection strategies based on its intrinsic parameter characteristics through orthogonal experiments and / or empirical rules. These strategies are verified to meet the accuracy and precision requirements and are encoded as optimal detection strategy labels. The encoding method of the labels is not specifically limited in this embodiment of the invention; for example, multi-digit combinations can be used to represent dilution factors, digestion temperature curve codes, color development time levels, wavelength selection codes, etc.

[0058] Subsequently, the intrinsic parameter samples of each historical sample (including conductivity, pH, turbidity, color, ambient temperature data, and one-heat encoding of the planting area identifier) ​​are used as input feature vectors, and the corresponding optimal detection strategy label is used as the output target to construct a training dataset. Based on this, an initial random forest model containing several decision trees is constructed. Supervised learning training is performed on the model using the training dataset. Model performance is optimized by adjusting hyperparameters such as the number of trees, maximum depth, and splitting criteria, ultimately obtaining a strategy generation model capable of accurately predicting applicable detection strategies based on the input intrinsic parameters. The trained model is deployed to the embedded industrial control unit of the detection system. During actual operation, the recommended first detection strategy is obtained by inputting the real-time collected intrinsic parameters into the model.

[0059] The specific reasons for selecting the above-mentioned specific indicators as preferred intrinsic parameters in this embodiment of the invention, i.e., as inputs to the strategy generation model, include: For conductivity, the conductivity value reflects the total soluble salt concentration in the nutrient solution. High conductivity means that the sample matrix contains a large number of interfering ions. If such samples are directly digested and colorimetrically analyzed, scale is easily formed on the digestion tube wall, affecting heat transfer efficiency. At the same time, significant background absorption interference occurs during colorimetric analysis. Therefore, the model of this invention tends to output a first detection strategy that includes a higher dilution factor or an additional masking agent addition step. For pH, the pH value affects the equilibrium state of the colorimetric reaction and the stability of the complex. Different colorimetric systems have their optimal reaction pH window. For example, the phosphomolybdic blue method requires a near-neutral reaction environment. If the sample is strongly acidic and not neutralized, the colorimetric reaction will be inhibited. Therefore, the model can adjust the addition ratio of acid or alkali reagents according to the pH value. Turbidity and color are used to characterize the content of suspended particulate matter and dissolved pigments in a sample. Therefore, samples with high turbidity or high color will significantly increase the background absorbance during colorimetric analysis, causing the net absorbance of the target element to be submerged. In this case, the model will tend to select the dual-wavelength correction method or the standard addition method as part of the colorimetric analysis parameters.

[0060] Furthermore, ambient temperature data is incorporated into the model as an auxiliary feature because the chemical reaction rate constant follows the Arrhenius equation with temperature. The completeness of the colorimetric reaction and the stability of the chromogenic agent are both affected by temperature. In high-temperature environments, appropriately shortening the colorimetric reaction time or lowering the digestion set temperature can prevent premature decomposition of the chromogenic agent or linear range shifts caused by excessively rapid reactions. Therefore, ambient temperature data provides the model with crucial context for predicting reaction kinetics. Additionally, the identification of the sample's origin growing area is used as another auxiliary feature to introduce prior agronomic knowledge. It is understandable that different growing areas often use specific nutrient solution formulations over a long period. For example, strawberry growing areas commonly use high-potassium formulations, while leafy vegetable growing areas primarily use high-nitrogen formulations. By learning the correlation pattern between the growing area identification and the detection strategy, the model can predict the approximate concentration range of elements before the specific concentration of the sample is observed. This allows for pre-adjustment of the dilution factor of relevant channels or the sensitivity setting of the colorimetric wavelength during strategy generation, further improving the accuracy of strategy matching and detection efficiency.

[0061] Through the synergistic input of the aforementioned multidimensional intrinsic parameters and auxiliary features, the strategy generation model of this invention can simulate the decision-making process of professional analysts and customize the optimal first detection strategy for each specific sample without human intervention, laying a solid technical foundation for the fully automated and intelligent detection of multiple elements in soilless cultivation nutrient solution.

[0062] Furthermore, anomaly detection includes anomaly detection in both the digestion stage and the colorimetric reaction stage; specifically, the anomaly detection process in the digestion stage includes:

[0063] When the real-time temperature feedback value of the heating subunit of the multi-channel detection unit is within the preset heating stage, and the deviation from the set temperature value exceeds the preset threshold and continues for a preset duration, a first abnormal signal is generated.

[0064] When the pressure inside the digestion tube of the multi-channel detection unit exceeds the preset safety threshold during the atmospheric pressure digestion stage, a second abnormal signal is generated.

[0065] Furthermore, the abnormality detection process during the colorimetric reaction stage includes:

[0066] When the rate curve of absorbance change over time in the colorimetric reaction stage fails to reach the preset stable plateau value within the preset time window, a third abnormal signal is generated.

[0067] When the blank absorbance of the nutrient solution sample to be tested exceeds the preset cleanliness threshold, a fourth abnormal signal is generated.

[0068] Furthermore, the process of detecting and adjusting according to the type of abnormal signal when an abnormal signal is detected specifically includes:

[0069] When abnormal signals occur during the digestion stage, a joint analysis is performed on the generation of abnormal signals during the digestion stage to obtain the first joint analysis result, and a detection adjustment strategy is determined based on the first joint analysis result.

[0070] When no abnormal signal appears during the digestion stage but an abnormal signal appears during the colorimetric reaction stage, a joint analysis is performed on the generation of abnormal signals during the colorimetric reaction stage to obtain a second joint analysis result, and a detection adjustment strategy is determined based on the second joint analysis result.

[0071] Specifically, when the present invention detects the nutrient solution sample to be tested based on the first detection strategy, it collects the strategy execution process data of the first channel group in real time during the detection process, and makes anomaly judgment on the strategy execution process based on these data, thereby adjusting the detection strategy accordingly.

[0072] It is understandable that, since the digestion stage and the color reaction stage are sequential in time and the hardware modules and chemical processes involved are fundamentally different, this invention divides the anomaly judgment logic into two parts: anomaly judgment in the digestion stage and anomaly judgment in the color reaction stage, and adopts differentiated detection and adjustment strategies according to the stage of anomaly occurrence and the specific signal type.

[0073] In this embodiment of the invention, the core task of the digestion stage is to complete the chemical pretreatment of the sample under controlled temperature and pressure conditions. Therefore, temperature and pressure are two key process parameters characterizing the operating status of this stage. This invention continuously monitors the real-time temperature feedback value of the heating subunit in the multi-channel detection unit and compares it with the set temperature value within the preset heating stage. When the absolute value of the deviation between the real-time temperature feedback value and the set temperature value exceeds a preset threshold, for example, ±3 degrees Celsius, and this deviation persists for more than a preset duration, for example, more than 10 seconds, it indicates that the temperature control system cannot adjust the actual temperature to near the target value. Possible causes of this phenomenon include an open circuit or short circuit fault in the heating element leading to inability to heat normally, temperature sensor drift or disconnection causing distorted feedback values, or solid-state relay sticking causing continuous heating that cannot be stopped. However, regardless of the specific source of the fault, the channel can no longer provide an accurate temperature environment as required by the segmented digestion procedure, thus generating a first abnormal signal.

[0074] To address the generation conditions of the second abnormal signal, this invention uses a miniature pressure sensor positioned above the digestion tube to monitor the pressure inside the tube in real time. During the atmospheric pressure digestion stage, i.e., when the digestion process does not require a sealed high-pressure environment, the pressure inside the digestion tube should be maintained at a level close to atmospheric pressure. Once the pressure inside the tube exceeds a preset safety threshold, such as exceeding 1.5 times the standard atmospheric pressure, it indicates that gas discharge is obstructed or the liquid volume is abnormally expanded. Possible causes of this phenomenon include blockage in the digestion tube outlet pipe preventing gas discharge, reagent addition exceeding the preset range causing violent boiling, or the sample containing a high concentration of carbonate substances reacting with acid to produce a large amount of carbon dioxide gas. Any of these situations may lead to digestion liquid splashing, cross-contamination, or even hardware damage, thereby generating the second abnormal signal.

[0075] The colorimetric reaction stage occurs after digestion is complete and the reaction solution has cooled to a suitable temperature. Its core task is to add the colorimetric reagent and monitor the absorbance change over time until a stable plateau value is reached. Therefore, the kinetic characteristics of absorbance changes and the background absorbance under blank conditions are key parameters characterizing the operational status of this stage.

[0076] The logic behind the generation of the third abnormal signal is understandable. Under normal colorimetric reaction conditions, absorbance should rise rapidly within a certain time and gradually stabilize; that is, the rate curve should decrease to near zero within a preset time window and remain near a plateau value. Therefore, if the absorbance has not reached the preset stable plateau value by the end of the preset time window, exhibiting a continuous slow rise or irregular fluctuations, it indicates that the colorimetric reaction has not been completed as expected. Forcibly reading the absorbance value and substituting it into the standard curve to calculate the concentration at this time will produce significant errors, thus generating the third abnormal signal. The causes of this abnormality may include the colorimetric reagent becoming partially or completely ineffective due to improper storage or exceeding its expiration date; the presence of interfering substances in the sample matrix that compete with the colorimetric reagent and inhibit the main reaction; or the reaction solution temperature deviating from the optimal temperature range for the colorimetric reaction, resulting in an excessively slow reaction rate.

[0077] The generation logic of the fourth abnormal signal can be further understood as follows: blank absorbance reflects the overall background level of the cuvette window cleanliness, light source intensity stability, and the sample's own color and turbidity. Therefore, when the blank absorbance exceeds a preset cleanliness threshold, such as an absorbance exceeding 0.1 absorbance units at the target wavelength, it indicates significant light attenuation caused by non-target analytes in the optical system. The causes of this phenomenon may include decreased transmittance due to residues adhering to the cuvette wall, light source aging leading to attenuation of output light intensity, or excessively dark sample color that has not been effectively diluted. In such cases, direct colorimetric determination will severely interfere with the calculation of the net absorbance of the target element.

[0078] Based on the aforementioned phased abnormal signal generation, this invention further performs targeted joint analysis and detection adjustments according to the occurrence stage and combination pattern of the abnormal signals. Since the digestion stage is executed before the colorimetric reaction stage in sequence, and abnormalities in the digestion stage usually involve physical failures of hardware actuators or fluid systems, once any abnormal signal is detected in the digestion stage, the abnormality of that stage will be processed first, without waiting for the execution result of the colorimetric reaction stage.

[0079] Specifically, when an abnormal signal occurs during the digestion stage, if only the first abnormal signal is generated and no second abnormal signal is generated, it indicates a temperature control-related fault while the pressure is normal. In this case, it can be inferred that there is a local fault in the heating element or temperature sensor, but there is no risk of blockage or boiling over in the digestion tube. In this situation, it is determined that the temperature control function of the current channel is unreliable, and continued use of the channel will lead to incomplete digestion or reagent decomposition. Therefore, the preferred detection and adjustment strategy is to immediately stop the operation of the current channel, generate a prompt message indicating that the heating system needs maintenance, and automatically switch the current sample residue in this channel to a pre-configured redundant backup channel with the same structure. The backup channel then takes over the unfinished digestion and subsequent detection tasks.

[0080] If only the second abnormal signal is generated and the first abnormal signal is not generated, it indicates that the temperature control function is normal, but the pressure inside the digestion tube is abnormally high. In this case, it can be inferred that the heating element and temperature sensor are working properly, and the root cause of the pressure increase is flow path blockage or excessive reagent addition, rather than temperature runaway. In this situation, it is determined that there is a fault in the fluid system of the current channel, but the heating and temperature control hardware itself is still working. Therefore, the preferred detection and adjustment strategy is to suspend the digestion program of the current channel, activate the pressure relief valve to release the pressure inside the tube, and perform an enhanced cleaning process to clear the pipeline. After the pressure returns to normal, the digestion program can be restarted in the same channel without using the redundant backup channel.

[0081] If the first and second abnormal signals are generated simultaneously, it indicates a dual fault of temperature runaway and pressure anomaly. In this case, temperature runaway is often the root cause, with continuous overheating leading to violent boiling of the liquid and thus causing a pressure increase. Therefore, this combined pattern is considered a serious heating system runaway fault. The preferred detection and adjustment strategy is to immediately cut off the heating power supply to this channel to ensure safety, terminate all unfinished tasks in this channel, generate an emergency alarm, and switch the sample retention solution to a redundant backup channel for retesting.

[0082] Furthermore, it should be noted that the multi-channel detection unit of this invention is designed with at least two structurally identical independent detection channels for each target element (such as N, P, and K). One channel serves as the primary channel, and the other as a redundant backup channel. When a hardware failure is detected in a primary channel (such as element K), and the system switches to the redundant backup channel based on the detection adjustment strategy, the backup channel takes over the detection, while other normally functioning N channels, P channels, etc., remain unaffected and continue to perform their respective detection tasks. This single-channel fault-tolerant design avoids the waste of resources that would otherwise require retesting the entire batch of samples due to a single channel failure.

[0083] Through the above-mentioned hierarchical joint analysis, the embodiments of the present invention can accurately distinguish the root cause of the fault, avoid unnecessary channel switching or reagent waste caused by misjudgment of a single signal, and significantly improve fault tolerance and operating efficiency.

[0084] On the other hand, when no abnormal signals appear during the entire digestion stage, but abnormal signals appear during the subsequent colorimetric reaction stage, this invention continues to jointly analyze the generation of the third and fourth abnormal signals. Since the digestion process has been successfully completed at this point, the root cause of the abnormality is mainly limited to non-hardware destructive factors such as colorimetric reagents, optical systems, or sample background interference.

[0085] Specifically, if only the third abnormal signal is generated and not the fourth abnormal signal, it indicates that the blank absorbance is normal, the cuvette and light source are in good condition, but the kinetics of the colorimetric reaction are abnormal. In this case, it is inferred that the colorimetric reagent is inactive or that specific interfering substances in the sample matrix are inhibiting the colorimetric reaction, but the optical hardware system is intact. In this situation, the colorimetric detection result of the current channel is deemed unreliable, but hardware switching is not required. The preferred detection adjustment strategy is to generate a prompt message, pause the detection task of the current channel, mark the channel as needing colorimetric reagent replacement or masking agent optimization calibration, and suggest that the operator manually verify the sample using the standard addition method.

[0086] If only the fourth abnormal signal is generated but the third abnormal signal is not, it indicates that the colorimetric reaction can reach a stable plateau normally, but the background absorbance in the blank state is too high. In this case, it is inferred that the cuvette is contaminated or the light source is aging, while the colorimetric reagent is of normal activity. To address this situation, the preferred detection adjustment strategy is to automatically trigger the cuvette cleaning program, repeatedly rinsing the cuvette with the built-in cleaning solution and re-measuring the blank value. If the blank absorbance falls below the threshold after rinsing, the colorimetric measurement of the current sample continues. If the blank value still exceeds the limit after rinsing, a prompt message is generated indicating that the optical system needs maintenance and calibration.

[0087] When both the third and fourth abnormal signals are generated, it indicates the coexistence of abnormal color reaction and abnormal background in the optical system, signifying extremely severe interference from the sample matrix or a long-term lack of system maintenance. In response to this complex situation, the optimal detection adjustment strategy is to suspend the current channel task, simultaneously trigger the cleaning procedure and reagent status self-check process, and generate a comprehensive alarm message to prompt operators to conduct a thorough overhaul.

[0088] Through the aforementioned hierarchical joint analysis and adjustment of anomalies in the colorimetric reaction stage, this invention can minimize unnecessary hardware switching actions and reduce the occupation of backup channel resources while ensuring the accuracy of detection results, making daily maintenance work more targeted and efficient. Furthermore, by dividing anomaly judgment into a resolution stage and a colorimetric reaction stage, and performing refined joint analysis and differentiated adjustment strategies for different combinations of abnormal signals within each stage, this invention also achieves accurate identification and intelligent response to various faults and interferences during the detection process. This not only ensures the accuracy and reliability of the detection results but also significantly improves the long-term operational stability of the detection system.

[0089] Furthermore, after outputting the first concentration detection results of each target element in the nutrient solution sample to be tested, a rationality check based on agronomic logic is also performed on the first concentration detection results;

[0090] If the rationality verification result fails, a second detection strategy is determined based on the rationality verification result and the first detection strategy, and the same nutrient solution sample to be tested is tested based on the second detection strategy to obtain the second concentration detection results of each target element.

[0091] Furthermore, the process of determining the second detection strategy includes:

[0092] Based on the items that fail the rationality check, determine the type of logical contradiction;

[0093] Based on the aforementioned logical contradiction type, suspicious target elements involved in the first concentration detection result are identified;

[0094] Based on the type of the suspicious target element and the original segmented resolution procedure corresponding to the suspicious target element in the first detection strategy, a second detection strategy is generated for the suspicious target element.

[0095] In this embodiment of the invention, when the entire detection process is completed and no abnormal signals are detected in the strategy execution data, the first concentration detection result of each target element in the nutrient solution sample to be tested will be output. This first concentration detection result is preliminary quantitative data obtained through the first detection strategy, namely the rapid detection strategy. Although it has a fast detection speed and consumes less reagent, due to the relatively simplified rapid digestion procedure, the measured values ​​of individual elements may have certain deviations when dealing with samples with complex matrices or containing special interfering substances.

[0096] To ensure both detection efficiency and the reliability of the final output results, this invention introduces a rationality verification mechanism based on agronomic logic after outputting the first concentration detection result. This rationality verification based on agronomic logic refers to using the distribution patterns of nutrient solution element concentrations and the correlations between elements, summarized through long-term production practice and scientific research in the field of hydroponics, to logically verify the detection results.

[0097] Specifically, the preparation of nutrient solutions for hydroponics typically follows a specific formulation system. Different crops at different growth stages exhibit relatively stable absorption ratios of various elements within a given range, and there are inherent agronomical constraints between the concentrations of each element within the same formulation system. For example, normally growing plants typically maintain nitrogen and potassium absorption ratios within a specific range, and the calcium-to-sulfur concentration ratio does not exhibit extreme imbalances in most nutrient solution formulations. Furthermore, the concentration of any element should not exceed its theoretical limit in conventional nutrient solutions. Therefore, in this embodiment of the invention, a rationality verification rule base is pre-constructed based on the aforementioned agronomical prior knowledge. This rule base stores reasonable concentration ranges for each element and reasonable ranges of ratios between elements for different planting area identifiers and different nutrient solution formulation types.

[0098] When performing agronomic logic-based rationality verification, the first step is to retrieve a subset of verification rules matching the commonly used nutrient solution formula type for that planting area, based on the sample source planting area identifier obtained in step S1. Then, the concentration values ​​of each target element in the first concentration detection result are compared one by one with the corresponding preset reasonable range in the rule subset, and the concentration ratio of any two target elements with agronomic correlation is calculated to determine whether it exceeds the preset correlation ratio range. In addition, it is determined whether the concentration value of any target element exceeds the theoretical limit concentration range of that element. If an extreme value significantly exceeds the physiological requirement limit, it is directly judged as data anomaly. Through the above multi-dimensional logical verification, the internal consistency of chemical detection results can be objectively evaluated from the perspective of agronomic laws, rather than simply relying on the instrument's own quality control indicators. This effectively compensates for the shortcomings of traditional detection instruments that only focus on the accuracy of a single value while neglecting the overall rationality of the results.

[0099] When the rationality verification result is passed, it indicates that the first concentration test result is not only normal in the chemical testing process, but also conforms to the general rules of nutrient solution in terms of agronomic logic. The first concentration test result is used as the final output value and a test report is generated.

[0100] If the rationality verification result fails, the present invention initiates the process of determining and executing the second detection strategy.

[0101] Specifically, the type of logical contradiction is determined based on the specific items that fail the rationality check. For example, if the failed item is a calcium-sulfur ratio exceeding a preset upper limit, the logical contradiction type is determined to be "sulfur may have a negative deviation or calcium may have a positive deviation"; if the failed item is an iron concentration exceeding the theoretical limit, the logical contradiction type is determined to be "iron detection value has a serious positive deviation"; if the failed item involves multiple element ratios that are simultaneously abnormal, the most likely source of the contradiction is identified according to preset priority rules or contradiction correlation diagrams. After determining the type of logical contradiction, the suspected target elements involved in the first concentration detection results are further located based on this logical contradiction type. Continuing the previous example, if the logical contradiction type points to an abnormal calcium-sulfur ratio, both calcium and sulfur are marked as suspected target elements. However, based on the chemical experience that sulfur in nutrient solution is more easily interfered with by organic matter and thus has incomplete color development, sulfur is determined to be the main suspected element, and calcium is used as a related verification element. If the logical contradiction type is an iron concentration exceeding the limit, iron is directly marked as a suspected target element.

[0102] After identifying the suspicious target element, a second detection strategy specifically targeting the suspicious target element is generated based on the type of the suspicious target element and the original segmented resolution procedure corresponding to the suspicious target element in the first detection strategy determined in step S1.

[0103] The core idea of ​​the second detection strategy in this invention is to address the shortcomings of the first detection strategy in dealing with specific matrix interference or reaction inertness by making targeted parameter enhancements or method changes, thereby eliminating or significantly reducing detection bias in retesting.

[0104] Specifically, in this embodiment of the invention, an enhanced detection strategy template library for each target element is pre-stored. This template library records the digestion procedure adjustment scheme and colorimetric parameter optimization scheme to be adopted under different interference scenarios. Specific operations include, but are not limited to, changing the total digestion time, changing the number of isothermal stages in the digestion temperature curve, changing the number of digestion temperature steps, adding masking agent components to the reagent types, changing the wavelength selection in the colorimetric analysis parameters, adopting the standard addition method for calibration, and adding activated carbon adsorption decolorization pretreatment steps, etc.

[0105] For example, when the suspected target element is sulfur, sulfur is easily masked by high concentrations of chloride ions or organic acids in the sample when measured using conventional barium sulfate turbidimetric or methylene blue colorimetric methods, resulting in incomplete color development and low results. Therefore, the second detection strategy is preferred to adopt the following parameter modification scheme compared to the first detection strategy: the total digestion time is extended from 15 minutes in the rapid detection mode to 30 minutes, and an isothermal stage is added to the digestion temperature curve. That is, after adding the masking agent, the temperature is first kept constant at 60 degrees Celsius for 10 minutes to fully release the sulfur bound to organic matter, and then the temperature is raised to 95 degrees Celsius for complete digestion. At the same time, masking agent components such as hydrogen peroxide or perchloric acid are added to the reagent types to destroy organic interference.

[0106] Through the precise strategy adjustment based on the types of suspected elements described above, the second detection strategy does not indiscriminately re-measure all target elements. Instead, it only performs enhanced re-testing on specific elements with potential problems identified through agronomic logic verification in the first concentration test results. The retained solution corresponding to the suspected target element from the same nutrient solution sample is diverted to the redundant detection channel for that element. A more stringent digestion and colorimetric process is then performed according to the second detection strategy to obtain the second concentration test result for the suspected target element. This second concentration test result replaces the corresponding element value in the first concentration test result and is combined with the first concentration test results of other elements that did not trigger re-testing to form the final test report.

[0107] Through the above operations, this embodiment of the invention significantly improves the accuracy and reliability of detection results while minimizing reagent consumption and time costs associated with repeated testing, enabling high-throughput detection capabilities to be maintained while ensuring data quality. Simultaneously, every instance of failed rationality verification and triggering of the second strategy is recorded and transmitted back to the cloud database for subsequent iterative optimization of the strategy generation model, forming a closed-loop evolution of the detection strategy and verification mechanism, further enhancing the intelligence level of detection and its long-term adaptive capabilities.

[0108] Furthermore, during the colorimetric analysis, the temperature inside the colorimetric cell is collected in real time, and the measured absorbance is corrected to the equivalent absorbance at the standard temperature according to the preset temperature compensation model.

[0109] Specifically, in order to further eliminate the influence of ambient temperature fluctuations and residual heat of the reaction solution after digestion on the accuracy of colorimetric determination during the colorimetric analysis operation, this embodiment of the invention deploys a high-precision miniature temperature sensor near the colorimetric cell of each detection channel to collect the actual temperature of the reaction solution in the colorimetric cell in real time.

[0110] Since the equilibrium constant of the colorimetric reaction and the molar absorptivity of the absorbing substance are both temperature-dependent, even for test solutions of the same concentration, the absorbance values ​​measured under different temperature conditions may show slight but not negligible differences. This is especially true for temperature-sensitive colorimetric systems, where temperature changes will cause the standard curve to shift or change its slope. Therefore, this invention adds an absorbance correction step based on a preset temperature compensation model after obtaining the measured absorbance and before substituting it into the standard curve to calculate the concentration.

[0111] For the temperature compensation model, during the initialization or periodic calibration phase, this invention, for each target element's colorimetric reaction system, experimentally obtains absorbance response data of a series of standard solutions of known concentrations under different temperature conditions. Specifically, the standard solution is placed in a constant-temperature colorimetric cell, and an external temperature control device is used to increase the cell temperature in increments within a expected working range, for example, 15°C to 40°C, recording the absorbance value of the standard solution at each temperature point. Subsequently, a standard reference temperature is selected, preferably 25°C, and the deviation ratio or difference of the absorbance at each temperature point relative to the absorbance at the reference temperature is calculated. A functional relationship between temperature and absorbance correction coefficients is established using a mathematical fitting method. Depending on the complexity of the temperature characteristics of the colorimetric system, a piecewise linear interpolation model is preferably used for this functional relationship. After fitting, the coefficients of the model are stored in the storage unit of the detection system as preset temperature compensation model parameters for the element detection channel.

[0112] Through the above operations, the embodiments of the present invention effectively eliminate the influence of colorimetric temperature fluctuations caused by diurnal temperature differences, seasonal changes, or the dissipation of residual heat on quantitative results, so that the detection accuracy and precision can be maintained under different environmental conditions, significantly improving the environmental adaptability and long-term operational stability of the instrument, and is especially suitable for use in hydroponic production sites that lack constant temperature laboratory conditions.

[0113] Example 2

[0114] The second aspect of this invention discloses a multi-element detection system for hydroponics nutrient solution based on automatic digestion, the system comprising:

[0115] The detection strategy determination module is used to acquire the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters.

[0116] The detection strategy execution module is used to allocate the nutrient solution sample to be tested to the first channel group in the multi-channel detection unit, detect the nutrient solution sample to be tested based on the first detection strategy, and collect the strategy execution process data of the first channel group in real time during the detection process.

[0117] The anomaly detection module is used to detect anomalies based on data from the strategy execution process. When an anomaly signal is detected, the module performs detection and adjustment according to the type of the anomaly signal.

[0118] The detection result generation module is used to generate the first concentration detection results of each target element in the nutrient solution sample based on colorimetric analysis data when no abnormal signals appear in the data during the entire strategy execution process of the detection process.

[0119] It should be noted that the specific implementation process of Example 2 is similar to that of Example 1, and will not be repeated in Example 2.

[0120] Finally, it should be noted that the above-described embodiments include multiple parallel implementations of the present invention. Deleting or otherwise adjusting one or more implementations will not affect the implementation of the solution. Furthermore, the multi-element detection method and system for hydroponics nutrient solution based on automatic digestion disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for multi-element detection in hydroponics nutrient solution based on automated digestion, the method comprising the following steps: S1. Obtain the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters. S2. The nutrient solution sample to be tested is assigned to the first channel group in the multi-channel detection unit, and the nutrient solution sample to be tested is detected based on the first detection strategy. The strategy execution process data of the first channel group is collected in real time during the detection process. S3. Based on the data from the strategy execution process, perform anomaly detection and adjustment according to the type of anomaly signal when an anomaly signal is detected; When no abnormal signals are found in the data during the entire detection process, the first concentration detection results of each target element in the nutrient solution sample are generated based on the colorimetric analysis data.

2. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 1, characterized in that, The determination of the first detection strategy based on intrinsic parameters includes: The intrinsic parameters are input into a pre-trained policy generation model, which then outputs a first detection policy; whereby... The training process of the policy generation model includes: Obtain intrinsic parameter samples of historical nutrient solution samples and corresponding optimal detection strategy labels that have been manually verified; An initial random forest model is constructed, with the intrinsic parameter samples as input and the optimal detection policy label as output. Supervised learning training is performed on the initial random forest model to obtain the policy generation model.

3. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 2, characterized in that, The intrinsic parameters include conductivity, pH, turbidity, color, ambient temperature data, and the identification of the planting area from which the sample originated; among them, The strategy generation model input layer simultaneously receives the ambient temperature data and the sample source planting area identifier, which are used as auxiliary features.

4. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 1, characterized in that, The anomaly detection includes anomaly detection in the digestion stage and the colorimetric reaction stage; wherein, the anomaly detection process in the digestion stage includes: When the real-time temperature feedback value of the heating subunit of the multi-channel detection unit is within the preset heating stage, and the deviation from the set temperature value exceeds the preset threshold and continues for a preset duration, a first abnormal signal is generated. When the pressure inside the digestion tube of the multi-channel detection unit exceeds the preset safety threshold during the atmospheric pressure digestion stage, a second abnormal signal is generated.

5. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 4, characterized in that, The abnormality detection process during the colorimetric reaction stage includes: When the rate curve of absorbance change over time in the colorimetric reaction stage fails to reach the preset stable plateau value within the preset time window, a third abnormal signal is generated. When the blank absorbance of the nutrient solution sample to be tested exceeds the preset cleanliness threshold, a fourth abnormal signal is generated.

6. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 5, characterized in that, The process of detecting and adjusting according to the type of abnormal signal when an abnormal signal is detected specifically includes: When abnormal signals occur during the digestion stage, a joint analysis is performed on the generation of abnormal signals during the digestion stage to obtain the first joint analysis result, and a detection adjustment strategy is determined based on the first joint analysis result. When no abnormal signal appears in the digestion stage but an abnormal signal appears in the colorimetric reaction stage, a joint analysis is performed on the generation of abnormal signals in the colorimetric reaction stage to obtain a second joint analysis result, and a detection adjustment strategy is determined based on the second joint analysis result.

7. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 1, characterized in that, After outputting the first concentration detection results of each target element in the nutrient solution sample to be tested, the method further includes performing a rationality check on the first concentration detection results based on agronomic logic. If the rationality verification result fails, a second detection strategy is determined based on the rationality verification result and the first detection strategy, and the same nutrient solution sample to be tested is tested based on the second detection strategy to obtain the second concentration detection results of each target element.

8. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 7, characterized in that, The process of determining the second detection strategy includes: Based on the items that fail the rationality check, determine the type of logical contradiction; Based on the aforementioned logical contradiction type, suspicious target elements involved in the first concentration detection result are identified; Based on the type of the suspicious target element and the original segmented resolution procedure corresponding to the suspicious target element in the first detection strategy, a second detection strategy is generated for the suspicious target element.

9. The method for multi-element detection of hydroponic nutrient solution based on automatic digestion according to claim 1, characterized in that, The method further includes: During colorimetric analysis, the temperature inside the colorimetric cell is collected in real time, and the measured absorbance is corrected to the equivalent absorbance at the standard temperature according to the preset temperature compensation model.

10. A multi-element detection system for hydroponics nutrient solution based on automatic digestion, said detection system being implemented based on the detection method according to any one of claims 1-9, characterized in that, The system includes: The detection strategy determination module is used to acquire the nutrient solution sample to be tested, collect the intrinsic parameters of the nutrient solution sample to be tested, and determine the first detection strategy based on the intrinsic parameters; the first detection strategy includes a segmented digestion procedure for different target elements and corresponding colorimetric analysis operation parameters. The detection strategy execution module is used to allocate the nutrient solution sample to be tested to the first channel group in the multi-channel detection unit, detect the nutrient solution sample to be tested based on the first detection strategy, and collect the strategy execution process data of the first channel group in real time during the detection process. The anomaly detection module is used to detect anomalies based on data from the strategy execution process. When an anomaly signal is detected, the module performs detection and adjustment according to the type of the anomaly signal. The detection result generation module is used to generate the first concentration detection results of each target element in the nutrient solution sample based on colorimetric analysis data when no abnormal signals appear in the data during the entire strategy execution process of the detection process.

Images