Water body comprehensive toxicity detection system based on algae

Abstract

The invention discloses an algae-based water body comprehensive toxicity detection system, which relates to the technical field of toxicity detection, and comprises an optical imaging module, a multi-parameter sensing module, an AI intelligent counting module, a growth curve analysis module and a parameter regulation and control module, cells and grids are recognized through the AI model, a multi-view averaging algorithm is combined, manual errors are reduced, errors caused by subjective factors and fatigue of a counter are avoided, test repeatability is improved, pesticide toxicity test standards are met, high-frequency automatic detection can be achieved, a continuous growth curve is generated in real time, inhibition inflection points of algae growth are accurately captured, and the accuracy of the pesticide toxicity test is improved. Meanwhile, the time stamps of concentration, temperature and illumination data are completely synchronized and stored in a unified database, the problems that in a traditional test, parameters are recorded dispersedly and are difficult to trace are solved, a unified counting standard and a parameter recording specification are adopted, the method difference of different laboratories and different operators is avoided, the test result has transverse comparability, and the test efficiency is improved. The pesticide environment risk is met.

Description

Technical Field

[0001] This invention relates to the field of toxicity detection technology, and in particular to a comprehensive water toxicity detection system based on algae. Background Technology

[0002] Algal growth inhibition tests are a crucial method for assessing the toxicity of pesticides and chemicals to aquatic ecosystems. The core principle is to determine the toxicity level of the test substance by measuring changes in algal concentration and calculating the half-maximal inhibitory concentration (ICP-C). Currently, the industry commonly uses the traditional method of "microscope + hemocytometer + manual counting," which has the following significant drawbacks: 1. Large error in manual counting: Hemocytometers require manual differentiation of algal cells from impurities and counting of cells in specific grids. Affected by the subjective judgment of the counter and visual fatigue, the error of a single count often exceeds 10%, and the error between different operators can reach more than 15%, resulting in poor experimental repeatability. Second, it is inefficient and cannot be monitored in real time: manual counting takes 30-60 minutes per operation, making it difficult to achieve high-frequency sampling and monitoring. It can only draw discrete growth curves through intermittent sampling, and cannot capture the dynamic changes in algal growth. 3. Dispersed parameter recording: Algal growth is significantly affected by light and temperature, but in traditional experiments, light intensity and culture temperature need to be recorded by separate instruments, which are in different documents from concentration data. This can easily lead to problems such as timestamp mismatch and difficulty in data traceability, which does not meet the standardized testing requirements of pesticide toxicity tests. IV. Experimental conditions cannot be dynamically adjusted: Because the growth status of algae cannot be monitored in real time, when environmental parameters cause abnormal growth, the experimental conditions cannot be adjusted in time, which may lead to the failure of the entire experiment and increase costs and time losses. In summary, existing technologies are insufficient to meet the requirements of pesticide toxicity testing for "accuracy, real-time performance, and standardization," and there is an urgent need for a detection system that can achieve automated counting, real-time monitoring, and multi-parameter synchronization. To address the aforementioned technical deficiencies, a solution is proposed. Summary of the Invention

[0003] The purpose of this invention is to: identify cells and grids using an AI model, combined with a multi-view averaging algorithm, to reduce human error and avoid errors caused by subjective factors and fatigue of the counter, thereby improving experimental repeatability and meeting pesticide toxicity testing standards. It also enables high-frequency automatic detection, real-time generation of continuous growth curves, and precise capture of the inhibition inflection point of algal growth. When environmental parameters are abnormal, experimental conditions can be adjusted to avoid experimental failures due to parameter fluctuations, reducing retrial costs. Furthermore, it integrates temperature and light sensors to achieve complete synchronization of concentration, temperature, and light data timestamps, storing them in a unified database. This solves the problem of scattered parameter recording and difficulty in tracing in traditional experiments. Moreover, it adopts a unified counting standard and parameter recording specification to avoid methodological differences between different laboratories and operators, making experimental results horizontally comparable and compliant with pesticide environmental risk standards.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a comprehensive water toxicity detection system based on algae, comprising an optical imaging module, a multi-parameter sensing module, an AI intelligent counting module, a growth curve analysis module, and a parameter control module, wherein: The optical imaging module consists of a biological microscope, a CCD industrial camera, and an electric focuser. It is used to acquire images from the hemocytometer and, based on the OpenCV algorithm library, preprocesses the acquired images before sending them to the AI ​​intelligent counting module. The biological microscope uses a 100x objective lens to be fitted with a hemocytometer to ensure clear imaging of the grid lines and cells; The CCD camera has a resolution of ≥1920×1080 and a frame rate of ≥30fps. It connects to an industrial computer via USB to achieve real-time image transmission. The electric focusing device is used in conjunction with the biological camera, and the software controls the autofocus to avoid the deviation of manual focusing. The multi-parameter sensing module integrates temperature and light sensors and is installed in the algae culture chamber to acquire temperature and light data, which are then sent to the experimental control module. The temperature sensor has a measurement range of -40 to 125℃ and an accuracy of ±0.2℃. It uses an RS485 bus to transmit data and is used to collect the temperature of the culture environment in real time. The light sensor has a measurement range of 0–1000 μmol / m 2 •s, with an accuracy of ±5%, used for synchronous recording of light intensity data; The AI ​​intelligent counting module uses an improved U-Net convolutional neural network to build and train an image recognition counting model to achieve automatic counting. The growth curve analysis module is used to automatically generate algal growth curves based on real-time counted concentration data, and to determine the growth stage through slope analysis. It sets a concentration judgment threshold based on the normal control group. If the concentration change rate is lower than the concentration judgment threshold, it is determined that the growth is inhibited and an abnormality alert is triggered. The parameter control module is used to establish a linkage mechanism between growth status and environmental parameters. When the generated curve shows abnormality, it automatically generates temperature adjustment commands and light intensity adjustment commands and sends them to the experimental control module. It also supports a manual intervention mode, where users can manually modify parameter thresholds according to software prompts.

[0005] Furthermore, it also includes a data management module, which uses a MySQL database to store data, realizes the synchronization of the three parameters of concentration, temperature and light intensity timestamps, supports real-time viewing of data curves and export of data in Excel format, and automatically generates a standardized report after the experiment, including IC50 calculation results and parameter change trend graphs.

[0006] Furthermore, the test control module includes a PID temperature control submodule and an LED illumination adjustment submodule, wherein: The PID temperature control submodule is used to connect to the incubator temperature control element, with a temperature control range of 1~0℃ and a temperature control accuracy of ±0.1℃. It can receive software commands to adjust the temperature. The LED illumination adjustment submodule is used to connect to the LED light source inside the incubator, and the light intensity adjustment range is 0~800μmol / m². 2 •s, supports setting the illumination cycle via software.

[0007] Furthermore, the preprocessing includes the following steps: S101, Denoising: Use a 5×5 Gaussian filter to eliminate image noise and avoid misidentification of impurities; S102, Enhancement: Improves the contrast between cells and background by histogram equalization, highlighting cell boundaries; S103. Segmentation: An adaptive threshold segmentation algorithm is used to separate the hemocytometer grid, algal cells, and background, eliminating interference.

[0008] Furthermore, the specific process of constructing an image recognition counting model and implementing automatic counting is as follows: S201. Obtain multiple sets of images of different types of algae on a hemocytometer, and mark the grid areas and cell positions on the images. After integration, use them as training samples, and divide the training samples into training set, validation set and test set according to an 8:1:1 ratio. S202. Construct an image recognition counting model based on the improved U-Net convolutional neural network, download the weight file and load it onto the corresponding network to initialize the transfer network parameters; S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the cell position mask and grid recognition result, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters. Retrain the entire network to obtain the image recognition counting model. S204. During the training process, small batches of samples are randomly and non-repeatedly drawn from the training set for training. One training cycle is completed after all training samples are drawn. The training is completed after a certain number of cycles, and the image recognition counting model is obtained. S205. Input the preprocessed hemocytometer image into the image recognition and counting model, output the cell position mask and grid recognition results, and automatically count the total number of cells in 25 grids. Calculate the algae concentration based on the hemocytometer specifications, and further reduce the error by averaging multiple fields of view to ensure that the counting accuracy of the validation set is greater than the preset deviation threshold. Algae concentration = total number of cells × dilution factor / (number of fields of view × number of small squares per field of view × volume of each small square).

[0009] Furthermore, the specific process for determining that growth is inhibited and triggering an abnormality alert is as follows: S301. Using the same detection procedure as the experimental group, obtain algal growth data of the normal control group, and generate the growth curve of the normal control group by the growth curve analysis module. S302. The growth curve analysis module sets a concentration judgment threshold based on the growth curve data of the normal control group and the concentration change rate during the logarithmic growth phase. The concentration judgment threshold represents the minimum healthy growth rate that algae should have under non-toxic stress. S303. Obtain the growth curve of the water sample to be tested. The growth curve analysis module determines the current growth stage of the algae through slope analysis. After determining that the algae in the water sample to be tested have entered the growth period, the growth curve analysis module will execute the following determination logic: The concentration change rate of algae growth curve in the water sample to be tested is calculated in real time, that is, the slope of the current growth curve, which is used to represent the concentration increment within a specific time interval. The real-time concentration change rate of the water sample to be tested is compared with the concentration judgment threshold. If the real-time concentration change rate is lower than the concentration judgment threshold, the algae in the water sample to be tested are determined to be in a state of inhibited growth. S304 The growth curve analysis module immediately triggers an abnormality alert signal to notify the operator that the water sample has potential toxicity. At the same time, the abnormality alert signal is sent to the parameter control module, which automatically generates temperature adjustment commands and light intensity adjustment commands according to the preset linkage mechanism.

[0010] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This algae-based comprehensive water toxicity detection system uses an AI model to identify cells and grids, combined with a multi-field averaging algorithm to reduce human error and avoid errors caused by subjective factors and fatigue of the counter, thereby improving the repeatability of the test and meeting the pesticide toxicity test standards. It can also achieve high-frequency automatic detection, generate continuous growth curves in real time, accurately capture the inflection point of algae growth inhibition, and adjust the test conditions when environmental parameters are abnormal to avoid test failures caused by parameter fluctuations and reduce retry costs.

[0011] 2. This algae-based comprehensive water toxicity detection system integrates temperature and light sensors to achieve complete synchronization of concentration, temperature, and light data timestamps, which are stored in a unified database. This solves the problem of scattered parameter recording and difficulty in tracing in traditional experiments. Furthermore, it adopts a unified counting standard and parameter recording specification to avoid methodological differences between different laboratories and operators, making the test results comparable across different regions and in line with pesticide environmental risk standards. Attached Figure Description

[0012] Figure 1 A schematic diagram of the overall structure of the present invention is shown. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Example: like Figure 1 As shown, a comprehensive water toxicity detection system based on algae includes an optical imaging module, a multi-parameter sensing module, an AI intelligent counting module, a growth curve analysis module, and a parameter control module, wherein: The optical imaging module consists of a biological microscope, a CCD industrial camera, and an electric focuser. It is used to acquire images from the hemocytometer and, based on the OpenCV algorithm library, preprocesses the acquired images before sending them to the AI ​​intelligent counting module. Preprocessing includes the following steps: S101, Denoising: Use a 5×5 Gaussian filter to eliminate image noise and avoid misidentification of impurities; S102, Enhancement: Improves the contrast between cells and background by histogram equalization, highlighting cell boundaries; S103. Segmentation: An adaptive threshold segmentation algorithm is used to separate the hemocytometer grid, algal cells, and background, eliminating interference.

[0015] The biological microscope uses a 100x objective lens to be fitted with a hemocytometer to ensure clear imaging of the grid lines and cells; The CCD camera has a resolution of ≥1920×1080 and a frame rate of ≥30fps. It connects to an industrial computer via USB to achieve real-time image transmission. The electric focusing device is used in conjunction with the biological camera, and the software controls the autofocus to avoid the deviation of manual focusing. The multi-parameter sensing module integrates temperature and light sensors and is installed in the algae culture chamber to acquire temperature and light data, which are then sent to the experimental control module. The temperature sensor has a measurement range of -40 to 125℃ and an accuracy of ±0.2℃. It uses an RS485 bus to transmit data and is used to acquire the temperature of the culture environment in real time. The light sensor has a measurement range of 0–1000 μmol / m 2 •s, with an accuracy of ±5%, used for synchronous recording of light intensity data; It also includes a data management module, which uses a MySQL database to store data and synchronizes the three parameters of concentration, temperature and light intensity with timestamps. It supports real-time viewing of data curves and exporting data in Excel format. At the same time, it automatically generates a standardized report after the experiment, which includes IC50 calculation results and parameter change trend graphs.

[0016] The test control module includes a PID temperature control submodule and an LED illumination adjustment submodule, wherein: The PID temperature control submodule is used to connect to the incubator temperature control element, with a temperature control range of 1~0℃ and a temperature control accuracy of ±0.1℃. It can receive software commands to adjust the temperature. The LED illumination adjustment submodule is used to connect to the LED light source inside the incubator, and the light intensity adjustment range is 0~800μmol / m². 2 •s, supports setting the illumination cycle via software.

[0017] The AI ​​intelligent counting module uses an improved U-Net convolutional neural network to build and train an image recognition counting model to achieve automatic counting. The specific process of constructing an image recognition counting model and implementing automatic counting is as follows: S201. Obtain multiple sets of images of different types of algae on a hemocytometer, and mark the grid areas and cell positions on the images. After integration, use them as training samples, and divide the training samples into training set, validation set and test set according to an 8:1:1 ratio. S202. Construct an image recognition counting model based on the improved U-Net convolutional neural network, download the weight file and load it onto the corresponding network to initialize the transfer network parameters; S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the cell position mask and grid recognition result, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters. Retrain the entire network to obtain the image recognition counting model. S204. During the training process, small batches of samples are randomly and non-repeatedly drawn from the training set for training. One training cycle is completed after all training samples are drawn. The training is completed after a certain number of cycles, and the image recognition counting model is obtained. S205. Input the preprocessed hemocytometer image into the image recognition and counting model, output the cell position mask and grid recognition results, and automatically count the total number of cells in 25 grids. Calculate the algae concentration based on the hemocytometer specifications, and further reduce the error by averaging multiple fields of view to ensure that the counting accuracy of the validation set is greater than the preset deviation threshold. Algae concentration = total number of cells × dilution factor / (number of fields of view × number of small squares per field of view × volume of each small square).

[0018] The growth curve analysis module is used to automatically generate algal growth curves based on real-time counted concentration data, and to determine the growth stage through slope analysis. It sets a concentration judgment threshold based on the normal control group. If the concentration change rate is lower than the concentration judgment threshold, it is determined that the growth is inhibited and an abnormality alert is triggered. The specific process for determining that growth is inhibited and triggering an abnormal alert is as follows: S301. Using the same detection procedure as the experimental group, obtain algal growth data of the normal control group, and generate the growth curve of the normal control group by the growth curve analysis module. S302. The growth curve analysis module sets a concentration judgment threshold based on the growth curve data of the normal control group and the concentration change rate during the logarithmic growth phase. The concentration judgment threshold represents the minimum healthy growth rate that algae should have under no toxic stress. S303. Obtain the growth curve of the water sample to be tested. The growth curve analysis module determines the current growth stage of the algae through slope analysis. After determining that the algae in the water sample to be tested have entered the growth period, the growth curve analysis module will execute the following judgment logic: The concentration change rate of algae growth curve in the water sample to be tested is calculated in real time, that is, the slope of the current growth curve, which is used to represent the concentration increment within a specific time interval. The real-time concentration change rate of the water sample to be tested is compared with the concentration judgment threshold. If the real-time concentration change rate is lower than the concentration judgment threshold, the algae in the water sample to be tested are determined to be in a state of inhibited growth. S304 The growth curve analysis module immediately triggers an abnormality alert signal to notify the operator that the water sample has potential toxicity. At the same time, the abnormality alert signal is sent to the parameter control module, which automatically generates temperature adjustment commands and light intensity adjustment commands according to the preset linkage mechanism.

[0019] The parameter control module is used to establish a linkage mechanism between growth status and environmental parameters. When the generated curve shows abnormality, it automatically generates temperature adjustment commands and light intensity adjustment commands and sends them to the experimental control module. It also supports a manual intervention mode, where users can manually modify parameter thresholds according to software prompts.

[0020] The specific implementation process is as follows: (1) Image acquisition: When the set time is reached, the software controls the electric focusing device to automatically focus, and the CCD camera takes an image of the algae sample on the hemocytometer (the sample is added to the counting plate by a 0.1 mL pipette and covered with a coverslip), and the image is transmitted to the computer in real time; (2) AI counting: The image preprocessing module completes denoising and segmentation, the AI ​​model identifies cells and counts them, and calculates the concentration. The counts in the three fields of view are 245, 252 and 248 respectively, with an average of 248. The concentration = 248 × 10 (dilution factor) / (3 × 16 × 0.1) = 5167 cells / mL; (3) Parameter monitoring: The sensor transmits temperature and light data in real time, and the software synchronously stores the data in the MySQL database to update the growth curve; (4) Dynamic control: At the 24th hour of the experiment, the software monitored that the concentration change rate of the 100mg / L pesticide group was only 60% of that of the control group, that is, the growth was severely inhibited. At the same time, the temperature dropped to 24.3℃, and the software automatically sent a command to the PID temperature control module to heat up to 25℃. (5) Data output: After the experiment lasts for 72 hours, the software automatically fits the growth curve, calculates the IC50 of each concentration group, and generates a standardized report containing the concentration curve, temperature / light curve and IC50 results.

[0021] Verify the results: 1. Counting accuracy verification: Twenty samples of Scenedesmus obliquus were selected and counted using both the system and manual methods. The results showed that the relative error between the average count of the system and the average count of the manual methods was ≤2%, which met the experimental accuracy requirements. 2. Real-time verification: During 72 hours of continuous monitoring, the system showed no data loss, and the growth curve showed a ≥95% agreement with the manually plotted curve from intermittent sampling. 3. Standardized Validation: Three laboratories used this system to test the same pesticide sample, IC... 50 The result deviation is ≤3%, which is much lower than the 10% deviation of traditional methods, proving the effectiveness of standardization.

[0022] This invention uses an AI model to identify cells and grids, combined with a multi-view averaging algorithm, to reduce human error and avoid errors caused by subjective factors and fatigue of the counter, thus improving experimental repeatability and meeting pesticide toxicity testing standards. It also enables high-frequency automatic detection, real-time generation of continuous growth curves, and precise capture of the inflection point of algal growth inhibition. When environmental parameters are abnormal, experimental conditions can be adjusted to avoid experimental failures due to parameter fluctuations, reducing retrial costs. Furthermore, it integrates temperature and light sensors to achieve complete synchronization of concentration, temperature, and light data timestamps, storing them in a unified database. This solves the problems of scattered parameter recording and difficulty in traceability in traditional experiments. Moreover, it adopts a unified counting standard and parameter recording specification to avoid methodological differences between different laboratories and operators, making experimental results horizontally comparable and compliant with pesticide environmental risk standards.

[0023] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0024] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A comprehensive water toxicity detection system based on algae, characterized in that, It includes an optical imaging module, a multi-parameter sensing module, an AI intelligent counting module, a growth curve analysis module, and a parameter control module, among which: The optical imaging module consists of a biological microscope, a CCD industrial camera, and an electric focuser. It is used to acquire images from the hemocytometer and, based on the OpenCV algorithm library, preprocesses the acquired images before sending them to the AI ​​intelligent counting module. The multi-parameter sensing module integrates temperature and light sensors and is installed in the algae culture chamber to acquire temperature and light data, which are then sent to the experimental control module. The AI ​​intelligent counting module uses an improved U-Net convolutional neural network to build and train an image recognition counting model to achieve automatic counting. The growth curve analysis module is used to automatically generate algal growth curves based on real-time counted concentration data, and to determine the growth stage through slope analysis. It sets a concentration judgment threshold based on the normal control group. If the concentration change rate is lower than the concentration judgment threshold, it is determined that the growth is inhibited and an abnormality alert is triggered. The parameter control module is used to establish a linkage mechanism between growth status and environmental parameters. When the generated curve shows abnormality, it automatically generates temperature adjustment commands and light intensity adjustment commands and sends them to the experimental control module. It also supports a manual intervention mode, where users can manually modify parameter thresholds according to software prompts.

2. The comprehensive water toxicity detection system based on algae according to claim 1, characterized in that, It also includes a data management module, which uses a MySQL database to store data, realizes the synchronization of the three parameters of concentration, temperature and light intensity timestamps, supports real-time viewing of data curves and export of data in Excel format, and automatically generates a standardized report after the experiment, including IC50 calculation results and parameter change trend graphs.

3. The comprehensive water toxicity detection system based on algae according to claim 1, characterized in that, The test control module includes a PID temperature control submodule and an LED illumination adjustment submodule, wherein: The PID temperature control submodule is used to connect to the incubator temperature control element, with a temperature control range of 1~0℃ and a temperature control accuracy of ±0.1℃. It can receive software commands to adjust the temperature. The LED illumination adjustment submodule is used to connect to the LED light source inside the incubator, and the light intensity adjustment range is 0~800μmol / m². 2 •s, supports setting the illumination cycle via software.

4. The comprehensive water toxicity detection system based on algae according to claim 1, characterized in that, The preprocessing includes the following steps: S101, Denoising: Use a 5×5 Gaussian filter to eliminate image noise and avoid misidentification of impurities; S102, Enhancement: Improves the contrast between cells and background by histogram equalization, highlighting cell boundaries; S103. Segmentation: An adaptive threshold segmentation algorithm is used to separate the hemocytometer grid, algal cells, and background, eliminating interference.

5. The comprehensive water toxicity detection system based on algae according to claim 1, characterized in that, The specific process of constructing an image recognition counting model and implementing automatic counting is as follows: S201. Obtain multiple sets of images of different types of algae on a hemocytometer, and mark the grid areas and cell positions on the images. After integration, use them as training samples, and divide the training samples into training set, validation set and test set according to an 8:1:1 ratio. S202. Construct an image recognition counting model based on the improved U-Net convolutional neural network, download the weight file and load it onto the corresponding network to initialize the transfer network parameters; S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the cell position mask and grid recognition result, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters. Retrain the entire network to obtain the image recognition counting model. S204. During the training process, small batches of samples are randomly and non-repeatedly drawn from the training set for training. One training cycle is completed after all training samples are drawn. The training is completed after a certain number of cycles, and the image recognition counting model is obtained. S205. Input the preprocessed hemocytometer image into the image recognition counting model, output the cell location mask and grid recognition results, automatically count the total number of cells in 25 grids, calculate the algae concentration based on the hemocytometer specifications, and further reduce the error by averaging multiple fields of view to ensure that the counting accuracy of the validation set is greater than the preset deviation threshold.

6. The comprehensive water toxicity detection system based on algae according to claim 1, characterized in that, The specific process for determining that growth is inhibited and triggering an abnormal alert is as follows: S301. Using the same detection procedure as the experimental group, obtain algal growth data of the normal control group, and generate the growth curve of the normal control group by the growth curve analysis module. S302. The growth curve analysis module sets a concentration judgment threshold based on the growth curve data of the normal control group and the concentration change rate during the logarithmic growth phase. The concentration judgment threshold represents the minimum healthy growth rate that algae should have under non-toxic stress. S303. Obtain the growth curve of the water sample to be tested. The growth curve analysis module determines the current growth stage of the algae through slope analysis. After determining that the algae in the water sample to be tested have entered the growth period, the growth curve analysis module will execute the following determination logic: The concentration change rate of algae growth curve in the water sample to be tested is calculated in real time, that is, the slope of the current growth curve, which is used to represent the concentration increment within a specific time interval. The real-time concentration change rate of the water sample to be tested is compared with the concentration judgment threshold. If the real-time concentration change rate is lower than the concentration judgment threshold, the algae in the water sample to be tested are determined to be in a state of inhibited growth. S304 The growth curve analysis module immediately triggers an abnormality alert signal to notify the operator that the water sample has potential toxicity. At the same time, the abnormality alert signal is sent to the parameter control module, which automatically generates temperature adjustment commands and light intensity adjustment commands according to the preset linkage mechanism.

Images