Algae efficient cultivation control method and system

Abstract

The invention provides an efficient algae cultivation control method and system, and relates to the technical field of data processing, and the method comprises the steps: dividing a water body into a plurality of equal-interval depth hierarchies, obtaining hierarchy division data, calculating an oxygen concentration difference value between adjacent hierarchies and a gas diffusion trend value of a unit depth, determining a disturbance tension value of each hierarchie, and determining a disturbance tension value of each hierarchie; the disturbance ratio of each level is calculated, first control data is obtained by combining the oxygen concentration recovery time of each level in the historical period, the stirring frequency coefficient and the duration value are calculated, the oxygen concentration recovery amplitudes in different time periods are extracted, the actual disturbance response value is calculated, difference comparison is conducted on the actual disturbance response value and the disturbance tension data, and first control data are obtained; the method comprises the steps of obtaining disturbance error data, analyzing the change direction of the disturbance error, performing error calculation and direction superposition on the change direction, calculating an error score value of the disturbance error of each level, and performing iterative updating on the first control data. The device can dynamically adjust the stirring frequency to ensure balanced growth of algae.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for efficiently cultivating and controlling algae. Background Art

[0002] In the existing industrial cultivation process of algae, in order to achieve standardization and efficient output of algae species, a constant temperature and constant light environment as well as multi-parameter online monitoring and control systems such as dissolved oxygen, water temperature, and pH value are usually configured in the production workshop. These systems, combined with automatic agitators and bottom aeration devices, realize real-time regulation of the algae growth environment, enabling the algae to reproduce rapidly under constant water quality and physical conditions.

[0003] Certain algae species require maintaining a dissolved oxygen concentration between 6-8 mg / L, often requiring a combination of continuous bottom aeration and stirring. However, due to the gradient of dissolved oxygen concentration distribution at different depths in the water column, constant frequency stirring alone may not be able to eliminate localized overoxygen or hypoxic areas. For example, in large algae seed tanks, if the agitator speed is not adjusted according to the dynamic changes in water level and temperature, it may lead to overexposure in the upper layer and hypoxia at the bottom, affecting the balanced distribution and growth of algae. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for efficiently cultivating and controlling algae, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] In a first aspect, a method for efficiently cultivating and controlling algae comprises:

[0007] Collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer division data;

[0008] According to the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated, the disturbance tension value of each layer is determined, and the disturbance tension data is obtained;

[0009] Calculate the disturbance ratio of each level based on the disturbance tension data and temperature data, and calculate the stirring frequency coefficient and duration value in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data;

[0010] Extracting the oxygen concentration recovery amplitude in different time periods according to the first control data, calculating the actual disturbance response value, and performing a difference comparison between the actual disturbance response value and the disturbance tension data to obtain disturbance error data;

[0011] According to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed on it. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data;

[0012] According to the disturbance assessment data, the first control data is iteratively updated to generate the second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set.

[0013] Furthermore, based on the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated to determine the disturbance tension value of each layer, thereby obtaining disturbance tension data, including:

[0014] According to the hierarchical division data and oxygen concentration data, the oxygen concentration values ​​of adjacent levels are extracted, and the oxygen concentration differences between adjacent levels are calculated to obtain an oxygen concentration difference sequence;

[0015] Based on the concentration difference sequence and the equally spaced depths of each layer, the gas diffusion offset value of each layer unit depth is calculated to obtain the diffusion trend data;

[0016] Based on the oxygen concentration data, the mean and standard deviation of the oxygen concentration differences between each layer are calculated; based on the mean and the oxygen concentration difference between adjacent layers, the weighted density of the concentration difference of the current layer relative to the mean is calculated to obtain the disturbance density term; based on the gas diffusion offset value in the diffusion trend data, the abnormal degree of the current layer disturbance in the fluctuation range is calculated to obtain the concentration deviation fluctuation term; based on the oxygen concentration between adjacent layers, the gas diffusion coupling strength between the two layers is calculated to obtain the disturbance energy concentration term;

[0017] The disturbance density term, concentration deviation fluctuation term and disturbance energy aggregation term are weighted and fused to calculate the disturbance tension value of each level and obtain the disturbance tension data.

[0018] Furthermore, the disturbance ratio of each level is calculated based on the disturbance tension data and temperature data, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data, including:

[0019] According to the disturbance tension data, the maximum disturbance tension value and the average disturbance tension value are extracted, and the disturbance tension values ​​of each level are normalized to obtain a disturbance ratio sequence;

[0020] According to the disturbance ratio sequence and temperature data, the product of the disturbance ratio and the temperature value is calculated to obtain the thermal disturbance score data;

[0021] Based on the thermal disturbance score data, the difference in thermal disturbance scores between adjacent layers is calculated to obtain the water layer thermal disturbance difference sequence;

[0022] According to the water layer thermal disturbance difference sequence, the thermal disturbance difference of each layer and its corresponding thermal disturbance score are normalized to obtain the disturbance adjustment sequence;

[0023] According to the disturbance adjustment sequence, the adjustment level of each level is determined to obtain the disturbance adjustment factor data.

[0024] Furthermore, the disturbance ratio of each level is calculated based on the disturbance tension data and temperature data, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data, which also includes:

[0025] Obtain the oxygen concentration recovery time of each level in the historical period, and combine it with the disturbance adjustment factor data to calculate the basic frequency value to obtain the basic frequency data;

[0026] According to the basic frequency data, the minimum and maximum basic frequency values ​​are extracted, a numerical interval is determined, and the numerical interval is equally divided into multiple frequency level intervals;

[0027] According to the basic frequency data, the frequency level interval of each basic frequency value is determined, and the corresponding frequency level label is assigned to it to obtain the stirring frequency coefficient sequence;

[0028] According to the stirring frequency coefficient sequence and its corresponding disturbance ratio, the inverse product processing is performed to obtain the stirring time value;

[0029] Linear mapping is performed according to the stirring duration value to limit its maximum and minimum stirring duration values ​​to obtain a duration value, and first control data is obtained according to the duration value and the stirring frequency coefficient sequence.

[0030] Furthermore, based on the first control data, the oxygen concentration recovery amplitude in different time periods is extracted, the actual disturbance response value is calculated, and the difference between the actual disturbance response value and the disturbance tension data is compared to obtain disturbance error data, including:

[0031] According to the stirring frequency coefficient and duration value of each level in the first control data, the stirring device is controlled to perform stirring operation at the corresponding level, and the execution cycle number is recorded;

[0032] According to the execution cycle number, oxygen concentration data is collected in a fixed time window before and after the execution cycle, and the difference between the before and after data is calculated to obtain the oxygen concentration recovery amplitude sequence;

[0033] According to the oxygen concentration recovery amplitude sequence, the oxygen concentration recovery value per unit time at each level is standardized to obtain the actual disturbance response value data;

[0034] Performing difference calculation on the actual disturbance response value data and the disturbance tension value of each level in the disturbance tension data to obtain the actual disturbance difference sequence;

[0035] According to the actual disturbance difference sequence, the disturbance error value between the control execution result and the expected disturbance result is determined to obtain the disturbance error data.

[0036] Furthermore, based on the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data, including:

[0037] According to the disturbance error values ​​of each level in the disturbance error data, the disturbance error values ​​in the continuous execution cycles are extracted, and the error change values ​​between adjacent execution cycles are calculated to obtain the error change sequence;

[0038] According to the positive and negative signs of the error change values ​​in the error change sequence, the change direction of the current disturbance error is judged to obtain the disturbance direction sequence;

[0039] According to the disturbance direction sequence, the continuous segments with consistent directions in adjacent time periods are counted to obtain the number of segments with consistent directions;

[0040] Based on the disturbance error value and its mean, the standardized deviation score of the disturbance error is calculated; based on the error increment and error change direction of the disturbance error in adjacent periods in the time series, the disturbance error growth term is calculated; the second-order change ratio of the disturbance error is calculated to obtain the disturbance trend growth term; based on the standardized deviation score term, the disturbance error growth term and the disturbance trend growth term, the dynamic offset fluctuation score term is obtained;

[0041] According to the change direction of the disturbance error and the amount of direction consistency segments, the direction consistency item is calculated; according to the proportion of the error amplitude relative to the overall disturbance system, the disturbance amplitude normalization evaluation item is calculated; according to the direction consistency item and the disturbance amplitude normalization evaluation item, the trend amplitude adjustment score item is obtained;

[0042] The dynamic offset fluctuation scoring item and the trend amplitude adjustment scoring item are integrated to calculate the error score value of each level and obtain the disturbance assessment data.

[0043] Furthermore, the first control data is iteratively updated according to the disturbance assessment data to generate second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set, including:

[0044] Obtaining a control update factor set based on the error score values ​​of each level in the disturbance assessment data and the stirring frequency coefficient and duration value of the corresponding level in the first control data;

[0045] According to each element in the control update factor set, parameter iteration processing is performed on the stirring frequency coefficient and the duration value in the first control data to obtain an updated control parameter set;

[0046] According to the updated control parameter set, the target frequency value, the target duration value and the control cycle number are respectively encapsulated at each level to obtain second control data;

[0047] According to the second control data, the control parameters of each level are combined and packaged into an instruction format to generate a control instruction, which is then output to the stirring device. The execution time, level number, and parameter value are recorded to obtain a control data set.

[0048] In a second aspect, a high-efficiency algae cultivation control system is provided, comprising:

[0049] The water body segmentation module is used to collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer segmentation data;

[0050] The disturbance tension module is used to calculate the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth based on the layer division data and oxygen concentration data, determine the disturbance tension value of each layer, and obtain the disturbance tension data;

[0051] A control parameter module is used to calculate the disturbance ratio of each layer according to the disturbance tension data and the temperature data, and calculate the stirring frequency coefficient and the duration value in combination with the oxygen concentration recovery time of each layer in the historical period to obtain the first control data;

[0052] A disturbance error module is used to extract the oxygen concentration recovery amplitude in different time periods according to the first control data, calculate the actual disturbance response value, and compare the actual disturbance response value with the disturbance tension data to obtain disturbance error data;

[0053] The error evaluation module is used to analyze the change direction of the disturbance error based on the disturbance error data, perform error calculation and direction superposition, calculate the error score value of the disturbance error at each level, and obtain the disturbance evaluation data;

[0054] The parameter updating module is used to iteratively update the first control data according to the disturbance assessment data, generate the second control data, and output the control instructions generated therefrom to the stirring device to obtain the control data set.

[0055] The above solution of the present invention includes at least the following beneficial effects:

[0056] This invention divides the water body into equally spaced vertical layers and combines oxygen concentration data from each layer to achieve three-dimensional structural modeling of the water body, forming the core foundation for dynamic control of the algae growth environment. By monitoring oxygen concentration in layers and further calculating the oxygen concentration difference between adjacent layers and the gas diffusion trend per unit depth, a disturbance tension model is formed. This fundamentally quantifies the disturbance demand at each layer in the water body, constructing a disturbance control system with higher spatial resolution and greater control accuracy. Subsequent stirring control actions can be differentiated according to the disturbance characteristics of different layers, so that the dissolved oxygen concentration of each layer can be stably controlled within the set range, thereby ensuring that the algae population is evenly distributed throughout the entire reactor volume.

[0057] The present invention quantifies the abnormal density of oxygen concentration differences, the degree of deviation fluctuation, and the intensity of gas diffusion coupling between adjacent layers, respectively reflecting the local gradient characteristics of water disturbances, the deviation of disturbance trends, and the strength of diffusion coupling, and obtaining the disturbance tension value of each layer. This not only makes the analysis of disturbance sources more scientific, but also enables the stirring control system to obtain a more accurate disturbance magnitude assessment before execution, thereby formulating a more refined and dynamic stirring strategy, which can effectively improve the system's disturbance response capability and regulation effect.

[0058] The present invention realizes adaptive optimization of stirring frequency and duration by introducing historical cycle oxygen concentration recovery time. Specifically, the maximum and average disturbance tension values ​​of each level are extracted, and the thermal disturbance score and adjustment level are obtained in combination with temperature data, thereby determining the stirring frequency coefficient sequence under different frequency level intervals. Then, the stirring duration of each level is calculated to ensure that areas with greater disturbance intensity can obtain local disturbances with higher frequency and shorter duration, while areas with less disturbance intensity adopt a lower frequency and longer duration control mode, which significantly improves the utilization efficiency of stirring energy consumption and reduces the operating load of the stirring motor while ensuring the effectiveness of control.

[0059] The present invention establishes a dynamic disturbance prediction and adjustment scoring system through trend analysis of disturbance errors. It not only describes the rate of change of disturbance errors, but also reflects the fluctuation trend of errors in continuous cycles, which helps to identify potential disturbance stability risks. The system can judge whether there is a systematic deviation in the current control strategy, so as to correct the control strategy in advance and avoid the disturbance out-of-control problem caused by continuous error accumulation. The system can quantify the intensity and direction consistency of disturbance fluctuations, providing a data basis for subsequent control iterations.

[0060] This invention iteratively optimizes the original control parameters by constructing a set of control update factors, generating updated secondary control data. This completes a closed-loop system for disturbance control, from modeling, execution, feedback, and re-optimization. This system can gradually approach the optimal control strategy over time, reducing the need for manual adjustments and significantly improving the adaptability of disturbance control under diverse environmental conditions. This control mechanism provides strong technical support for intelligent algae cultivation and exhibits excellent versatility and portability. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flowchart of an efficient algae cultivation control method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0063] like Figure 1 As shown, an embodiment of the present invention provides a method for efficiently cultivating and controlling algae, the method comprising:

[0064] Collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer division data;

[0065] According to the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated, the disturbance tension value of each layer is determined, and the disturbance tension data is obtained;

[0066] Calculate the disturbance ratio of each level based on the disturbance tension data and temperature data, and calculate the stirring frequency coefficient and duration value in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data;

[0067] Extracting the oxygen concentration recovery amplitude in different time periods according to the first control data, calculating the actual disturbance response value, and performing a difference comparison between the actual disturbance response value and the disturbance tension data to obtain disturbance error data;

[0068] According to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed on it. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data;

[0069] According to the disturbance assessment data, the first control data is iteratively updated to generate the second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set.

[0070] In an embodiment of the present invention, water level data, oxygen concentration data, and temperature data are collected and vertically segmented based on the water level data to divide the water body into multiple equally spaced depth levels, obtaining layer division data. This constructs a basic model of the three-dimensional structure of the water body, enabling subsequent operations to independently analyze and process regions at different depths, effectively avoiding the problem of local control failure caused by the overall processing approach in traditional systems. Based on the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent levels and the gas diffusion trend value per unit depth are calculated to determine the disturbance tension value for each level, obtaining disturbance tension data. This comprehensively characterizes the non-uniformity and directionality of water disturbances, accurately identifies the disturbance requirements of each level, and provides a scientific basis for subsequent adaptive stirring control. Based on the disturbance tension data and temperature data, the disturbance ratio of each level is calculated. Combined with the oxygen concentration recovery time of each level within a historical period, the stirring frequency coefficient and duration value are calculated to obtain first control data. This transforms the stirring frequency and duration from static parameters into dynamically adjustable parameters, improving the accuracy and energy efficiency of the stirring operation and adapting to the changing requirements of algae metabolic rates under different temperature conditions.

[0071] According to the first control data, the oxygen concentration recovery amplitude in different time periods is extracted, the actual disturbance response value is calculated, and the difference is compared with the disturbance tension data to obtain the disturbance error data. A comparison mechanism between the theoretical disturbance demand and the actual disturbance effect is established, and an objective standard for evaluating whether the stirring strategy is effective is provided; according to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed on it. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data, accurately identify the disturbance imbalance level and the long-term control deviation trend, and provide direction for subsequent optimization parameters; according to the disturbance evaluation data, the first control data is iteratively updated to generate the second control data, and the control instructions generated based on it are output to the stirring device to obtain the control data set, so that the control strategy can continuously optimize itself during continuous execution.

[0072] Among them, water level data, oxygen concentration data and temperature data are collected, and vertical segmentation processing is performed based on the water level data to divide the water body into multiple equally spaced depth levels, and the hierarchical division data is obtained, which specifically includes:

[0073] Multiple sensor modules installed within the algae cultivation container collect parameter data in real time. Water level data is continuously detected by an ultrasonic water level sensor located above the container. This sensor is set to measure at least once every 30 seconds, with an accuracy of better than ±1mm to accurately capture dynamic water level changes. A wave shield is installed at the sensor installation location to reduce measurement errors caused by water surface fluctuations. The electrical signals collected by the sensor are digitized by an A / D conversion module, ultimately forming the actual water level data within the container.

[0074] The collection of oxygen concentration data relies on dissolved oxygen optical probes, which are arranged in a multi-point vertical layout. That is, multiple fixed probe brackets are set along the depth direction of the inner wall of the container, and multiple dissolved oxygen sensors are installed on the brackets. The intervals between each sensor are set to 20cm~30cm. The measurement range covers the entire water depth. Each probe will collect the instantaneous oxygen concentration value at the layer where it is located and transmit the data to the central processing unit via wired / wireless means. To ensure data synchronization, the sampling time windows of all dissolved oxygen probes remain consistent, and the data within each sampling period are marked with depth position and timestamp.

[0075] The method of acquiring temperature data is similar to that of oxygen concentration. PT100 temperature sensors are deployed at multiple points and installed on an independent bracket next to the dissolved oxygen sensor. The temperature sensor is connected to the temperature measurement module through a three-wire system. The temperature sampling frequency is synchronized with the oxygen concentration. The temperature data generated within a unit sampling period will correspond one-to-one with the depth at which it is located, and is used to describe the temperature distribution of each water layer. Since algae are sensitive to temperature changes, the sensors are equipped with an automatic calibration function to ensure that the measurement error is controlled within ±0.2°C.

[0076] After completing the above data collection, the system analyzes the water level data to obtain the total depth value from the current liquid surface to the bottom, and then divides the depth interval into equal intervals, with the division interval being a set value (such as 20cm). The total number of levels is obtained based on the depth interval and the division interval. The system marks each level with a number to form a hierarchical index sequence. At the same time, based on the multi-point distributed oxygen concentration and temperature sensor data, the depth of each sensor point is matched with the nearest level, thereby realizing the binding and integration of oxygen concentration data and temperature data in the hierarchical structure. Finally, the hierarchical division data is generated in units of "level number + corresponding depth + oxygen concentration and temperature data at that depth".

[0077] In a preferred embodiment of the present invention, based on the layer division data and the oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated, and the disturbance tension value of each layer is determined to obtain the disturbance tension data, including:

[0078] According to the hierarchical division data and oxygen concentration data, the oxygen concentration values ​​of adjacent levels are extracted, and the oxygen concentration differences between adjacent levels are calculated to obtain an oxygen concentration difference sequence;

[0079] Based on the concentration difference sequence and the equally spaced depths of each layer, the gas diffusion offset value of each layer unit depth is calculated to obtain the diffusion trend data;

[0080] Based on the oxygen concentration data, the mean and standard deviation of the oxygen concentration differences between each layer are calculated; based on the mean and the oxygen concentration difference between adjacent layers, the weighted density of the concentration difference of the current layer relative to the mean is calculated to obtain the disturbance density term; based on the gas diffusion offset value in the diffusion trend data, the abnormal degree of the current layer disturbance in the fluctuation range is calculated to obtain the concentration deviation fluctuation term; based on the oxygen concentration between adjacent layers, the gas diffusion coupling strength between the two layers is calculated to obtain the disturbance energy concentration term;

[0081] The disturbance density term, concentration deviation fluctuation term and disturbance energy aggregation term are weighted and fused to calculate the disturbance tension value of each level and obtain the disturbance tension data.

[0082] In an embodiment of the present invention, based on the hierarchical division data and oxygen concentration data, the oxygen concentration values ​​of adjacent layers are extracted, and the oxygen concentration differences between adjacent layers are calculated to obtain an oxygen concentration difference sequence, which quantifies the oxygen concentration gradient between layers and provides an important basis for subsequent disturbance intensity judgment; based on the concentration difference sequence and the equally spaced depths of each layer, the gas diffusion offset value of the unit depth of each layer is calculated to obtain diffusion trend data, which reflects the intensity of the concentration gradient change within the unit depth and makes up for the limitation of using only the original difference for judgment; based on the oxygen concentration data, the mean and standard deviation of the oxygen concentration difference between each layer are calculated to determine the central trend and fluctuation range of the concentration change in the current vertical distribution of the water body; based on the mean and the oxygen concentration difference between adjacent layers, the relative value of the concentration difference of the current layer to the mean is calculated. Weighted density is used to obtain the disturbance density term, which reflects whether there is a sudden change point in oxygen concentration at a certain level. It is an important means to judge the local high gradient area. According to the gas diffusion offset value in the diffusion trend data, the abnormal degree of the current level disturbance in the fluctuation range is calculated to obtain the concentration deviation fluctuation term, which accurately identifies the abnormal diffusion phenomenon and determines whether the water body is affected by external sudden disturbances. According to the oxygen concentration between adjacent levels, the gas diffusion coupling strength between the two layers is calculated to obtain the disturbance energy accumulation term, which determines whether there is a strong diffusion trend at certain levels. The disturbance density term, concentration deviation fluctuation term and disturbance energy accumulation term are weighted and fused to calculate the disturbance tension value of each level and obtain the disturbance tension data, which comprehensively reflects the strength of the disturbance driving force at each level and serves as the direct basis for setting the subsequent stirring frequency and duration.

[0083] The calculation formula of the disturbance tension value is: ,

[0084] in, For the The disturbance tension value of the layer, is the index of the level, For the The oxygen concentration difference between the layers, , For the The oxygen concentration value of the level, For the The oxygen concentration value of the level, is the equally spaced depth between levels, is the average of the oxygen concentration differences at each level, is the standard deviation of the oxygen concentration difference at each level, For the Gas diffusion offset value per unit depth of the layer, , is the weight coefficient.

[0085] in, is the perturbation density term, is the concentration deviation fluctuation term, is the disturbance energy aggregation term. is the weight coefficient, and its sum is 1.

[0086] During the initial inoculation period (the initial algae proliferation stage), the oxygen concentration is generally low, and the focus is on rapid identification of weak disturbances. During this stage, the algae have not yet formed a stable growth structure, and the oxygen concentration fluctuates slightly but is sensitive. A small concentration anomaly may affect the distribution uniformity. Therefore, it is necessary to enhance the detection of abnormal local concentration gradients, i.e., the disturbance density term. The response capability of the system can quickly identify potential interference points and implement early intervention. Diffusion fluctuations and coupling strength are not obvious at this stage, so their weights can be appropriately reduced. The values ​​are 0.6, 0.3 and 0.1.

[0087] During the rapid growth period (active algae metabolism phase), the growth rate is high, the algae density increases rapidly, and the dissolved oxygen fluctuates violently. During this phase, the photosynthesis and respiration of algae are significantly enhanced, resulting in dramatic fluctuations in water oxygen concentration between day and night or between the upper and lower layers. Frequent disturbance areas are very likely to appear. At this time, we should focus on responding to the fluctuation characteristics and strengthen the control of the water oxygen concentration. The identification of the system can be used to determine in advance whether the system may experience over-disturbance or under-disturbance. The density deviation is secondary to the treatment, and the energy coupling value also begins to show its importance, and the weight is moderately increased. The values ​​are 0.3, 0.5 and 0.2.

[0088] During the high-density stable period (algae distribution tends to be balanced), the water system is stable, but there is a risk of long-term accumulation of local concentrations. This stage has entered the stable operation period, and single-layer disturbances no longer occur frequently, but oxygen may remain in specific areas for a long time or migrate poorly, forming hidden dead spots. At this stage, the modeling of diffusion paths and migration trends should be strengthened, that is, attention should be paid to Calculation can be used to determine whether there is a potential for energy accumulation or migration imbalance, and timely directional disturbance adjustments can be made to improve the overall diffusion efficiency of the system. The values ​​are 0.2, 0.2 and 0.6.

[0089] In abnormal system conditions (such as equipment fluctuations, external disturbances), data changes are abnormal and the source of the disturbance needs to be quickly located. In this scenario, it is necessary to enhance the sensitivity of both local density and fluctuation terms to quickly locate the problem area in the early stages of the disturbance. The importance of coupling terms is relatively reduced because the system is not on the normal operating path. Error isolation and real-time response capabilities should be prioritized. The values ​​are 0.4, 0.4 and 0.2.

[0090] In a preferred embodiment of the present invention, the disturbance ratio of each level is calculated based on the disturbance tension data and temperature data, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data, including:

[0091] According to the disturbance tension data, the maximum disturbance tension value and the average disturbance tension value are extracted, and the disturbance tension values ​​of each level are normalized to obtain a disturbance ratio sequence;

[0092] According to the disturbance ratio sequence and temperature data, the product of the disturbance ratio and the temperature value is calculated to obtain the thermal disturbance score data;

[0093] Based on the thermal disturbance score data, the difference in thermal disturbance scores between adjacent layers is calculated to obtain the water layer thermal disturbance difference sequence;

[0094] According to the water layer thermal disturbance difference sequence, the thermal disturbance difference of each layer and its corresponding thermal disturbance score are normalized to obtain the disturbance adjustment sequence;

[0095] According to the disturbance adjustment sequence, the adjustment level of each level is determined to obtain the disturbance adjustment factor data.

[0096] In an embodiment of the present invention, according to the disturbance tension data, the maximum disturbance tension value and the average disturbance tension value are extracted, and the disturbance tension values ​​of each layer are normalized to obtain a disturbance ratio sequence, and the disturbance tension data is mapped to a dimensionless ratio, which effectively improves the stability and versatility of subsequent calculations; according to the disturbance ratio sequence and temperature data, the product of the disturbance ratio and the temperature value is calculated to obtain thermal disturbance score data, and the temperature is introduced into the adjustment logic of the disturbance control parameter to avoid the problem of over-disturbance caused by mistaken strong stirring under high temperature conditions; according to the thermal disturbance score data, the difference in thermal disturbance scores between adjacent layers is calculated to obtain a water layer thermal disturbance difference sequence, and the thermal disturbance score difference is identified. The spatial variation gradient of the disturbance score is used to discover possible abnormal areas of disturbance transition, providing a basic basis for subsequent disturbance regulation; according to the water layer thermal disturbance difference sequence, the thermal disturbance differences of each layer and their corresponding thermal disturbance scores are normalized to obtain a disturbance regulation sequence, which integrates the severity of the thermal disturbance and the disturbance intensity into a standardized indicator, removes the interference of different dimensions on the calculation results, and thus lays a unified standard for the hierarchical control of subsequent regulation levels; according to the disturbance regulation sequence, the regulation level of each layer is determined, and the disturbance regulation factor data is obtained to achieve refinement and differentiation of the systematic disturbance response, and improve the accuracy of local disturbance regulation and the coordination of overall control.

[0097] Among them, according to the disturbance ratio sequence and temperature data, the product of the disturbance ratio and the temperature value is calculated to obtain the thermal disturbance score data, which specifically includes:

[0098] The system simultaneously reads the disturbance ratio and temperature value for each layer, then directly multiplies the two to obtain a thermal disturbance score. This multiplication process not only considers the disturbance intensity of the water layer but also introduces temperature as a key factor, thereby comprehensively evaluating the sensitivity of each layer to the disturbance operation in the current physical environment. As a key factor affecting the oxygen diffusion rate, temperature directly affects the physical response of the disturbance. Therefore, this multiplication method is physically reasonable and process-calculable. Finally, the system integrates the thermal disturbance scores of all layers into a thermal disturbance score dataset for subsequent difference calculation and adjustment level classification.

[0099] Among them, according to the water layer thermal disturbance difference sequence, the thermal disturbance difference of each layer and its corresponding thermal disturbance score are normalized to obtain the disturbance adjustment sequence, which specifically includes:

[0100] The system calculates the score difference for each pair of adjacent levels to form a thermal disturbance difference sequence. The difference data reflects the degree of thermal disturbance mutation in the vertical direction of the water body. In order to uniformly evaluate the thermal disturbance difference and the thermal disturbance score intensity, the system needs to further construct a disturbance adjustment sequence. First, the absolute value operation of the difference sequence is performed to reflect its change intensity. Then, the thermal disturbance difference and its corresponding score value are jointly normalized. The system calculates all thermal disturbance differences separately. The maximum value and rating values The maximum value Perform standardized operations, , the obtained disturbance adjustment value It can be used to comprehensively evaluate the regulation priority or regulation intensity of each level.

[0101] In a preferred embodiment of the present invention, the disturbance ratio of each level is calculated based on the disturbance tension data and temperature data, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data, which also includes:

[0102] Obtain the oxygen concentration recovery time of each level in the historical period, and combine it with the disturbance adjustment factor data to calculate the basic frequency value to obtain the basic frequency data;

[0103] According to the basic frequency data, the minimum and maximum basic frequency values ​​are extracted, a numerical interval is determined, and the numerical interval is equally divided into multiple frequency level intervals;

[0104] According to the basic frequency data, the frequency level interval of each basic frequency value is determined, and the corresponding frequency level label is assigned to it to obtain the stirring frequency coefficient sequence;

[0105] According to the stirring frequency coefficient sequence and its corresponding disturbance ratio, the inverse product processing is performed to obtain the stirring time value;

[0106] Linear mapping is performed according to the stirring duration value to limit its maximum and minimum stirring duration values ​​to obtain a duration value, and first control data is obtained according to the duration value and the stirring frequency coefficient sequence.

[0107] In an embodiment of the present invention, the oxygen concentration recovery time of each level in the historical period is obtained, and the basic frequency value is calculated in combination with the disturbance adjustment factor data to obtain the basic frequency data, thereby realizing a dynamic assignment mechanism for the stirring frequency and effectively avoiding the incompatibility problem caused by the same frequency control for different levels; according to the basic frequency data, the minimum and maximum basic frequency values ​​are extracted, the numerical interval is determined, and the numerical interval is equally divided into multiple frequency level intervals. The continuous frequency value is discretized by equally dividing the intervals, which provides a basis for subsequent parameter mapping and frequency grading; according to the basic frequency data, the frequency level interval in which each basic frequency value is located is determined, and its corresponding frequency level is assigned. The label is obtained to obtain the stirring frequency coefficient sequence, and the continuous frequency value is converted into a discrete label and mapped into a frequency coefficient, so as to accurately identify the stirring intensity to be executed at each level and avoid the execution error caused by the high precision of the frequency data; according to the stirring frequency coefficient sequence and its corresponding disturbance ratio, an inverse product processing is performed to obtain the stirring time value, thereby realizing the dynamic adjustment of the stirring time and avoiding the water quality fluctuation caused by the disturbance redundancy; according to the stirring time value, a linear mapping processing is performed to limit its maximum and minimum stirring time values ​​to obtain the continuous time value, and according to the continuous time value and the stirring frequency coefficient sequence, the first control data is obtained to ensure the executability and stability of the stirring control strategy.

[0108] Among them, the oxygen concentration recovery time of each level in the historical period is obtained, and combined with the disturbance adjustment factor data, the basic frequency value is calculated to obtain the basic frequency data, which specifically includes:

[0109] First, the oxygen concentration changes at each water layer over multiple historical control cycles must be recorded. During each control cycle, after the agitator is activated, the system uses an oxygen concentration sensor to record in real time the time it takes for the oxygen concentration in that layer to rise from its initial state to the target oxygen concentration (e.g., set to 6 mg / L). This time is defined as the oxygen concentration recovery time for that layer. This data is stored in a structured format along with the execution cycle number and layer number to represent the oxygen response performance of that layer during that cycle. The recovery time is then combined with the disturbance adjustment factor data obtained in the previous stage. The disturbance adjustment factor is pre-classified based on parameters such as temperature gradient, disturbance intensity, and historical change trends. For example, three disturbance levels (low, medium, and high) are set, each with a corresponding adjustment weight. The system uses this disturbance adjustment factor as a weighting parameter to adjust the oxygen concentration recovery time, thereby quantifying the base frequency requirement for the current layer. The adjusted result is converted into a numerical base frequency value, which is then aggregated into a complete base frequency data sequence.

[0110] Among them, according to the stirring frequency coefficient sequence and its corresponding disturbance ratio, an inverse product process is performed to obtain the stirring time value, which specifically includes:

[0111] The disturbance ratio indicates the relative proportion of the current disturbance intensity of each level in the total level. The frequency coefficient indicates the stirring frequency that should be adopted by the level within a certain frequency level range. In order to achieve high-frequency stirring in a shorter time in the area with strong disturbance and low-frequency stirring in a longer time in the area with weak disturbance, the system will use the frequency coefficient Ratio to disturbance Perform the inverse product operation, that is, the formula Calculate, where is the original stirring time of this level, is a configurable proportional constant that controls the magnitude of the overall duration. The system performs this calculation for each level in turn to obtain a preliminary estimate of the stirring duration.

[0112] Among them, linear mapping processing is performed according to the stirring time value, and its maximum and minimum stirring time values ​​are limited to obtain the duration value, and the first control data is obtained according to the duration value and the stirring frequency coefficient sequence, which specifically includes:

[0113] After the mixing time value is processed by the inverse product, there may be extremely large or small abnormal values. If these values ​​are used directly, it may interfere with the execution of the control system and even affect the mechanical life of the equipment. Therefore, the system needs to perform linear mapping processing on the mixing time value and control all time values ​​within the preset execution range, such as setting the minimum value to 5 seconds and the maximum value to 90 seconds. Specifically, the system will convert each original mixing time into a linear mapping. If the input value exceeds the upper limit of the linear interval, it is forced to be the maximum value; if it is below the lower limit, it is assigned the minimum value; the original value remains unchanged in the rest of the range, thus obtaining a new stirring duration value. The system then combines this duration value with the obtained stirring frequency coefficient and combines it with identification information such as the level number and control cycle number to form the first control data.

[0114] In a preferred embodiment of the present invention, based on the first control data, the oxygen concentration recovery amplitude in different time periods is extracted, the actual disturbance response value is calculated, and the difference between the actual disturbance response value and the disturbance tension data is compared to obtain disturbance error data, including:

[0115] According to the stirring frequency coefficient and duration value of each level in the first control data, the stirring device is controlled to perform stirring operation at the corresponding level, and the execution cycle number is recorded;

[0116] According to the execution cycle number, oxygen concentration data is collected in a fixed time window before and after the execution cycle, and the difference between the before and after data is calculated to obtain the oxygen concentration recovery amplitude sequence;

[0117] According to the oxygen concentration recovery amplitude sequence, the oxygen concentration recovery value per unit time at each level is standardized to obtain the actual disturbance response value data;

[0118] Performing difference calculation on the actual disturbance response value data and the disturbance tension value of each level in the disturbance tension data to obtain the actual disturbance difference sequence;

[0119] According to the actual disturbance difference sequence, the disturbance error value between the control execution result and the expected disturbance result is determined to obtain the disturbance error data.

[0120] In an embodiment of the present invention, according to the stirring frequency coefficient and duration value of each level in the first control data, the stirring device is controlled to perform stirring operation at the corresponding level, and the execution cycle number is recorded, and the precise disturbance process is started, so that the stirring device no longer operates at a fixed rhythm, but performs differentiated disturbance in response to the disturbance tension characteristics, thereby providing a prerequisite for subsequent recovery monitoring and effect evaluation; according to the execution cycle number, oxygen concentration data is collected within a fixed time window before and after the execution cycle, and the difference between the before and after data is calculated to obtain the oxygen concentration recovery amplitude sequence, quantify the direct effect of the disturbance operation on the change of dissolved oxygen in the water body, and provide a reliable data basis for subsequent disturbance response analysis; according to the oxygen Concentration recovery amplitude sequence, standardizes the oxygen concentration recovery value per unit time of each level to obtain the actual disturbance response value data, eliminates the deviation caused by the different duration of stirring at different levels, and realizes horizontal comparison of disturbance response levels across levels; performs difference calculation on the actual disturbance response value data and the disturbance tension value of each level in the disturbance tension data to obtain the actual disturbance difference sequence, quantifies the difference between the actual disturbance effect and the predicted disturbance demand, and reflects whether the stirring measures have achieved the expected goal; according to the actual disturbance difference sequence, determines the disturbance error value between the control execution result and the expected disturbance result, and obtains the disturbance error data, which reflects the effect of the disturbance operation in a single cycle.

[0121] Among them, according to the actual disturbance difference sequence, the disturbance error value between the control execution result and the expected disturbance result is determined to obtain the disturbance error data, which specifically includes:

[0122] First, each difference item in the difference sequence is processed to identify its relative degree of deviation within the current control context. To enhance the adaptability and comparability of error assessment, the system constructs a reference range for expected disturbance changes based on the distribution of disturbance tension values ​​across all levels. For example, the mean and standard deviation of disturbance tension across all levels can be used to set a reasonable fluctuation range. If a difference item exceeds the range of the mean plus the standard deviation, it can be considered a highly sensitive disturbance difference and warrants special attention.

[0123] These difference items are then aggregated by layer, labeled with their execution cycle numbers, and sorted by execution order to form a disturbance error time series. On this basis, dynamic weighting factors are further introduced to strengthen the assessment of the error impact of key layers or high-frequency disturbance areas. For example, the error difference of a layer with high disturbance tension values ​​can be given a higher weight because it has a more direct impact on the overall system stability.

[0124] During the error calculation process, the system also combines control parameters such as the stirring frequency coefficient and duration in the current control data to perform causal relationship inference. If the disturbance difference at a certain level is consistently high and the corresponding stirring frequency deviates significantly from the normal value range, it can be preliminarily determined that the disturbance intensity configuration is abnormal. Conversely, if the frequency is moderate but the error is still significant, it may be caused by changes in the external environment or a delayed system response, suggesting the need to strengthen the environmental perception module or optimize the model structure. Finally, after integrating the above calculations with the results of logical reasoning, the disturbance error data is generated.

[0125] In a preferred embodiment of the present invention, according to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed on it, and the error score value of the disturbance error at each level is calculated to obtain the disturbance assessment data, including:

[0126] According to the disturbance error values ​​of each level in the disturbance error data, the disturbance error values ​​in the continuous execution cycles are extracted, and the error change values ​​between adjacent execution cycles are calculated to obtain the error change sequence;

[0127] According to the positive and negative signs of the error change values ​​in the error change sequence, the change direction of the current disturbance error is judged to obtain the disturbance direction sequence;

[0128] According to the disturbance direction sequence, the continuous segments with consistent directions in adjacent time periods are counted to obtain the number of segments with consistent directions;

[0129] Based on the disturbance error value and its mean, the standardized deviation score of the disturbance error is calculated; based on the error increment and error change direction of the disturbance error in adjacent periods in the time series, the disturbance error growth term is calculated; the second-order change ratio of the disturbance error is calculated to obtain the disturbance trend growth term; based on the standardized deviation score term, the disturbance error growth term and the disturbance trend growth term, the dynamic offset fluctuation score term is obtained;

[0130] According to the change direction of the disturbance error and the amount of direction consistency segments, the direction consistency item is calculated; according to the proportion of the error amplitude relative to the overall disturbance system, the disturbance amplitude normalization evaluation item is calculated; according to the direction consistency item and the disturbance amplitude normalization evaluation item, the trend amplitude adjustment score item is obtained;

[0131] The dynamic offset fluctuation scoring item and the trend amplitude adjustment scoring item are integrated to calculate the error score value of each level and obtain the disturbance assessment data.

[0132] In an embodiment of the present invention, according to the disturbance error values ​​of each level in the disturbance error data, the disturbance error values ​​within the continuous execution cycle are extracted, and the error change values ​​between adjacent execution cycles are calculated to obtain an error change sequence, thereby realizing time series analysis of the disturbance error and providing a quantitative basis for subsequent judgment of whether the disturbance tends to be stable, deteriorated or fluctuates repeatedly; according to the positive and negative signs of the error change values ​​in the error change sequence, the change direction of the current disturbance error is judged to obtain a disturbance direction sequence, and it is identified whether the disturbance error is continuously increasing, decreasing or maintaining a stable state, thereby distinguishing the type of disturbance change; according to the disturbance direction sequence, the continuous segments with consistent directions in adjacent time periods are counted to obtain the direction consistent segment quantity; according to the disturbance error value and its mean, the standardized deviation score item of the disturbance error is calculated to reflect the relative position of the current disturbance error in the overall historical distribution, providing a basis for subsequent control parameter adjustment; according to the error increment and error change direction of the disturbance error in adjacent cycles in the time series, the disturbance error growth item is calculated to identify whether the disturbance is intensifying; the second-order change ratio of the disturbance error is calculated to obtain the disturbance trend growth item, It reflects the trend of error aggravation; the dynamic offset fluctuation score item is obtained based on the standardized deviation score item, the disturbance error growth item and the disturbance trend growth item, which characterizes the deviation amplitude, growth rate and change trend of the current disturbance error, and is a key indicator for dynamically measuring disturbance stability; the direction consistency item is calculated based on the disturbance error change direction and the direction consistency segment, which reveals whether the disturbance trend is continuous and clear, and determines whether the error fluctuation presents a stable growth or stable decay pattern; the disturbance amplitude normalization evaluation item is calculated based on the proportion of the error amplitude relative to the overall disturbance system, which reflects the relative strength of the current error relative to the maximum error of the system; the trend amplitude adjustment score item is obtained based on the direction consistency item and the disturbance amplitude normalization evaluation item, which reflects the overall trend of the disturbance control strategy and provides a more comprehensive adjustment score for the system; the dynamic offset fluctuation score item and the trend amplitude adjustment score item are integrated to calculate the error score values ​​of each level to obtain the disturbance assessment data, which comprehensively reflects the actual execution effect, deviation degree and trend direction of the disturbance control at the current level, and is the core input data for subsequent control parameter adjustment, stirring frequency and duration iteration.

[0133] Among them, according to the positive and negative signs of the error change values ​​in the error change sequence, the change direction of the current disturbance error is judged to obtain the disturbance direction sequence, which specifically includes:

[0134] First, the perturbation error values ​​of each layer over multiple consecutive execution cycles are extracted and a chronological error sequence is constructed. Then, based on the error values ​​of two adjacent cycles, the error change values ​​are calculated to obtain an error change sequence, which represents the amplitude and directionality of the error fluctuations between adjacent cycles.

[0135] The system then performs a sign judgment operation on each error change value and constructs a disturbance direction sequence based on its sign. If the error change value is greater than zero, it means that the disturbance error value of the current cycle has increased compared to the previous cycle. The system defines this change direction as a positive disturbance and the disturbance direction value in the disturbance direction sequence is +1. If the error change value is greater than zero, it means that the disturbance error has decreased compared to the previous cycle. It is defined as a negative disturbance and the disturbance direction value is -1. If the error change value is equal to zero, that is, the disturbance error value remains unchanged between the two cycles, the disturbance change of the cycle is defined as no direction change, and the disturbance direction value is 0. By processing all periodic error change values ​​in sequence, the system finally forms a complete disturbance direction sequence.

[0136] The calculation formula of the error score value is: ,

[0137] in, For the The error score value of the level, is the index of the level, For the The perturbation error value of the level, is the mean of the perturbation error values ​​of all levels, is the standard deviation of the perturbation error values ​​at all levels, For the The error variation between adjacent execution cycles of the level, , For the The change in the second-order error between adjacent execution cycles of the hierarchy, , For the The direction of change of the error variation between adjacent execution cycles of the hierarchy, For the The direction of the change of the level error change is consistent with the segment quantity, and is the weight coefficient.

[0138] in, is the standardized deviation score item, is the disturbance error growth term, is the disturbance trend growth term, Scoring item for dynamic offset fluctuation; is the direction consistency term, is the normalized evaluation item of disturbance amplitude, Adjust the scoring item for trend magnitude. and is the weight coefficient, and its sum is 1.

[0139] Among them, in the scenario of severe environmental fluctuations and frequent external interference, the immediate response to the disturbance error is crucial. Due to the instability of the external environment (temperature, light, water disturbance), the intensity of the disturbance response has a greater impact on the growth of algae. Therefore, the system needs to quickly identify and correct the current error. In this scenario, more attention should be paid to the fluctuation, intensity and mutation of the current period disturbance error. Setting it to a larger value increases the sensitivity of the scoring mechanism to the immediate error response, thereby improving the real-time control capability. and The values ​​of are 0.8 and 0.2 respectively.

[0140] In the early stages of operation or training, when initial deployment or parameters are not yet stable, the volatility of disturbance data may be strong, and the error of a single cycle is not enough to reflect whether the system is in the correct control direction. Therefore, it is necessary to balance the proportion between short-term deviation and trend, so that the system can avoid over-adjustment while capturing effective trends. Setting it as the median value can maintain the balance of the score and help the system form a preliminary stable operation foundation. and The values ​​of are 0.5 and 0.5 respectively.

[0141] In scenarios where the system tends to be stable and long-term optimization is pursued, the control system has entered a steady state, the disturbance amplitude is small, and the error changes smoothly. The importance of real-time error fluctuations decreases, and the system should pay more attention to whether the control parameters have a stable trend. Especially in long-term batch algae production, maintaining trend consistency helps reduce system adjustment frequency and energy consumption. In this case, the proportion of trend scores should be appropriately increased to allow the system to iteratively evolve in a stable and predictable direction. Therefore and The values ​​of are 0.3 and 0.7 respectively.

[0142] In a preferred embodiment of the present invention, the first control data is iteratively updated according to the disturbance assessment data to generate the second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set, including:

[0143] Obtaining a control update factor set based on the error score values ​​of each level in the disturbance assessment data and the stirring frequency coefficient and duration value of the corresponding level in the first control data;

[0144] According to each element in the control update factor set, parameter iteration processing is performed on the stirring frequency coefficient and the duration value in the first control data to obtain an updated control parameter set;

[0145] According to the updated control parameter set, the target frequency value, the target duration value and the control cycle number are respectively encapsulated at each level to obtain second control data;

[0146] According to the second control data, the control parameters of each level are combined and packaged into an instruction format to generate a control instruction, which is then output to the stirring device. The execution time, level number, and parameter value are recorded to obtain a control data set.

[0147] In an embodiment of the present invention, a control update factor set is obtained based on the error score value of each level in the disturbance assessment data and the stirring frequency coefficient and duration value of the corresponding level in the first control data, so as to make compensatory adjustments to the level with larger error response in a more refined manner, thereby achieving layer-by-layer correction of the disturbance accuracy; according to each element in the control update factor set, the stirring frequency coefficient and duration value in the first control data are respectively subjected to parameter iteration processing to obtain an updated control parameter set, which significantly enhances the system's adaptability and dynamic optimization capabilities, and ensures system stability and equipment operation safety; according to the updated control parameter set, the target frequency value, target duration value and control cycle number are respectively encapsulated according to each level to obtain the second control data, thereby realizing a structured conversion of the control strategy from the optimization process to the actual control parameters, and ensuring that each control adjustment has a cycle number identification, version traceability and maintainability; according to the second control data, the control parameters of each level are combined and packaged into an instruction format, a control instruction is generated, and it is output to the stirring device, and the execution time, level number, and parameter value are recorded to obtain a control data set, effectively preventing system control failure due to communication anomalies, equipment response failure and other problems.

[0148] Among them, according to the error score value of each level in the disturbance assessment data and the stirring frequency coefficient and duration value of the corresponding level in the first control data, a control update factor set is obtained, which specifically includes:

[0149] First, based on the level number, the error score is matched with the stirring frequency coefficient and duration value of the corresponding level in the first control data. The first control data is the control parameter set executed in the previous cycle, which records the stirring frequency and duration used at each level. After the matching is completed, the system constructs the corresponding control update factor based on the error score value of each level. This control update factor represents the adjustment weight of the disturbance deviation on the current control parameter in the form of a function. The calculation process considers the adjustment trend of the stirring frequency and the direction of change of the stirring duration. For example, for a level with a positive error score, it means that the disturbance control at this level is insufficient and the disturbance intensity needs to be increased. Therefore, the current frequency coefficient is proportionally increased and the duration is proportionally decreased. Conversely, the amplitude is reduced or extended to avoid excessive disturbance affecting the stability of the water body. To improve control accuracy, a nonlinear weight function can be introduced into the update factor so that the adjustment amplitude of the high error score level is exponentially increased, thereby accelerating the convergence of the disturbance control strategy. Finally, the error score value of each level is coupled with its corresponding control parameter and the output is a set of control update factors, where each update factor contains three key fields: level number, frequency adjustment coefficient and duration adjustment coefficient.

[0150] According to each element in the control update factor set, parameter iteration processing is performed on the stirring frequency coefficient and the duration value in the first control data to obtain an updated control parameter set, which specifically includes:

[0151] The system reads each element in the control update factor set layer by layer, obtains the stirring frequency coefficient and duration value in the corresponding first control data, and iteratively updates them. The parameter iteration adopts a linear correction model and dynamically adjusts the current control parameters according to the trend control strategy of the disturbance error. The calculation formula of the updated stirring frequency coefficient can be expressed as: ,in is the original frequency coefficient, is the current error score value, is the frequency adjustment gain coefficient, which is used to control the adjustment rate; the update formula of the duration value is: ,in is the original duration value, The duration adjustment coefficient. During the parameter update process, it is also necessary to impose boundary constraints on the frequency and duration values. For example, setting a maximum frequency threshold and a minimum duration threshold to ensure that the updated parameters are within the physical capabilities of the device and avoid the occurrence of unexecutable instructions. The updated frequency coefficient and duration value will be encapsulated together with the current system operation cycle number to form the update control parameter set.

[0152] An embodiment of the present invention further provides an algae efficient cultivation control system, the system comprising:

[0153] The water body segmentation module is used to collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer segmentation data;

[0154] The disturbance tension module is used to calculate the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth based on the layer division data and oxygen concentration data, determine the disturbance tension value of each layer, and obtain the disturbance tension data;

[0155] A control parameter module is used to calculate the disturbance ratio of each layer according to the disturbance tension data and the temperature data, and calculate the stirring frequency coefficient and the duration value in combination with the oxygen concentration recovery time of each layer in the historical period to obtain the first control data;

[0156] A disturbance error module is used to extract the oxygen concentration recovery amplitude in different time periods according to the first control data, calculate the actual disturbance response value, and compare the actual disturbance response value with the disturbance tension data to obtain disturbance error data;

[0157] The error evaluation module is used to analyze the change direction of the disturbance error based on the disturbance error data, perform error calculation and direction superposition, calculate the error score value of the disturbance error at each level, and obtain the disturbance evaluation data;

[0158] The parameter updating module is used to iteratively update the first control data according to the disturbance assessment data, generate the second control data, and output the control instructions generated therefrom to the stirring device to obtain the control data set.

[0159] It should be noted that this system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0160] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the above-described method. All implementations in the above-described method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0161] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, causes the computer to execute the above-described method. All implementations in the above-described method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0162] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for efficiently cultivating and controlling algae, characterized in that: The method comprises: Collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer division data; According to the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated, the disturbance tension value of each layer is determined, and the disturbance tension data is obtained; Calculate the disturbance ratio of each level based on the disturbance tension data and temperature data, and calculate the stirring frequency coefficient and duration value in combination with the oxygen concentration recovery time of each level in the historical period to obtain the first control data; Extracting the oxygen concentration recovery amplitude in different time periods according to the first control data, calculating the actual disturbance response value, and performing a difference comparison between the actual disturbance response value and the disturbance tension data to obtain disturbance error data; According to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed on it. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data; According to the disturbance assessment data, the first control data is iteratively updated to generate the second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set.

2. The method for efficient algae cultivation and control according to claim 1, characterized in that: According to the layer division data and oxygen concentration data, the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth are calculated, and the disturbance tension value of each layer is determined to obtain the disturbance tension data, including: According to the hierarchical division data and oxygen concentration data, the oxygen concentration values ​​of adjacent levels are extracted, and the oxygen concentration differences between adjacent levels are calculated to obtain an oxygen concentration difference sequence; Based on the concentration difference sequence and the equally spaced depths of each layer, the gas diffusion offset value of each layer unit depth is calculated to obtain the diffusion trend data; Based on the oxygen concentration data, the mean and standard deviation of the oxygen concentration differences between each layer are calculated; based on the mean and the oxygen concentration difference between adjacent layers, the weighted density of the concentration difference of the current layer relative to the mean is calculated to obtain the disturbance density term; based on the gas diffusion offset value in the diffusion trend data, the abnormal degree of the current layer disturbance in the fluctuation range is calculated to obtain the concentration deviation fluctuation term; based on the oxygen concentration between adjacent layers, the gas diffusion coupling strength between the two layers is calculated to obtain the disturbance energy concentration term; The disturbance density term, concentration deviation fluctuation term and disturbance energy aggregation term are weighted and fused to calculate the disturbance tension value of each level and obtain the disturbance tension data.

3. The method for efficient algae cultivation and control according to claim 2, characterized in that: According to the disturbance tension data and temperature data, the disturbance ratio of each layer is calculated, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each layer in the historical period to obtain the first control data, including: According to the disturbance tension data, the maximum disturbance tension value and the average disturbance tension value are extracted, and the disturbance tension values ​​of each level are normalized to obtain a disturbance ratio sequence; According to the disturbance ratio sequence and temperature data, the product of the disturbance ratio and the temperature value is calculated to obtain the thermal disturbance score data; Based on the thermal disturbance score data, the difference in thermal disturbance scores between adjacent layers is calculated to obtain the water layer thermal disturbance difference sequence; According to the water layer thermal disturbance difference sequence, the thermal disturbance difference of each layer and its corresponding thermal disturbance score are normalized to obtain the disturbance adjustment sequence; According to the disturbance adjustment sequence, the adjustment level of each level is determined to obtain the disturbance adjustment factor data.

4. The method for efficient algae cultivation and control according to claim 3, characterized in that: According to the disturbance tension data and temperature data, the disturbance ratio of each layer is calculated, and the stirring frequency coefficient and duration value are calculated in combination with the oxygen concentration recovery time of each layer in the historical period to obtain the first control data, which also includes: Obtain the oxygen concentration recovery time of each level in the historical period, and combine it with the disturbance adjustment factor data to calculate the basic frequency value to obtain the basic frequency data; According to the basic frequency data, the minimum and maximum basic frequency values ​​are extracted, a numerical interval is determined, and the numerical interval is equally divided into multiple frequency level intervals; According to the basic frequency data, the frequency level interval of each basic frequency value is determined, and the corresponding frequency level label is assigned to it to obtain the stirring frequency coefficient sequence; According to the stirring frequency coefficient sequence and its corresponding disturbance ratio, the inverse product processing is performed to obtain the stirring time value; Linear mapping is performed according to the stirring duration value to limit its maximum and minimum stirring duration values ​​to obtain a duration value, and first control data is obtained according to the duration value and the stirring frequency coefficient sequence.

5. The method for efficient algae cultivation and control according to claim 4, characterized in that: According to the first control data, the oxygen concentration recovery amplitude in different time periods is extracted, the actual disturbance response value is calculated, and the difference between the actual disturbance response value and the disturbance tension data is compared to obtain the disturbance error data, including: According to the stirring frequency coefficient and duration value of each level in the first control data, the stirring device is controlled to perform stirring operation at the corresponding level, and the execution cycle number is recorded; According to the execution cycle number, oxygen concentration data is collected in a fixed time window before and after the execution cycle, and the difference between the before and after data is calculated to obtain the oxygen concentration recovery amplitude sequence; According to the oxygen concentration recovery amplitude sequence, the oxygen concentration recovery value per unit time at each level is standardized to obtain the actual disturbance response value data; Performing difference calculation on the actual disturbance response value data and the disturbance tension value of each level in the disturbance tension data to obtain the actual disturbance difference sequence; According to the actual disturbance difference sequence, the disturbance error value between the control execution result and the expected disturbance result is determined to obtain the disturbance error data.

6. The method for efficient algae cultivation and control according to claim 5, characterized in that: According to the disturbance error data, the change direction of the disturbance error is analyzed, and the error calculation and direction superposition are performed. The error score value of the disturbance error at each level is calculated to obtain the disturbance evaluation data, including: According to the disturbance error values ​​of each level in the disturbance error data, the disturbance error values ​​in the continuous execution cycles are extracted, and the error change values ​​between adjacent execution cycles are calculated to obtain the error change sequence; According to the positive and negative signs of the error change values ​​in the error change sequence, the change direction of the current disturbance error is judged to obtain the disturbance direction sequence; According to the disturbance direction sequence, the continuous segments with consistent directions in adjacent time periods are counted to obtain the number of segments with consistent directions; Based on the disturbance error value and its mean, the standardized deviation score of the disturbance error is calculated; based on the error increment and error change direction of the disturbance error in adjacent periods in the time series, the disturbance error growth term is calculated; the second-order change ratio of the disturbance error is calculated to obtain the disturbance trend growth term; based on the standardized deviation score term, the disturbance error growth term and the disturbance trend growth term, the dynamic offset fluctuation score term is obtained; According to the change direction of the disturbance error and the amount of direction consistency segments, the direction consistency item is calculated; according to the proportion of the error amplitude relative to the overall disturbance system, the disturbance amplitude normalization evaluation item is calculated; according to the direction consistency item and the disturbance amplitude normalization evaluation item, the trend amplitude adjustment score item is obtained; The dynamic offset fluctuation scoring item and the trend amplitude adjustment scoring item are integrated to calculate the error score value of each level and obtain the disturbance assessment data.

7. The method for efficient algae cultivation and control according to claim 6, characterized in that: According to the disturbance assessment data, the first control data is iteratively updated to generate the second control data, and a control instruction is generated based on the second control data and output to the stirring device to obtain a control data set, including: Obtaining a control update factor set based on the error score values ​​of each level in the disturbance assessment data and the stirring frequency coefficient and duration value of the corresponding level in the first control data; According to each element in the control update factor set, parameter iteration processing is performed on the stirring frequency coefficient and the duration value in the first control data to obtain an updated control parameter set; According to the updated control parameter set, the target frequency value, the target duration value and the control cycle number are respectively encapsulated at each level to obtain second control data; According to the second control data, the control parameters of each level are combined and packaged into an instruction format to generate a control instruction, which is then output to the stirring device. The execution time, level number, and parameter value are recorded to obtain a control data set.

8. An efficient algae cultivation control system, characterized in that: The system is used to perform the method according to any one of claims 1 to 7, and the system comprises: The water body segmentation module is used to collect water level data, oxygen concentration data, and temperature data, and perform vertical segmentation processing based on the water level data to divide the water body into multiple equally spaced depth levels to obtain layer segmentation data; The disturbance tension module is used to calculate the oxygen concentration difference between adjacent layers and the gas diffusion trend value per unit depth based on the layer division data and oxygen concentration data, determine the disturbance tension value of each layer, and obtain the disturbance tension data; A control parameter module is used to calculate the disturbance ratio of each layer according to the disturbance tension data and the temperature data, and calculate the stirring frequency coefficient and the duration value in combination with the oxygen concentration recovery time of each layer in the historical period to obtain the first control data; A disturbance error module is used to extract the oxygen concentration recovery amplitude in different time periods according to the first control data, calculate the actual disturbance response value, and compare the difference between the actual disturbance response value and the disturbance tension data to obtain disturbance error data; The error evaluation module is used to analyze the change direction of the disturbance error based on the disturbance error data, perform error calculation and direction superposition, calculate the error score value of the disturbance error at each level, and obtain the disturbance evaluation data; The parameter updating module is used to iteratively update the first control data according to the disturbance assessment data, generate the second control data, and output the control instructions generated therefrom to the stirring device to obtain the control data set.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images