Leachate treatment method based on microalgae coupling nitrification process

Abstract

The invention discloses a percolate treatment method based on a microalgae coupling nitration process, and belongs to the field of wastewater biological treatment and intelligent control, and the method comprises the following steps: constructing initial reaction conditions suitable for microalgae proliferation and nitration reaction; operating parameters such as ammonia nitrogen concentration, nitrite concentration, dissolved oxygen, temperature and microalgae concentration are collected to form real-time operating data; based on the weighted multi-parameter fitting relationship, calculating an optimal combination of the illumination intensity, the carbon source feeding rate, the cycle ratio and the reaction duration; when the ammonia nitrogen removal rate is continuously lower than a threshold value, calling historical data to carry out trend analysis, and dynamically correcting a fitting weight; applying the correction result to control the illumination condition and the water inlet and outlet proportion; after the operation cycle is finished, updating the fitting model by adopting an incremental learning mode to realize continuous optimization of the control strategy; according to the invention, the nitrogen element removal efficiency of the percolation liquid and the system stability are improved, and the self-adaptive adjustment capability is relatively strong.

Description

Technical Field

[0001] The present invention relates to the field of wastewater biological treatment and intelligent control, and in particular to a leachate treatment method based on a microalgae-coupled nitrification process. Background Art

[0002] Leachate, a highly concentrated organic wastewater generated during the landfill or storage of municipal solid waste, is a typical difficult-to-treat wastewater due to its high ammonia nitrogen concentration, high organic load, complex water composition, and high toxicity. If discharged directly without effective treatment, it can easily cause environmental problems such as eutrophication and groundwater contamination. Therefore, the efficient removal of nitrogen pollutants from leachate has become a key research topic in solid waste management.

[0003] At present, the treatment of ammonia nitrogen in leachate mainly adopts traditional biological denitrification processes such as nitrification-denitrification. However, such processes usually rely on external carbon sources and have high requirements for operating conditions. In addition, under leachate characteristics such as low C / N ratio and high salinity, the denitrification efficiency decreases significantly, the system stability is poor, and the operating cost is high.

[0004] Microalgae are autotrophic organisms that, under sunlight, can utilize carbon dioxide for photosynthesis while absorbing nutrients such as ammonia nitrogen and phosphorus. Their oxygen production provides dissolved oxygen for nitrification, giving them excellent denitrification potential. However, using microalgae alone to treat high-ammonia nitrogen leachate can result in growth inhibition and incomplete nitrogen removal.

[0005] To address the above issues, some studies have attempted to couple microalgae technology with the autotrophic nitrification process. However, the current coupling path is imperfect, and there is a lack of precise regulation of the synergistic relationship between microalgae growth and the nitrification process. Especially under conditions such as high ammonia nitrogen, low carbon, and high toxicity in the leachate, it is still difficult to achieve stable and efficient denitrification. Summary of the Invention

[0006] The purpose of the present invention is to provide a leachate treatment method based on a microalgae-coupled nitrification process to address the shortcomings of the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a leachate treatment method based on a microalgae-coupled nitrification process, comprising: The leachate is introduced into a reaction environment with controllable light conditions to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction; During the reaction process, the operating parameters of ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, pH value, temperature and microalgae concentration are obtained to form real-time operating data; Based on the real-time operation data obtained, the optimal combination of light intensity, carbon source injection rate, circulation ratio and reaction time is calculated through the established weighted multi-parameter fitting relationship; When the monitoring results show that the ammonia nitrogen removal rate is lower than the preset threshold for two consecutive cycles, the historical operation data is called for trend analysis, and the weight parameters in the weighted multi-parameter fitting relationship are dynamically modified based on the time series prediction method to update the operation control strategy; Control lighting conditions and water inlet and outlet ratios based on the revised weight parameters; After each operation cycle, the treated leachate is output, and all the operation data of this cycle are used to update the weighted multi-parameter fitting relationship.

[0008] Preferably, the step of introducing the leachate into a reaction environment with controllable light conditions to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction comprises: The leachate was pretreated through a coarse grid and then introduced into the initial reaction unit, and the pH value was adjusted to 6.8-7.5 using a buffer solution; Based on the influent ammonia nitrogen concentration and total organic carbon value, the required carbon-nitrogen ratio is calculated and the amount of external carbon source added is determined; After the carbon source is added, a pre-illumination period is set to gradually increase the light intensity in the reaction environment to the target illumination.

[0009] Preferably, the optimal combination of light intensity, carbon source addition rate, circulation ratio and reaction time calculated by establishing a weighted multi-parameter fitting relationship includes: The obtained ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, microalgae concentration and temperature were normalized; Each normalized parameter is assigned a value according to the preset weight, and a weighted multi-parameter fitting function is constructed. The fitting function takes the target control variable as the output, and the optimal combination scheme is selected using the minimum deviation principle.

[0010] Preferably, the construction of a weighted multi-parameter fitting function, wherein the fitting function takes the target control variable as output and adopts the minimum deviation principle to select the optimal combination scheme includes: Each weight parameter is substituted into the multivariate nonlinear fitting model, and the fitting model is trained using a stepwise regression algorithm to output the predicted values ​​of light intensity, carbon source injection rate, circulation ratio and reaction time in the control variable group; For each set of predicted values, the deviation between the objective function value in the previous cycle and the current predicted objective function value is calculated. The objective function includes pollutant removal rate and microalgae growth rate. The set of control variables with the smallest deviation value is selected as the optimal output solution.

[0011] Preferably, when the monitoring results show that the ammonia nitrogen removal rate is lower than a preset threshold value for two consecutive cycles, calling historical operation data for trend analysis includes: Record the actual measured values ​​of ammonia nitrogen removal rate in two consecutive operation cycles and determine whether they are both lower than the set threshold. If so, call historical operation data with similar environmental conditions to the reaction stage to form a time series data set; The trend analysis of the time series data set was performed, and the short-term trend change value of the ammonia nitrogen removal rate was calculated using the exponential smoothing prediction method. The deviation between the predicted trend and the current measured value was used as the trigger factor for adjusting the fitting weight. The weights of the operating parameters in the current weighted multi-parameter fitting relationship are dynamically modified to update the calculation path of the control variables in the next cycle.

[0012] Preferably, dynamically modifying the weight of each operating parameter in the current weighted multi-parameter fitting relationship includes: The Pearson correlation coefficient between each operating parameter and the ammonia nitrogen removal rate in the ten historical cycles was calculated, and a correlation matrix between the parameters and the target performance indicators was constructed. The parameters were then divided into three groups: high correlation, medium correlation, and low correlation. The fluctuation range of each operating parameter in the current cycle is combined with the correlation category to which it belongs, and the weight adjustment function is applied to perform value correction. The weight increase of the parameters with large fluctuation range in the high correlation group is prioritized; the weight of the parameters in the low correlation group with small fluctuation range in the current cycle is linearly reduced. The modified weight set is used to reconstruct the multi-parameter fitting relationship to update the next cycle control output of light intensity, carbon source injection rate, circulation ratio and reaction time.

[0013] Preferably, the control of lighting conditions and inlet and outlet water ratios according to the corrected weight parameters includes: determining the target lighting intensity and leachate inlet and outlet water ratio based on the output result of the weighted fitting function after weight correction; adjusting the light source power supply time and the illumination intermittent period to match the target lighting intensity, and at the same time setting the water inlet and outlet rates by adjusting the operating time and start and stop frequency of the peristaltic pump.

[0014] Preferably, using all the operating data of the current cycle to update the weighted multi-parameter fitting relationship includes: updating the weighted multi-parameter fitting relationship by adopting an incremental learning method, and inputting the data of the new cycle as an extended sample.

[0015] Preferably, the normalized operating parameters collected in the current operating cycle and their corresponding target output values ​​are combined into a training sample pair, and a sliding sample window is constructed together with the historical samples retained in the most recent cycles; the Euclidean distance between the new sample and the historical sample in the parameter distribution space is compared, and samples with high differences are screened out and preferentially added to the weighted multi-parameter fitting relationship.

[0016] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The proposed leachate treatment method, based on a weighted multi-parameter fitting relationship, effectively achieves precise calculation and optimized control of key control variables such as light intensity, carbon source injection rate, circulation ratio, and reaction time by introducing multi-source operating parameter normalization, a dynamic weight adjustment mechanism, and an incremental learning strategy. This method not only improves the synergistic efficiency of microalgae and nitrification, but also enhances the system's adaptability to leachate quality fluctuations, significantly improving ammonia nitrogen removal efficiency and microalgae biomass stability.

[0017] 2. Compared with existing processes that use fixed control parameters or static models, this invention offers significant advantages in data-driven intelligence: through time series trend prediction, dynamic weight correction, and sliding sample window management, it establishes a sustainable, adaptively optimized control path, improving system operational stability. Furthermore, this invention boasts excellent algorithm scalability and can be widely applied to the resource-based and ecological management of high-ammonia-nitrogen, difficult-to-treat wastewater. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] For examples, see Figure 1 As shown, the leachate treatment method based on the microalgae-coupled nitrification process described in this embodiment includes: The leachate is introduced into a reaction environment with controllable light conditions to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction; During the reaction process, the operating parameters of ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, pH value, temperature and microalgae concentration are obtained to form real-time operating data; Based on the real-time operation data obtained, the optimal combination of light intensity, carbon source injection rate, circulation ratio and reaction time is calculated through the established weighted multi-parameter fitting relationship; When the monitoring results show that the ammonia nitrogen removal rate is lower than the preset threshold for two consecutive cycles, the historical operation data is called for trend analysis, and the weight parameters in the weighted multi-parameter fitting relationship are dynamically modified based on the time series prediction method to update the operation control strategy; According to the modified weight parameters, the lighting conditions and the ratio of inlet and outlet water are controlled to achieve a dynamic balance between the oxygen production rate of microalgae and the oxygen demand of nitrification reaction. After each operation cycle, the treated leachate is output, and all the operation data of this cycle are used to update the weighted multi-parameter fitting relationship.

[0022] In the present invention, the process of “introducing the leachate into a reaction environment with controllable light conditions, adjusting the pH value to a constant range, and supplementing an exogenous carbon source according to a preset carbon-nitrogen ratio to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction” specifically includes the following technical measures: First, leachate collected from landfills or other sources is screened through a mechanical screen to remove suspended particles, floating debris, and coarse impurities, preventing clogging or light interference in subsequent reaction units. This step not only ensures a stable reaction environment but also provides a cleaner growth medium for the microalgae.

[0023] After the grid pretreatment is completed, the leachate is introduced into a reaction unit with a controllable lighting device. Considering that the ammonia nitrogen concentration in the leachate is usually high (often hundreds to thousands of milligrams per liter), which has a certain inhibitory effect on the growth of microalgae, the pH value needs to be adjusted before the reaction starts to slow down the volatilization of ammonia nitrogen and provide a suitable enzyme activity environment for nitrifying bacteria. In this embodiment, a buffer solution formed by a compound of sodium bicarbonate and phosphate (such as potassium dihydrogen phosphate) is used to regulate the pH, wherein sodium bicarbonate is mainly used to provide alkalinity, and phosphate provides a phosphorus source while enhancing the pH buffering capacity. It is preferred to control the pH to be stable between 6.8 and 7.5. The adjustment method adopts drop-by-drop addition supplemented by a pH online monitoring probe to ensure that the set range is reached within the specified time.

[0024] Secondly, to ensure the smooth progress of the nitrification reaction, sufficient carbon source must be available during the ammonia oxidation phase to meet the initial growth needs of the microalgae and prevent premature initiation of the denitrification reaction. Therefore, in the present invention, the addition of an exogenous carbon source is dynamically calculated based on the carbon-nitrogen ratio (C / N), which is the ratio of the total organic carbon content of the carbon source to the ammonia nitrogen concentration. Specifically, a ratio analysis is performed based on the initial ammonia nitrogen concentration (in milligrams per liter) of the influent and the real-time monitored TOC (total organic carbon) value. When the C / N ratio falls below a set threshold (preferably 6), the carbon source is automatically added. In this invention, the added carbon source is preferably a biodegradable carbon source such as sodium acetate, glycerol, or ethanol. Prior to addition, it is diluted at a ratio of 1:10 using an online dilution system to prevent localized high concentrations that could cause microalgae cell membrane rupture or premature initiation of the denitrification chain reaction. The diluted carbon source is added using a precision peristaltic pump to ensure uniform addition and a controllable rate.

[0025] To further improve the adaptability of microalgae to the reaction environment, especially their ability to resist stress under initial conditions of high ammonia nitrogen, the present invention provides a pre-activation stage. Specifically, after the leachate enters the reaction unit and the pH and carbon source are regulated, a pre-illumination period is set, and the light intensity is gradually increased to the target value (preferably in the range of 30 to 80 micromoles of photons per square meter per second), and a pre-adaptation period of not less than 30 minutes is maintained. During this stage, the microalgae have not yet proliferated in large quantities but have begun to initiate photosynthesis. The oxygen they release can be naturally enriched in the reaction unit, providing an initial oxygen source support for the subsequent nitrification process. At the same time, the slowly increasing light intensity helps activate the photosynthetic pigment system, improve the physiological response of microalgae in a high ammonia nitrogen environment, and reduce inhibition.

[0026] In the present invention, after the leachate enters the reaction stage, multiple sets of online sensors are deployed at different key locations to collect operating parameters in real time. The above sensors include: Ammonia nitrogen and nitrite sensors: These use selective ion electrodes to respond to the concentrations of ammonia and nitrite ions in the reaction solution, enabling online continuous detection. This type of electrode has high selectivity and is immune to interference from complex background ions in the leachate. Dissolved oxygen, pH value and temperature sensors: Fluorescence dissolved oxygen probe, composite pH glass electrode and thermal resistance thermometer are used. All probes are encapsulated in a highly corrosion-resistant structure to adapt to the high salinity and high ammonia nitrogen environment of the leachate; Microalgae concentration collection device: Set up an optical density detection window and use colorimetric detection to obtain the optical density value (OD value) in the reaction solution as an indirect indicator of microalgae concentration.

[0027] To ensure data temporal consistency and responsiveness, all sensors synchronize data acquisition with a ten-minute baseline. Unlike traditional timed data collection methods, this invention further incorporates a dynamic sampling frequency adjustment mechanism. When the system detects that the fluctuation rate of any key parameter (such as ammonia nitrogen concentration or dissolved oxygen) exceeds a set threshold, the sampling frequency is automatically increased to once every five minutes. If the parameter stabilizes, the frequency is restored to the normal rate. This strategy balances system resource consumption with response accuracy, providing enhanced real-time adaptability.

[0028] When it comes to collecting microalgae concentrations, conventional optical density methods are susceptible to bubble interference, leading to erratic readings, due to the large amounts of suspended matter and biofoam in leachate. To address this, the present invention incorporates a sliding average and outlier rejection algorithm into the optical density reading process. This algorithm takes OD values ​​from three consecutive sampling cycles, performs a weighted average, and removes outlier data points that deviate from the mean by more than two standard deviations. This significantly improves the stability and reliability of microalgae concentration data in high-foam environments.

[0029] All collected data is organized in a unified structured data format. Each data set consists of three core elements: the timestamp (collection time), the measured value (e.g., an ammonia nitrogen concentration of 30 mg / L), and the measurement confidence level (determined by probe calibration accuracy and algorithm bias, and categorized as high, medium, and low). This ternary data structure not only facilitates subsequent modeling and control input but also supports data quality tracking, preventing deviations from overall control logic due to error propagation.

[0030] After data collection is complete, the real-time data verification and calibration phase begins. The present invention incorporates a multi-parameter linkage verification mechanism for this phase: This mechanism is triggered when two or more operating parameters are detected to deviate from their respective preset ranges simultaneously (for example, an increase in ammonia nitrogen concentration and a decrease in dissolved oxygen). At this point, the current abnormal data is compared with the data from the same point in time during the previous valid operating cycle. If the deviation exceeds 15%, the data is marked as "requires verification" and temporarily not used for control parameter input. An alarm is also issued. This mechanism effectively reduces control errors caused by single-point sensor failures or temporary interference, improving the system's stable operation capabilities.

[0031] Furthermore, to ensure long-term sensor stability under high ammonia nitrogen concentrations and complex backgrounds during data acquisition, the present invention employs a timed self-cleaning mechanism and intermittent calibration to ensure the accuracy of each sensor over extended periods of operation. The probe cleaning process is automatically triggered every 48 hours of operation, and calibration is performed manually or automatically every seven days using a standard solution or a simulant of known concentration.

[0032] In the present invention, the whole process of "calculating the optimal combination of control variables by establishing a weighted multi-parameter fitting relationship" includes the following technical details: After obtaining operational parameters such as ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, microalgae concentration, and temperature, each parameter must first be normalized to eliminate the effects of different physical dimensions and numerical ranges. Specifically, the present invention preferably employs a linear interval scaling method for normalization. This involves subtracting the historical minimum value of each parameter from its current value, and then dividing the result by the historical range (maximum minus minimum) of the parameter, thereby normalizing the result to a dimensionless standard value between 0 and 1.

[0033] For example, if the ammonia nitrogen concentration at a certain point in time is 300 mg / L, its historical minimum value is 100, and its maximum value is 500, then the normalized result is: .

[0034] All operating parameters involved in the calculation are standardized according to the above method to ensure that in the subsequent weighting and fitting process, each parameter has an equivalent mathematical expression basis, avoiding a certain parameter having a disproportionate impact on the output result due to its large value.

[0035] After normalization, the system assigns weights to each parameter. Rather than using manual experience to set weights, this method dynamically calculates weighting factors based on historical correlations between parameters and key performance indicators. In practice, the system uses statistical correlation coefficients for each operating parameter over several consecutive cycles with pollutant removal rates, microalgae concentration growth rates, and reaction solution pH stability as initial weight values.

[0036] After the weights are assigned, the parameter combination at the input of the fitting function is realized by constructing a weighted linear combination function or a nonlinear combination function. For example: If the normalized parameters are A1 (ammonia nitrogen), A2 (nitrite), A3 (DO), A4 (microalgae concentration), and A5 (temperature), and their corresponding weights are W1 to W5, then the input expression of the fitting function is: W1×A1 + W2×A2 + W3×A3+ W4×A4 + W5×A5; To further improve the adaptability and generalization of the function, this paper uses a stepwise regression algorithm to construct a multivariate nonlinear fitting model. Stepwise regression is a variable selection process based on significance testing that effectively eliminates redundant variables and improves the explanatory power of the model. The regression process includes variable introduction, variable elimination, and interaction term setting. The fitting objective is to predict the optimal range of control variables.

[0037] For example, the system can construct four nonlinear regression sub-models corresponding to light intensity, carbon source injection rate, circulation ratio, and reaction time. Each model takes a weighted combination of operating parameters as input and the historical optimal value of the control variable as output, ultimately obtaining four sets of predicted values.

[0038] The light intensity, carbon source injection rate, cycle ratio, and reaction time calculated by the fitting model are the candidate value set. Since the model has certain errors, in order to ensure the reliability of the actual operation effect, the present invention further designs a minimum deviation selection mechanism.

[0039] Specifically, the system retrieves the output value of the objective function of the previous operation cycle. The objective function includes but is not limited to: Ammonia nitrogen removal rate; microalgae biomass growth rate; The pH change of the reaction solution.

[0040] Then, each fitted prediction value is fed into the currently running control logic to predict the possible output value of the objective function. The predicted value is then compared with the actual value of the previous cycle to calculate the deviation. This deviation is preferably calculated using the "relative difference absolute value" method, that is, the predicted target value is subtracted from the actual target value, and the absolute value is taken after dividing it by the target value of the previous cycle.

[0041] After simulating and evaluating the deviations of each of the four control variable combinations, the set of control variables with the smallest deviation is selected as the final output solution for the next cycle of operational control. This approach not only ensures that the actual predicted values ​​meet system expectations but also enhances the control strategy's resilience to outliers and sudden operating conditions.

[0042] The final selected control parameters act on the reaction light source device, carbon source dosing controller, liquid reflux regulating valve and periodic drainage timer respectively to ensure that the reaction process operates under the new optimal control conditions.

[0043] During actual operation, the system records the ammonia nitrogen removal rate for each cycle and compares it with the set performance threshold. If the ammonia nitrogen removal rate is detected to be below the preset threshold (for example, below 85%) for two consecutive cycles (for example, two consecutive hours or two control cycles), the system will determine that the operating performance is abnormal and trigger the parameter weight adjustment.

[0044] To ensure the timeliness and relevance of weight adjustments, the system requires access to historical data with operating conditions similar to the current reaction phase (such as pH, temperature, and microalgae concentration range). A time series dataset covering the last ten valid cycles is preferably constructed. Each cycle includes real-time values ​​of the target output variable (ammonia nitrogen removal rate) and the operating parameters involved in the fitting (ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, microalgae concentration, temperature, etc.).

[0045] For this time series dataset, we used an exponential smoothing forecasting method to calculate short-term trends in ammonia nitrogen removal rates. Compared to simple averaging or linear regression, exponential smoothing is more accurate in capturing subtle trends in time series and is therefore suitable for short-term operating condition assessments.

[0046] The exponential smoothing forecast value is calculated as follows: The predicted value at the current moment = smoothing coefficient × current actual value + (1-smoothing coefficient) × predicted value at the previous moment.

[0047] The smoothing coefficient is a weighting factor between 0 and 1 (preferably 0.3 to 0.5 in the present invention), and can be dynamically adjusted according to the degree of system fluctuation.

[0048] The trend value of ammonia nitrogen removal rate obtained by this method will be compared with the current measured value. The difference between the two is the deviation factor, which is used to determine whether the actual performance of the system deviates seriously from the trend expectation.

[0049] When the absolute value of the deviation factor is greater than a set tolerance threshold (such as ±5%), the system triggers the weighted correction procedure and proceeds to the next step.

[0050] In order to improve the adaptability of the weighted multi-parameter fitting relationship, the present invention establishes a correlation matrix between operating parameters and ammonia nitrogen removal rate based on historical data to support the weight correction logic.

[0051] The specific operations are: For the historical data of the above ten cycles, calculate the Pearson correlation coefficient between each operating parameter and ammonia nitrogen removal rate This coefficient is used to measure the degree of linear correlation between two variables and ranges from -1 to +1. The closer the coefficient is to +1, the stronger the positive effect of the parameter on ammonia nitrogen removal efficiency.

[0052] According to the Pearson coefficient, all operating parameters involved in fitting are divided into three categories: high correlation group (|r| ≥ 0.7); medium correlation group (0.4 ≤ |r|<0.7); Low correlation group (|r|<0.4).

[0053] Analyze the fluctuation range of each operating parameter in the current cycle. The fluctuation range can be expressed as the ratio of the absolute value of the difference between the current parameter value and the previous cycle value to the mean value, which is used to reflect the degree of dynamic change of the parameter in the current cycle.

[0054] Then, based on the parameter's correlation category and its current volatility, the weight adjustment function is called to perform value correction. This function follows the following logic: If a parameter belongs to a highly correlated group and has a large fluctuation range in the current period (exceeding its historical standard deviation), its original weight value will be positively increased (for example, by 10% to 20%). If the parameter belongs to the low correlation group and has a small fluctuation range (within ±10% of the historical mean), its weight value will be linearly reduced (e.g., reduced by 5% to 10%). The parameters of the relevant groups are adjusted mildly according to the specific fluctuation range, with small increases or decreases or remaining unchanged.

[0055] All corrected weight values ​​will be standardized (for example, the sum is normalized to 1) and used in the weighted fitting function in the calculation of the control variables (light intensity, carbon source injection rate, cycle ratio, reaction time) of the next cycle.

[0056] After adopting the new set of weights, the system will rebuild the weighted fitting function to update the output values ​​of the control variables. This process can continue to use the aforementioned nonlinear regression model structure, without changing the original model architecture, and only replacing the parameter weights.

[0057] At the same time, the revised weights and their effective period will be marked in the data records and participate in subsequent model evaluation as part of the "sample labels" in the machine learning path for model iterative optimization.

[0058] Through the above-mentioned dynamic weight correction mechanism, the present invention can timely adjust the fitting strategy in a data-driven manner when abnormal fluctuations occur in system operation or the removal effect of target pollutants decreases, and has obvious adaptive control capabilities and fault response mechanisms.

[0059] After completing the fitting model weight correction (for example, through Pearson correlation coefficient grouping and amplitude-driven function correction), the system reapplies the new set of weights to the weighted fitting relationship. The fitting calculation results in an updated set of control variables, including light intensity (in micromoles of photons per square meter per second) and the ratio of water inlet and outlet (in units of a dimensionless ratio).

[0060] The light intensity output is limited to the effective physiological illumination range of 30 to 80 micromoles photons per square meter per second, which represents the optimal illumination range determined by this invention based on microalgae experiments. The inlet and outlet water ratio is set as the ratio of "inlet flow rate:outlet flow rate," preferably between 1:1.2 and 1:1.8, to ensure a reasonable residence time of the reaction solution within the reaction tank and prevent loss of microalgae.

[0061] Lighting control includes two dimensions: light intensity and light mode.

[0062] Light intensity control: Utilizing an adjustable-voltage LED light source, the light controller receives target illumination values, enabling continuous regulation of the light intensity. The controller is integrated into a closed-loop feedback system, with a light sensor monitoring illumination values ​​in real time, ensuring that the actual illumination remains within ±5% of the model output.

[0063] Lighting pattern control: Generate a lighting curve based on the model output value, combine the diurnal cycle with the response requirements, and set an intermittent lighting cycle (e.g., 20 minutes of operation and 5 minutes of rest) to simulate the natural circadian rhythm and reduce the risk of photoinhibition of microalgae.

[0064] Through the above method, the present invention not only achieves precise control of light intensity, but also introduces the "light rhythm" dimension, thereby enhancing the adaptability of microalgae to artificial lighting environment and photosynthesis efficiency.

[0065] The adjustment of the water inlet and outlet ratio relies on two sets of independently operated peristaltic pumps or solenoid valve flow control devices: The water inlet pump receives the target water inlet rate value output by the model before starting, and uses the flow sensor as the feedback reference during operation to achieve closed-loop flow control.

[0066] The outlet pump or drainage device automatically calculates the target water outlet rate based on the water inflow and the set ratio, and realizes discharge control by timing opening and closing or variable frequency operation.

[0067] For example, if the current cycle model outputs an inlet-outlet ratio of 1:1.5, the system is set to discharge 150 liters of outlet water for every 100 liters of water in. To prevent excessive fluctuations in the reaction liquid level during the drainage process, the drainage operation can be set to a pulsed mode, that is, discharging the water in five intermittent steps, each interval set between 2 and 3 minutes.

[0068] The water inlet and outlet operations are also monitored by liquid level probes. If the liquid level drops beyond the safety threshold, the water outflow will be temporarily suspended and a reminder will be issued.

[0069] The present invention specifically designs an operation feedback and parameter memory mechanism to determine whether the current control strategy has achieved the optimization goal and to store the effective strategy for subsequent use.

[0070] The evaluation of control results is based on the following two indicators: Microalgae concentration growth rate: Compare the change in microalgae concentration between the current cycle and the previous cycle (e.g., by optical density (OD) or cell number density); Ammonia nitrogen concentration decrease rate: Compare the difference between the inlet and outlet ammonia nitrogen concentrations and divide it by the treatment time to form a denitrification efficiency index per unit time.

[0071] When both of the above indicators are improved compared with the previous cycle and the system operates stably (such as pH fluctuation is less than ±0.3 and dissolved oxygen is maintained within the set range), the system will mark the current control variable combination (including light intensity and inlet and outlet water ratio) as "optimization effective" and write it into the control strategy library.

[0072] After each operation cycle is completed, the system automatically pre-processes the collected data during the entire cycle. This mainly includes the following: Normalization of operating parameters: including ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, microalgae concentration, temperature, pH value, etc., are all scaled according to the historical maximum and minimum values ​​so that each parameter is mapped to the [0, 1] interval; Target output value extraction: including the final ammonia nitrogen removal rate, microalgae biomass growth rate or other control target values ​​of the cycle.

[0073] Subsequently, the system constructs training sample pairs in the form of "input parameter vector + corresponding target output value", namely: Input: Normalized parameter vector of the current cycle (denoted as ) Output: The corresponding actual target value (denoted as ) Forming sample pairs ( , ), as the extended samples that can be used for training in this cycle.

[0074] To prevent the model from increasing computational complexity due to the continuous accumulation of training samples, this paper uses a sliding sample window mechanism to control the sample size. The sample window is set to a maximum size (for example, the latest 30 cycles). When new samples are input, the oldest set of samples is automatically removed, ensuring that the training data is always up to date and within a limited time series range, improving the timeliness and convergence efficiency of the model.

[0075] In order to avoid interference of sample redundancy or repetitive data on the model, the present invention introduces the Euclidean distance calculation method in the parameter distribution space to evaluate the degree of difference between the current new sample and the historical sample, and accordingly selects new data with greater representativeness for model updating.

[0076] The specific implementation steps are as follows: For new samples ( ) and each historical sample in the sample window ( ) calculates the Euclidean distance, which is defined as: ; where n is the number of operating parameters, Calculate the minimum distance between the new sample and the historical sample set , and compare it with the set threshold, for example, if (Indicating that the new sample is significantly different from the historical sample in the parameter space), the sample is judged to be a "high difference" sample.

[0077] Only new samples that are judged to be highly different are included in the training data to avoid repeatedly using samples that are too close for model updates and reduce the risk of overfitting in model training.

[0078] Accepted new samples are used to incrementally update the existing weighted multi-parameter fitting relationship. Unlike retraining, incremental updates do not reconstruct the entire model structure. Instead, they preserve the existing fitting relationship and achieve parameter correction and model performance enhancement through local adjustments.

[0079] The essence of the fitting relationship is a mapping function from an input vector to an output variable. The function structure is generated by historical data training and is expressed in the form of multivariate weighted regression: ;in, to is the normalized input parameter, to is the fitting weight, and ε is the residual term.

[0080] Incremental updates use the following mechanism: With the new sample ( , ) is input and its prediction error under the existing fitting function is calculated, that is: error = |predicted value - actual value|; like If the threshold is higher than the set threshold (e.g. 5%), the The weights corresponding to the key parameters , fine-tuning is done by the least squares criterion or gradient update method.

[0081] like If it is lower than the set threshold, only the residual term is updated, keeping the main weight structure unchanged and reducing unnecessary model disturbances.

[0082] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A leachate treatment method based on microalgae coupled nitrification process, characterized by: include: The leachate is introduced into a reaction environment with controllable light conditions to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction; During the reaction process, the operating parameters of ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, pH value, temperature and microalgae concentration are obtained to form real-time operating data. Based on the real-time operating data obtained, the optimal combination of light intensity, carbon source addition rate, circulation ratio and reaction time is calculated through the established weighted multi-parameter fitting relationship. When the monitoring results show that the ammonia nitrogen removal rate is lower than the preset threshold for two consecutive cycles, the historical operating data is called for trend analysis, and the weight parameters in the weighted multi-parameter fitting relationship are dynamically modified based on the time series prediction method to update the operation control strategy. The lighting conditions and the ratio of inlet and outlet water are controlled according to the modified weight parameters; after each operation cycle, the treated leachate is output, and all the operation data of this cycle are used to update the weighted multi-parameter fitting relationship.

2. The leachate treatment method based on microalgae coupled nitrification process according to claim 1, characterized in that: The leachate is introduced into a reaction environment with controllable light conditions to establish initial conditions suitable for microalgae proliferation and ammonia oxidation reaction, including: The leachate was pretreated through a coarse grid and then introduced into the initial reaction unit, and the pH value was adjusted to 6.8-7.5 using a buffer solution; Based on the influent ammonia nitrogen concentration and total organic carbon value, the required carbon-nitrogen ratio is calculated and the amount of external carbon source added is determined; After the carbon source is added, a pre-illumination period is set to gradually increase the light intensity in the reaction environment to the target illumination.

3. The leachate treatment method based on microalgae coupled nitrification process according to claim 1, characterized in that: The optimal combination of light intensity, carbon source addition rate, circulation ratio and reaction time calculated by establishing a weighted multi-parameter fitting relationship includes: The obtained ammonia nitrogen concentration, nitrite concentration, dissolved oxygen concentration, microalgae concentration and temperature were normalized; Each normalized parameter is assigned a value according to the preset weight, and a weighted multi-parameter fitting function is constructed. The fitting function takes the target control variable as the output, and the optimal combination scheme is selected using the minimum deviation principle.

4. The leachate treatment method based on microalgae coupled nitrification process according to claim 3, characterized in that: The method of constructing a weighted multi-parameter fitting function, wherein the fitting function takes the target control variable as output and adopts the minimum deviation principle to select the optimal combination scheme includes: Each weight parameter is substituted into the multivariate nonlinear fitting model, and the fitting model is trained using a stepwise regression algorithm to output the predicted values ​​of light intensity, carbon source injection rate, circulation ratio and reaction time in the control variable group; For each set of predicted values, the deviation between the objective function value in the previous cycle and the current predicted objective function value is calculated. The objective function includes pollutant removal rate and microalgae growth rate. The set of control variables with the smallest deviation value is selected as the optimal output solution.

5. The leachate treatment method based on microalgae coupled nitrification process according to claim 1, characterized in that: When the monitoring results show that the ammonia nitrogen removal rate is lower than the preset threshold value for two consecutive cycles, calling the historical operation data for trend analysis includes: Record the actual measured values ​​of ammonia nitrogen removal rate in two consecutive operation cycles and determine whether they are both lower than the set threshold. If so, call historical operation data with similar environmental conditions to the reaction stage to form a time series data set; The trend analysis of the time series data set was performed, and the short-term trend change value of the ammonia nitrogen removal rate was calculated using the exponential smoothing prediction method. The deviation between the predicted trend and the current measured value was used as the trigger factor for adjusting the fitting weight. The weights of the operating parameters in the current weighted multi-parameter fitting relationship are dynamically modified to update the calculation path of the control variables in the next cycle.

6. The leachate treatment method based on microalgae coupled nitrification process according to claim 5, characterized in that: Dynamically modifying the weights of each operating parameter in the current weighted multi-parameter fitting relationship includes: The Pearson correlation coefficient between each operating parameter and the ammonia nitrogen removal rate in the ten historical cycles was calculated, and a correlation matrix between the parameters and the target performance indicators was constructed. The parameters were then divided into three groups: high correlation, medium correlation, and low correlation. The fluctuation range of each operating parameter in the current cycle is combined with the correlation category to which it belongs, and the weight adjustment function is applied to perform value correction. The weight increase of the parameters with large fluctuation range in the high correlation group is prioritized; the weight of the parameters in the low correlation group with small fluctuation range in the current cycle is linearly reduced. The modified weight set is used to reconstruct the multi-parameter fitting relationship to update the next cycle control output of light intensity, carbon source injection rate, circulation ratio and reaction time.

7. The leachate treatment method based on microalgae coupled nitrification process according to claim 6, characterized in that: The control of the lighting conditions and the inlet and outlet water ratio according to the corrected weight parameters includes: determining the target lighting intensity and the inlet and outlet water ratio of the leachate based on the output result of the weighted fitting function after weight correction; adjusting the light source power supply time and the illumination intermittent period to match the target lighting intensity, and at the same time setting the water inlet and outlet rates by adjusting the operating time and start and stop frequency of the peristaltic pump.

8. The leachate treatment method based on microalgae coupled nitrification process according to claim 1, characterized in that: The using of all the operation data of the current cycle to update the weighted multi-parameter fitting relationship includes: updating the weighted multi-parameter fitting relationship by an incremental learning method, and inputting the data of the new cycle as an extended sample.

9. The leachate treatment method based on microalgae coupled nitrification process according to claim 8, characterized in that: The normalized operating parameters collected in the current operating cycle and their corresponding target output values ​​are combined into training sample pairs, and a sliding sample window is constructed together with the historical samples retained in the most recent cycles; the Euclidean distance between the new samples and the historical samples in the parameter distribution space is compared, and samples with high differences are screened out and preferentially added to the weighted multi-parameter fitting relationship.

Images