Intelligent aeration control system for biological sewage treatment

Abstract

The invention relates to the technical field of automatic control of sewage treatment, in particular to an intelligent aeration control system for biological sewage treatment. The system comprises a water quality classification module, an intelligent analysis module and a pulse execution module. Water quality types are divided through the water quality classification module, different response measures are taken for different water qualities through the intelligent analysis module, the pulse execution module adjusts the frequency of a frequency converter and the start-stop period of an air blower according to pulse parameters, the state of an aeration pipe network is monitored in real time, dynamic characteristics are extracted to calculate the health degree index, and the control quantity is corrected through feedforward compensation. The influence of pipe network abnormity on oxygen supply is counteracted, pulse parameters are dynamically adjusted by combining the water inlet load and dissolved oxygen deviation, energy consumption waste caused by low-load and high-aeration is avoided, insufficient dissolved oxygen caused by high-load and low-aeration is prevented, and balance of energy conservation and treatment effects is achieved.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for wastewater treatment, and more specifically, to an intelligent aeration control system for biological wastewater treatment. Background Technology

[0002] In the field of biological wastewater treatment, biodegradability is a key indicator for assessing whether wastewater is easily degraded by microorganisms (activated sludge). Highly biodegradable wastewater refers to wastewater in which organic matter is easily recognized, adsorbed, decomposed, and ultimately converted into CO2, H2O, and its own cellular material by microorganisms. It usually contains a large amount of sugars, proteins, fats, alcohols, organic acids, and other foods that are easily utilized by microorganisms. Domestic sewage, food processing wastewater, and brewing wastewater are typical examples. Lowly biodegradable wastewater (or recalcitrant wastewater) refers to wastewater in which a large amount of organic matter is not easily degraded by microorganisms, or even has an inhibitory or toxic effect on microorganisms. It usually contains complex, stable, or toxic organic matter such as halogenated hydrocarbons, some synthetic dyes, pesticides, and antibiotics. Wastewater from the chemical, pharmaceutical, pesticide, and printing and dyeing industries are typical examples.

[0003] Currently, the aeration control systems commonly used in the industry are mainly based on fixed dissolved oxygen setpoints and use traditional PID control algorithms for adjustment. These systems typically employ fixed parameter control models, which cannot effectively cope with drastic fluctuations in influent water quality (especially biodegradability). For easily degradable domestic sewage and difficult-to-degrade industrial wastewater, the microbial reaction kinetics differ significantly, making it difficult for a single control strategy to simultaneously address both, resulting in unstable treatment effects. Control methods based on dissolved oxygen (DO) feedback have inherent response delays. When encountering instantaneous peak oxygen demand, the system often fails to respond in time, affecting treatment efficiency. Maintaining high aeration rates for extended periods to avoid this problem leads to continuous energy waste. Continuous and uniform aeration modes are inefficient at low influent loads, with large amounts of oxygen escaping unused, resulting in significant energy waste. Therefore, we propose an intelligent aeration control system for biological wastewater treatment. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent aeration control system for biological wastewater treatment, which solves any of the technical problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an intelligent aeration control system for biological wastewater treatment, the system comprising: The water quality classification module is used to collect water quality data at the inlet, including five-day biochemical oxygen demand, chemical oxygen demand and dissolved oxygen concentration in the aeration tank. The ratio of five-day biochemical oxygen demand to chemical oxygen demand is compared with a preset threshold to classify the inlet water quality into high biodegradable wastewater and low biodegradable wastewater. The intelligent analysis module is used to intelligently adjust the parameters of the PID controller based on the dynamic deviation between the measured dissolved oxygen value and the set value through fuzzy reasoning. For wastewater with low biodegradability, the PID parameter correction is converted into optimized pulse control parameters through the dominant factor mapping method. For wastewater with high biodegradability, the control quantity is calculated based on the PID parameter correction using the incremental PID algorithm, and then converted into pulse control parameters by the pulse modulator. The pulse execution module is used to convert pulse control parameters into control commands, control the speed and start-stop cycle of the blower according to the control commands, monitor the status data of the aeration pipeline network in real time, extract the dynamic response characteristics within each pulse cycle, calculate the health index, and perform feedforward compensation on the output control quantity based on the health index.

[0006] As a further improvement to this technical solution, the preset threshold in the water quality classification module is determined based on the historical operating data of the sewage treatment plant. It is determined by correlation analysis of the ratio of five-day biochemical oxygen demand to chemical oxygen demand in the influent and the inflection point of the organic matter removal rate, or by measuring the critical point of the oxygen consumption rate of activated sludge with the change of the ratio of five-day biochemical oxygen demand to chemical oxygen demand through a microbial respiration rate experiment.

[0007] As a further improvement to this technical solution, the intelligent analysis module includes: The fuzzy control unit is used to calculate the deviation between the dissolved oxygen concentration in the aeration tank and the set value, as well as the rate of change of the deviation. The data is input to the fuzzy machine, which performs inference based on the fuzzy rule base and outputs the correction amount of the PID parameters, including the correction amount of the proportional coefficient, the correction amount of the integral coefficient, and the correction amount of the derivative coefficient. The classification and control unit targets wastewater with low biodegradability. Based on the physical semantic association of PID parameter correction, the pulse parameters are obtained. The proportional coefficient correction represents the instantaneous control force requirement and is mapped to the pulse intensity; the integral coefficient correction represents the anti-integral saturation requirement and is mapped to the pulse interval; and the derivative coefficient correction represents the damping and prediction requirements and is mapped to the pulse width. A linear mapping function is used to obtain the pulse parameters. For wastewater with high biodegradability, the PID parameter correction is superimposed on the initial PID parameter. The incremental PID algorithm is used to calculate the real-time change of the control quantity and the rate of change of the control quantity. The pulse modulator reflects the output intensity based on the total control quantity, and the rate of change of the control quantity reflects the urgency of the regulation, and outputs pulse parameters.

[0008] As a further improvement to this technical solution, the fuzzy control unit encodes each rule in the fuzzy rule base as a binary chromosome, and performs selection, crossover and mutation operations on the chromosomes through a genetic algorithm to globally search for the optimal rule set and automatically adjust the fuzzy rule base.

[0009] As a further improvement to this technical solution, when the classification and control unit maps the PID parameter correction amount to pulse parameters, it formulates quantitative constraint boundaries for the pulse parameters based on the physical limitations of the aeration tank, the survival requirements of microorganisms, and the safety boundaries of the equipment. Among these constraints, the pulse intensity corresponds to the output power of the aeration equipment, and the constraint originates from equipment safety and the oxygen supply baseline for microorganisms; the pulse interval corresponds to the time interval between two aerations, and the constraint originates from equipment lifespan and dissolved oxygen response speed; and the pulse width corresponds to the duration of a single aeration, and the constraint originates from aeration effectiveness and energy consumption control.

[0010] As a further improvement to this technical solution, when the pulse parameters change, the classification control unit sets the maximum single change amplitude of each parameter through a pulse smoothing transition mechanism, judges the urgency of adjustment by the rate of change of the control quantity, and dynamically adjusts the maximum allowable sudden change amplitude according to the urgency of adjustment.

[0011] As a further improvement to this technical solution, the pulse execution module includes: The aeration unit adjusts the output frequency of the frequency converter according to the pulse intensity command, controls the speed of the blower, and precisely controls the aeration intensity. According to the pulse width and pulse interval commands, it controls the start and stop cycle of the blower under a specific time sequence to perform pulse aeration. The monitoring and feedback unit monitors the real-time current of the blower motor, extracts dynamic response characteristics such as pressure rise slope, dissolved oxygen response delay, and pressure drop half-life, and converts the degree of deviation of the characteristics from the normal range into health loss, which is then scored using a health index.

[0012] As a further improvement to this technical solution, the aeration execution unit controls the slope and curve of the frequency change when the blower switches between aeration and aeration states, and controls the start and stop of the blower through the soft start / stop function of the frequency converter.

[0013] As a further improvement to this technical solution, the monitoring feedback unit establishes a classification of pipeline network status levels based on the health index, matches exclusive compensation coefficients for different levels of pipeline network status, establishes a correlation model between the health index and the compensation coefficients, and calculates the compensation coefficients in real time through the pipeline network health index to perform targeted correction on the pulse parameters output by the fuzzy PID parameter optimizer.

[0014] As a further improvement to this technical solution, the monitoring feedback unit dynamically adjusts the correction amount of the pulse parameter according to the deviation between the influent load and dissolved oxygen. By quantifying the influent load, a reference range for pulse parameter correction is defined. Within the reference range, the specific correction magnitude is calculated based on the magnitude and trend of the dissolved oxygen deviation.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This intelligent aeration control system for biological wastewater treatment collects five-day biochemical oxygen demand, chemical oxygen demand and dissolved oxygen concentration through a water quality classification module. It classifies water quality types by comparing the biodegradability ratio with a preset threshold, thus avoiding the problem of poor adaptability caused by using a uniform control strategy for high and low biodegradability wastewater. 2. The intelligent analysis module infers the PID parameter correction amount by analyzing dissolved oxygen deviation and change rate. It optimizes the fuzzy rule base with a genetic algorithm to improve the intelligence and accuracy of parameter adjustment, avoiding the limitations of manual experience rules. For wastewater with low biodegradability, it directly associates the PID correction amount with the pulse parameter through the dominant factor mapping method to simplify the calculation. For wastewater with high biodegradability, it uses incremental PID to calculate the control quantity, adapting to its rapid dissolved oxygen fluctuation characteristics. Different measures are taken for different water qualities to improve the system's adaptability. 3. The pulse execution module adjusts the inverter frequency and blower start / stop cycle based on pulse parameters. Combined with soft start / stop function, it avoids equipment impact and extends equipment life. At the same time, it ensures precise control of aeration intensity and timing, reduces ineffective aeration, monitors the aeration network status in real time, extracts dynamic characteristics to calculate the health index, and corrects the control quantity through feedforward compensation to offset the impact of network abnormalities on oxygen supply. It dynamically adjusts pulse parameters based on influent load and dissolved oxygen deviation, avoiding energy waste from low load and high aeration, and preventing insufficient dissolved oxygen from high load and low aeration, thus achieving a balance between energy saving and treatment effect.

[0016] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The invention will now be described in further detail with reference to the figures. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 This is a schematic diagram illustrating the overall detailed process of the present invention.

[0018] The meanings of the labels in the diagram are as follows: 100. Water quality classification module; 200. Intelligent analysis module; 210. Fuzzy control unit; 220. Classification and control unit; 300. Pulse execution module; 310. Aeration execution unit; 320. Monitoring and feedback unit. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Currently, the aeration control systems commonly used in the industry are mainly based on fixed dissolved oxygen setpoints and are adjusted using traditional PID control algorithms. These systems typically employ fixed-parameter control models, which cannot effectively cope with drastic fluctuations in influent water quality (especially biodegradability). For easily degradable domestic sewage and difficult-to-degrade industrial wastewater, the microbial reaction kinetics differ significantly, making it difficult for a single control strategy to simultaneously address both, resulting in unstable treatment effects. Control methods based on dissolved oxygen (DO) feedback have inherent response delays. When encountering instantaneous peak oxygen demand, the system often responds untimely, affecting treatment efficiency. Maintaining high aeration rates for extended periods to avoid this problem leads to continuous energy waste. Continuous and uniform aeration modes are inefficient at low influent loads, with large amounts of oxygen escaping without being fully utilized, resulting in significant energy waste.

[0021] Therefore, this embodiment of the invention proposes that a water quality classification module categorizes water quality according to its biodegradability, directly transmitting signals to an intelligent analysis module. The intelligent analysis module then automatically switches control modes accordingly. For low-biodegradable wastewater, a dominant factor mapping method is used, which has a slow response time; for high-biodegradable wastewater, incremental PID and pulse modulation are used, which adapts quickly to fluctuations, ensuring precise matching between the strategy and water quality and avoiding control mismatch. The intelligent analysis module, combining PID optimization and process constraints, generates compliant and stable pulse parameters, which are directly output to the pulse execution module. The pulse execution module requires no additional adjustments, adjusting the frequency converter and controlling the blower's start and stop according to instructions, and enabling soft start / stop functionality to avoid equipment impact. The pulse execution module monitors the pipeline network status in real time, calculates health status, analyzes the influent load and dissolved oxygen deviation, and outputs compensation requirements to the intelligent analysis module. The intelligent analysis module then corrects the parameters accordingly, forming a closed-loop optimization.

[0022] Specifically as follows: Please see Figure 1 As shown, this embodiment of the invention provides an intelligent aeration control system for biological wastewater treatment, including a water quality classification module 100, an intelligent analysis module 200, and a pulse execution module 300; First, the water quality classification module 100 collects five-day biochemical oxygen demand (BOD5), chemical oxygen demand (COD), and dissolved oxygen (DO) concentrations at the inlet. By comparing the ratio of BOD5 to COD (a biodegradability index) with a preset threshold, the influent is classified into high biodegradability wastewater and low biodegradability wastewater. This solves the problem of traditional aeration systems using a uniform control strategy for wastewater with different biodegradability, which leads to mismatched oxygen supply for low biodegradable wastewater (high in recalcitrant organic matter) and excessive aeration for high biodegradable wastewater (high in readily degradable organic matter). This allows for precise differentiation of water quality types, providing a targeted basis for subsequent regulation and avoiding poor treatment effects or energy waste caused by a one-size-fits-all approach. Then, the intelligent analysis module 200 infers the PID parameter correction amount through DO deviation and rate of change, and optimizes the fuzzy rule base by combining genetic algorithm. For wastewater with low biodegradability, the dominant factor mapping method is used to directly convert the PID correction amount into pulse parameters. For wastewater with high biodegradability, the incremental PID calculates the control amount and converts it into pulse parameters through a pulse modulator. At the same time, the quantitative constraint boundary (equipment safety, microbial requirements, etc.) and smooth transition mechanism of the pulse parameters are formulated to solve the problem that traditional PID control has poor adaptability to complex water quality and that parameter mutations are prone to equipment shock and DO fluctuations. The PID parameters are dynamically optimized to improve control accuracy. Finally, the pulse execution module 300 adjusts the frequency converter frequency (controlling the blower speed) and start / stop cycle (controlling the aeration sequence) according to the pulse parameters. It achieves smooth switching through the soft start / stop function, monitors the pipeline pressure, flow rate and DO response in real time, extracts dynamic features to calculate the health index, and dynamically corrects the pulse parameters based on the health, influent load and DO deviation to achieve feedforward compensation. This solves the problem of reduced oxygen supply efficiency caused by abnormalities such as blockage or leakage in the aeration pipeline. In addition, traditional execution terminals lack real-time feedback and cannot dynamically offset the impact of changes in operating conditions (such as load fluctuations). It accurately executes pulse commands, and the soft start / stop function extends the equipment life. Real-time monitoring and health assessment achieve feedforward compensation to offset the impact of pipeline abnormalities.

[0023] Among them, the water quality classification module 100 is used to collect water quality data at the inlet, including five-day biochemical oxygen demand, chemical oxygen demand and dissolved oxygen concentration in the aeration tank. The ratio of five-day biochemical oxygen demand to chemical oxygen demand is compared with a preset threshold to classify the inlet water quality into high biodegradable wastewater and low biodegradable wastewater. Five-day biochemical oxygen demand (BOD5) refers to the amount of dissolved oxygen consumed by microorganisms in a water sample when decomposing biodegradable organic matter under specified conditions (20℃±1℃). It is typically measured over a five-day period and is expressed in mg / L. It reflects the content of biodegradable organic matter in water and is a core indicator for assessing the biodegradability and pollution level of wastewater. A higher BOD5 indicates more biodegradable organic matter in the water and a greater oxygen demand from microorganisms. BOD5 can be indirectly calculated by rapidly measuring BOD values ​​using a rapid BOD sensor. Chemical oxygen demand (COD) refers to the amount of oxygen consumed by a strong oxidant to oxidize all reducing substances in a water sample under certain conditions. The unit is mg / L. It is commonly measured by the potassium dichromate method (CODcr). It reflects the total amount of all reducing substances in the water body and can quickly assess the overall pollution load of wastewater. COD values ​​can be collected periodically through online COD sensors. Based on the collected five-day biochemical oxygen demand (BOD5) and chemical oxygen demand (COD), the biodegradability ratio is calculated using the following formula: ; in, The value represents the biodegradability ratio, BOD5 is the five-day biochemical oxygen demand, and COD is the chemical oxygen demand.

[0024] The biodegradability ratio is compared with a preset threshold. If the biodegradability ratio is lower than the preset threshold, the inlet water quality is determined to be low biodegradability wastewater. If the biodegradability ratio is higher than the preset threshold, the inlet water quality is determined to be high biodegradability wastewater.

[0025] In order to better determine the preset threshold, the preset threshold in the water quality classification module 100 is determined based on the historical operation data of the sewage treatment plant by correlation analysis of the ratio of five-day biochemical oxygen demand to chemical oxygen demand of influent and the inflection point of organic matter removal rate. By using the operational data accumulated by the wastewater treatment plant, statistical analysis is used to find the point of abrupt change in the ratio and organic matter removal rate. When the ratio exceeds this point, the removal rate increases significantly, indicating that the biodegradability is better. Conversely, the removal rate is low and the biodegradability is poor. This point is the preset threshold. Extract continuous operating data of the plant area in recent years, including daily BOD5, COD and corresponding daily organic matter removal rate of the influent, and calculate the K value according to the formula for the biodegradability ratio. Group the K values ​​at 0.05 intervals, calculate the average removal rate for each group, plot K on the horizontal axis and the average removal rate on the vertical axis, and observe the trend of the curve. When K increases to a certain value, the slope of the removal rate suddenly becomes larger (the point where the second derivative turns from negative to positive). This K value is the inflection point, and the K value is determined to be the preset threshold.

[0026] Considering the lack of accumulated operational data from existing wastewater treatment plants or the construction of new wastewater treatment plants, a microbial respiration rate experiment was conducted to determine the critical point at which the oxygen consumption rate of activated sludge changes with the ratio of five-day biochemical oxygen demand to chemical oxygen demand, thereby determining the preset threshold. Through laboratory simulation, the activity of microorganisms under different K values ​​(reflected by oxygen consumption rate) was directly measured to find the critical point at which the oxygen consumption rate significantly increases. When K exceeds this point, the activity of microorganisms increases sharply and their biochemical properties are good; conversely, the activity is low and their biochemical properties are poor. This point is the preset threshold. Take the mixed liquor from the aeration tank during normal operation, let it settle and discard the supernatant, retain the concentrated sludge, prepare a series of simulated wastewater with different K values, and ensure that the COD concentration of each water sample is uniform. Add an equal amount of activated sludge to each water sample and place them in a constant temperature culture bottle (20℃). Use a DO sensor to monitor the DO change of the mixed liquor in real time and calculate the oxygen consumption rate OUR. Plot a graph with K on the horizontal axis and OUR on the vertical axis, observe the trend of the curve, and the critical point is the K value at which the growth slope of OUR suddenly changes, that is, determine the K value as the preset threshold. Based on historical data from the plant itself or local sludge experiments, thresholds can be set to match the unique composition of the influent (such as when industrial wastewater accounts for a high proportion). Unlike general values, which are easily out of touch with reality, preset thresholds can guide the system to accurately switch control strategies (using a fast response strategy for high biodegradability and a stable oxygen supply strategy for low biodegradability), which not only avoids wasting aeration energy consumption but also ensures the efficiency of organic matter degradation.

[0027] In addition, the intelligent analysis module 200 is used to intelligently adjust the parameters of the PID controller based on the dynamic deviation between the measured dissolved oxygen value and the set value through fuzzy reasoning. If it is low biodegradable wastewater, the PID parameter correction amount is converted into optimized pulse control parameters through the dominant factor mapping method. If it is high biodegradable wastewater, the control amount is calculated based on the PID parameter correction amount through the incremental PID algorithm, and then converted into pulse control parameters through the pulse modulator. Among them, such as Figure 2 As shown, the intelligent analysis module 200 includes: a fuzzy control unit 210 for calculating the deviation between the dissolved oxygen concentration in the aeration tank and the set value and the rate of change of the deviation, inputting it to the fuzzy machine, performing inference according to the fuzzy rule base, and outputting the correction amount of the PID parameters, including the proportional coefficient correction amount, integral coefficient correction amount and derivative coefficient correction amount; Dissolved oxygen concentration (DO) in an aeration tank refers to the amount of dissolved oxygen in the mixed liquor of the aeration tank, measured in mg / L. It directly reflects the balance between the oxygen supply capacity of the aeration system and the oxygen consumption demand of microorganisms. An online fluorescence DO sensor (real-time monitoring) is used. To reflect the overall DO distribution in the aeration tank, samples need to be taken at 3 to 5 representative points, including the inlet, middle, and outlet of the aeration tank, as well as areas with dense and sparse aerator distribution. Dissolved oxygen deviation reflects the difference between the current dissolved oxygen (DO) and the set value, and its calculation formula is as follows: ; in, Dissolved oxygen deviation, in mg / L. Dissolved oxygen setpoint, unit: mg / L The dissolved oxygen value is the measured value, in mg / L.

[0028] The dissolved oxygen setpoint is determined by extracting the DO concentration corresponding to the time when the effluent quality is stable and meets the standards and the energy consumption is lowest from the previous operating data of the wastewater treatment plant, and using this concentration as a fixed setpoint, or by determining it according to the standard process setpoint. The rate of change of deviation reflects the speed at which the deviation changes, and its calculation formula is as follows: ; in, The rate of change of deviation For the present Timing deviation, for The deviation from the previous moment, The sampling interval is denoted as .

[0029] The dissolved oxygen deviation and deviation change rate are fuzzified. Fuzzification is the process of converting precise figures such as dissolved oxygen deviation (e) and deviation change rate (ec) in the aeration tank into fuzzy language descriptions such as negative large and positive small. The steps are as follows: To determine the fuzzy vocabulary, first identify the precise variables to be described, and then assign a set of qualitative terms and corresponding sets to each variable. Taking dissolved oxygen deviation e as an example, the terms include negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). These terms together form the fuzzy set of e. For example, negative large means that DO is much higher than the set value, positive large means that DO is much lower than the set value, and zero means that DO is close to the set value. The deviation change rate ec also uses the same terms and set, but the meaning becomes the speed at which the deviation increases or decreases. For example, positive large means that the deviation increases rapidly. Quantification involves determining the degree of association between precise values ​​and fuzzy terms. A membership function is used to calculate the degree to which each precise value belongs to a particular fuzzy term (between 0 and 1, where 1 represents complete association and 0 represents no association). For example, if we set the range of e from -2 to +2 mg / L, using a trigonometric function, we define that when e = -2 mg / L, it completely belongs to negative large (membership 1); when e = +0.5 mg / L, the degree to positive small is 0.8, the degree to zero is 0.2, and the degree to other terms is 0. This calculates the correlation between precise numbers and fuzzy terms. To determine the fuzzy result, for each precise value, select the fuzzy term with the highest membership degree as the final result. For example, when e = +0.5 mg / L, the membership degree of positive small is the largest at 0.8, so e is described as positive small, and the final deviation is positive small. Such fuzzy language pairs are used as input for subsequent fuzzy inference. Using the above fuzzy results as input, the PID parameter correction amount is inferred based on the fuzzy rule base. The steps are as follows: A fuzzy rule base is pre-programmed with if...then... logic based on aeration experience. For example, if the dissolved oxygen deviation is large (DO is much low) and the deviation change rate is small (deviation is slowly increasing), the proportional coefficient correction is set to large, the integral coefficient correction is set to medium, and the derivative coefficient correction is set to small. Each possible combination of deviation and deviation change rate corresponds to a set of PID correction schemes. Match the current fuzzy state, take the results obtained from the previous fuzzing (such as small positive deviation and small negative rate of change of deviation), find the corresponding rule in the rule base, and find that the rules corresponding to small positive deviation and small negative rate of change are small proportional correction, zero integral correction, and small differential correction. Determine the PID correction amount, and according to the matched rules, directly output the corresponding correction direction (positive for enhancement, negative for reduction) and intensity (large, medium, small), such as proportional coefficient +0.2, integral coefficient 0, derivative coefficient +0.1, to obtain the PID parameter correction amount; Dissolved oxygen fluctuates frequently due to water quality and load. Fuzzy control can adjust PID parameters in real time based on deviation and rate of change, unlike fixed PID which is prone to problems such as not being fast enough or not being stable enough. The fuzzy rule base is based on aeration experience, and the correction amount inferred conforms to the actual operation logic, which can quickly bring DO back to the set value and ensure the treatment effect.

[0030] In order to better optimize the fuzzy rule base, the fuzzy control unit 210 encodes each rule in the fuzzy rule base as a binary chromosome, and performs selection, crossover and mutation operations on the chromosomes through a genetic algorithm to globally search for the optimal rule set and automatically adjust the fuzzy rule base; The fuzzy rules are encoded as binary chromosomes. Each rule consists of a condition part (the fuzzy state of the input variable) and a conclusion part (the fuzzy state of the output variable). A unique binary code is assigned to each fuzzy set. The encoding of a rule is the encoding of the condition part and the encoding of the conclusion part. If there are N rules in the rule base, the chromosomes of the N rules are concatenated to form a long binary string, which is used as the operation object of the genetic algorithm. Genetic algorithms simulate natural selection, allowing the rule sets (chromosomes) with good control effects to be preserved and evolve, while gradually eliminating the rule sets with poor control effects. The specific steps are as follows: M chromosomes are randomly generated, each corresponding to a complete fuzzy rule base (containing N random rules) to form an initial population. The aeration system is controlled using the rule set, and the fitness value is calculated (the higher the score, the better the rule set). The core indicators are the absolute value of dissolved oxygen deviation (the smaller the better), the deviation fluctuation range (the smaller the better), and the aeration energy consumption (the lower the better). The formula for calculating fitness is as follows: ; in, For fitness value, for The weighting coefficients, for The weighting coefficients, The average of the absolute values ​​of the DO deviation over one control cycle. The maximum absolute value of the DO deviation within one control cycle. To find the minimum value, avoid having a denominator of 0.

[0031] Determine the weighting coefficients and The steps are as follows: Through minor adjustments and Observe the matching degree between changes in fitness and control effect, find the optimal value, and fix it. and The sum of is 1. Set 3-5 weight combinations. Each weight combination corresponds to 1 control cycle. Record the average deviation of DO, the maximum deviation and the effluent index. Select the combination that meets the effluent standard and has the smallest DO fluctuation. This weight combination is the optimal one. The fitness value is calculated according to the above formula. The best chromosomes are selected from the population. The chromosome with higher fitness has a higher probability of being selected (the higher the fitness ratio, the greater the chance of being selected). This ensures that the high-quality rule set is preserved. Two parent chromosomes (high-quality rule set) are randomly selected and their codes are swapped at random positions to generate offspring chromosomes. Some binary bits in the offspring chromosomes are randomly selected and flipped (0 becomes 1, 1 becomes 0). The offspring chromosomes are added to the population. The individual with the lowest fitness is eliminated. The population size remains unchanged. The above steps are repeated until the fitness no longer increases. The rule set corresponding to the chromosome at this time is the optimal fuzzy rule base. The population size is set as follows: Based on 7 fuzzy subsets each of the input deviation e and the deviation change rate ec, 49 rules are combined. For a fuzzy rule base containing N=49 rules, the encoding length of each rule is (number of bits for input fuzzy state encoding + number of bits for output fuzzy state encoding). If there are 7 fuzzy subsets each for input and output (encoding requires 3 bits of binary, 2³=8≥7), then the encoding length of a single rule is 3(e)+3(ec)+3(ΔKp)+3(ΔKi)+3(ΔKd)=15 bits. After concatenating the 49 rules, the total chromosome length is 49×15=735 bits. Therefore, a population size of 30-50 chromosomes can cover a sufficient number of rule set combinations. The crossover rate is set as follows: For a long chromosome with 735 positions, single-point crossover is adopted. A crossover rate of 0.7-0.8 can produce offspring in about 70%-80% of the parent combinations. This not only preserves high-quality regular segments, but also explores new rule sets through gene recombination, which is suitable for the search requirements of the nonlinear characteristics of the aeration system. The mutation rate is set as follows: For 735 chromosomes, a mutation rate of 0.01 means that each offspring chromosome undergoes an average of 7-8 mutations (735×0.01≈7). This allows for fine-tuning of the fuzzy state of some rules without significantly disrupting the high-quality rules, thus adapting to the randomness of DO fluctuations in the aeration system. To ensure the algorithm converges to the optimal rule set and avoids invalid iterations, a dual termination condition is set: The rate of change of the optimal fitness value of the population over five consecutive generations is less than or equal to 1%. The fitness value directly reflects the control effect of the fuzzy rule set (the higher the value, the better the rule set). When the change of the optimal fitness value is small over multiple generations, it indicates that the rule set is close to the optimal value, and the improvement of control effect by further iteration is negligible. According to the control accuracy target of the wastewater aeration system, the average deviation of DO is less than 0.2 mg / L, which corresponds to a fitness value that is stable above 0.85. The experiment shows that when the rate of change of fitness value over five consecutive generations is less than 1%, the fluctuation of the average deviation of DO is less than 0.01 mg / L, and the impact on the treatment effect is less than 0.5%, with no substantial room for optimization. The number of iterations reaches 80 generations: Even if the convergence stability constraint is not met, the iteration cycle must be limited by the maximum number of iterations to avoid the algorithm from getting into an infinite loop. At the same time, it should be adapted to the operating condition fluctuation cycle of the sewage aeration system. The typical fluctuation cycle of the influent load is 2-4 hours (such as load changes caused by peak water use in the morning and evening). The algorithm needs to complete the optimization within one fluctuation cycle to ensure that the rule set matches the real-time operating conditions. Based on the calculation that the iteration time of one generation is ≈1.5 minutes, the time for 80 iterations is ≈120 minutes (2 hours), which is in line with the operating condition fluctuation cycle. Genetic algorithms can automatically find a set of rules that are suitable for complex aeration conditions through global search. When the influent water quality or load changes, the algorithm iteration can be restarted to allow the rule base to be updated automatically, always maintaining the optimal control effect. The evolved rule set can more accurately match the dynamic changes of dissolved oxygen and reduce overshoot and lag.

[0032] After the fuzzy machine outputs the correction values ​​of the PID parameters, including the proportional coefficient correction value, integral coefficient correction value, and derivative coefficient correction value, different measures are taken for low biodegradable water quality and high biodegradable water quality. Among them, the classification control unit 220, for low biodegradable wastewater, correlates the pulse parameters based on the physical semantics of the PID parameter correction values. The proportional coefficient correction value represents the instantaneous control force requirement and is mapped to the pulse intensity; the integral coefficient correction value represents the anti-integral saturation requirement and is mapped to the pulse interval; the derivative coefficient correction value represents the damping and prediction requirements and is mapped to the pulse width. A linear mapping function is used to obtain the pulse parameters. The core characteristics of low biodegradability wastewater (such as chemical and dyeing wastewater) are slow DO response and low microbial degradation efficiency. It is necessary to directly correlate the physical semantics of PID correction with pulse parameters and use simple linear mapping to achieve stable control and avoid regulation lag or overshoot. Define the actual control requirements of each PID correction, and then map them to the functions of pulse parameters (intensity, interval, width). The physical meaning of the proportional coefficient correction is the instantaneous control force. When the DO deviation (the difference between the actual value and the set value) increases (such as severe hypoxia), the proportional coefficient correction increases accordingly. Its physical meaning is that the system needs a faster intervention force. At this time, the pulse intensity needs to be increased simultaneously. By increasing the fan frequency, the aeration rate per unit time (i.e., oxygen transfer rate) is increased. The stronger airflow is used to quickly make up for the oxygen gap and shorten the deviation convergence time. The physical meaning of the integral coefficient correction (ΔKi) is to resist integral saturation (avoid cumulative overshoot), corresponding to the pulse interval in the pulse parameters, which is the time difference between two aerations; the physical meaning of the differential coefficient correction (ΔKd) is to damping and prediction (prevent sudden rises and falls in DO), corresponding to the pulse width in the pulse parameters, which is the duration of a single aeration (a smaller width means less oxygen supply per aeration, thus enhancing damping). The core of linear mapping is to proportionally map the range of PID correction values ​​to the actual feasible range of pulse parameters. The formula is unified as follows: Actual value of pulse parameter = slope × PID correction amount + reference value; The steps for determining the slope and the benchmark value are as follows: The slope determines the magnitude of the impact of the PID correction on the pulse parameter. The baseline value is the basic value of the pulse parameter when the PID correction is 0. The essence of linear mapping is to make the extreme value of the PID correction correspond to the extreme value of the pulse parameter. Therefore, two boundary ranges need to be defined first. The range of the PID correction is obtained from historical optimization data or process experience of fuzzy control to ensure that it covers all scenarios that need adjustment. The feasible range of the pulse parameter is determined by the equipment capacity (such as the maximum power of the blower) and process requirements (such as avoiding frequent start-stop), and cannot exceed the safety or efficiency range. The baseline value is the initial value of the pulse parameter when the PID correction is 0 (no adjustment is needed). It is directly taken as the middle value of the feasible range of the pulse parameter to ensure that it is in a safe and neutral state when there is no correction. The formula is: Baseline value = (upper limit of pulse parameter + lower limit of pulse parameter) / 2; The slope determines how much the pulse parameter changes for every unit change in the PID correction. This needs to be inversely calculated using the extreme values ​​of the PID correction and the extreme values ​​of the pulse parameter. The formula has two cases: When the PID correction is positive (maximum value), the pulse parameter must reach the upper limit; when the correction is negative (maximum value), the pulse parameter must reach the lower limit. Substituting into the linear formula: Pulse parameter = slope × correction + reference value, the slope can be solved. When the amount of correction required for partial mapping increases, the pulse parameter decreases instead (e.g., the larger ΔKd is, the stronger the damping is required, and the shorter the single aeration time should be). At this time, the slope is negative, and the slope can be obtained by substituting it into the linear formula.

[0033] In order to better define the constraint boundary for the pulse parameter and ensure the normal operation of the aeration equipment, the classification control unit (220) maps the PID parameter correction amount to the pulse parameter, and sets the quantitative constraint boundary of the pulse parameter based on the physical limitations of the aeration tank, the survival requirements of microorganisms and the safety boundary of the equipment. The pulse intensity corresponds to the output power of the aeration equipment. The constraint stems from equipment safety and the minimum oxygen supply for microorganisms. Long-term full-load (100% power) operation of aeration equipment (such as blowers) will lead to motor overheating, bearing wear, and shortened lifespan. 80% is the safe long-term operating limit recommended by most blower manufacturers to avoid equipment overload damage. Therefore, the upper limit of pulse intensity is set at 80%. Based on the microbial oxygen supply experiment, an experiment was conducted to correlate the power of the aeration tank with dissolved oxygen. The blower power was gradually reduced (from 50% to 10%), and the stable DO value was recorded at different power levels. The power corresponding to when DO was just maintained at 1 mg / L (the minimum requirement for aerobic degradation by microorganisms) was found. If DO stabilized at 1.0 mg / L when the power was reduced to 20%, and DO dropped rapidly below 0.8 mg / L when the power was reduced to 20%, then 20% could be set as the lower limit of pulse intensity. The pulse interval corresponds to the time interval between two aeration cycles. The constraint stems from the equipment lifespan and dissolved oxygen response speed. Frequent start-stop of the aeration equipment will lead to excessive motor starting current, and frequent impacts will accelerate the aging of motor windings and damage to the contactor. Based on experimental data on the impact of fan start-stop frequency on lifespan, intervals of 10s, 20s, 30s, and 40s were set, and the equipment was run continuously for 24 hours. The motor starting current and contactor temperature rise were recorded. If the interval is <30s, the peak starting current exceeds the rated value by 5 times, the contactor temperature rise reaches 60℃, and the daily failure risk increases by 15%. When the interval is ≥30s, the starting current stabilizes within 3 times the rated value, and the temperature rise is ≤40℃. Therefore, 30s is the lower limit of the interval. To test the natural decay rate of dissolved oxygen (DO) in low biodegradable wastewater, the DO was first raised to the set value, then aeration was stopped, and the decay curve of DO over time was recorded. If the DO decreased from 2 mg / L to 1.2 mg / L after 120 seconds of aeration stopping, and decreased to 0.9 mg / L (below the bottom line) after 130 seconds of aeration stopping, then 120 seconds is the upper limit of the interval to avoid excessive DO decay. The pulse width corresponds to the duration of a single aeration cycle. The constraint stems from aeration effectiveness and energy consumption control. Oxygen utilization was tested for different aeration durations, with pulse widths set to 5s, 10s, and 15s respectively. Aeration was performed at the same intensity, and the increase in dissolved oxygen (DO) was recorded using a dissolved oxygen meter after aeration. If the pulse width was <10s, the oxygen dissolution efficiency was <30% (bubbles escaped before fully dissolving), and the DO increase was only 0.3 mg / L. If the pulse width was ≥10s, the dissolution efficiency was >60%, and the DO increase was 0.8 mg / L. Therefore, 10s is the lower limit for the pulse width to ensure effective aeration. The energy consumption and DO overshoot were statistically analyzed for different aeration widths. Widths of 30s, 40s, and 50s were set respectively, and the energy consumption (kWh) and DO peak value after aeration were recorded. When the width was 40s, the energy consumption was moderate (0.5kWh / cycle), and the DO peak value was 3.0mg / L (not exceeding the 4mg / L waste threshold). When the width was 50s, the energy consumption increased to 0.7kWh / cycle, and the DO peak value was 4.2mg / L (over-aeration). Therefore, 40s is the upper limit of the width to balance efficiency and energy consumption. All constraint data follow the principle of prioritizing experimental testing and supplementing with manual parameters. For equipment-related constraints (upper limit of strength, lower limit of interval): the equipment manual is consulted first, followed by on-site experimental verification. For microbial / process-related constraints (lower limit of strength, upper limit of interval, upper and lower limits of width): the critical values ​​that just meet the requirements and do not exceed the limits are found through parameter-effect correlation experiments. The final data can be fine-tuned according to the aeration tank volume, wastewater type, and equipment model to ensure that it fits the actual working conditions.

[0034] Specifically, for highly biodegradable wastewater, the PID parameter correction is superimposed on the initial PID parameter, and the incremental PID algorithm is used to calculate the real-time change of the control quantity and the rate of change of the control quantity. The pulse modulator reflects the output intensity based on the total control quantity, and the rate of change of the control quantity reflects the urgency of the adjustment, and outputs pulse parameters. The core characteristics of highly biodegradable wastewater (such as domestic sewage and food wastewater) are rapid DO response and large fluctuations in microbial oxygen consumption. Sensitive control is achieved through real-time correction of PID parameters, incremental calculation, and dynamic pulse modulation to avoid DO overshoot or lag. The steps are as follows: Based on process experience or on-site commissioning (such as COD / BOD degradation in ordinary aerobic stages), the initial values ​​need to be adapted to the characteristics of DO being prone to fluctuation and requiring rapid stabilization. Considering the fast DO response of highly biodegradable wastewater, the preset proportional coefficient is 1.5 to 3.0 to quickly respond to DO deviations; the integral coefficient is 0.3 to 0.8, not too large as it will result in overshoot; the derivative coefficient is 0.5 to 1.2 to anticipate DO changes in advance and stabilize fluctuations. Use step tests to fine-tune the preset initial values. For example, increase the DO setting from 2 mg / L to 3 mg / L, and adjust the parameters according to the DO change curve. If the response is slow, increase the proportional coefficient; if there is a lot of overshoot, decrease the integral coefficient or increase the derivative coefficient; if the DO differs greatly from the set value after stabilization, increase the proportional coefficient; if the DO fluctuates back and forth, decrease the proportional coefficient. The PID correction values ​​(ΔKp, ΔKi, ΔKd) from the fuzzy control output are superimposed on the initial parameters to dynamically optimize PID performance, adapt to real-time changes in DO, and compensate for the fixed defects of the initial parameters, allowing the PID parameters to be adjusted in real time with DO deviations (such as a sudden drop in DO due to a sudden high load). The advantage of incremental PID is that it only calculates the change in the control quantity (rather than the total control quantity), avoiding abrupt changes in the control quantity caused by integral accumulation. It is perfectly suited to the rapid fluctuations of highly biodegradable wastewater. The formula for the real-time change in the control quantity is as follows: ; in, This represents the increment of the control quantity that needs to be adjusted at the current moment. This is the scaling factor after superposition. The integral coefficients after superposition are... These are the differential coefficients after superposition. For the current deviation, This is the deviation from the previous moment. This represents the deviation between the first two time points.

[0035] If Δu(k) is positive, the aeration output needs to be increased (e.g., DO is too low); if it is negative, the output needs to be decreased (e.g., DO is too high); the larger the absolute value, the stronger the output required. The rate of change of the control quantity reflects the urgency of the regulation, and the formula is as follows: ; in, To control the rate of change of the quantity, This represents the increment of the control quantity that needs to be adjusted at the current moment. This is the increment of the control quantity that needs to be adjusted in the previous moment.

[0036] The pulse modulator takes the change in control quantity and the rate of change as input, and dynamically generates pulse parameters in combination with equipment and microbial constraints. The core logic is to match the intensity of the output and the urgency of the frequency. The pulse intensity is determined based on the change in the control quantity. The larger the change in the control quantity, the higher the pulse intensity, which directly matches the oxygen supply capacity. The smaller the change in the control quantity, the lower the intensity, to avoid overshoot and always limit it to the 20%-80% equipment safety range. The pulse interval is determined based on the rate of change of the control quantity. The larger the rate of change of the control quantity, the shorter the interval, and the faster the adjustment frequency. The smaller the rate of change of the control quantity, the longer the interval, and the less equipment starts and stops. The interval is controlled between 30s and 120s to balance equipment life and response speed. The pulse width is finely adjusted based on the change in the control quantity and the rate of change of the control quantity. When both the change in the control quantity and the rate of change of the control quantity are large (strong demand, high urgency), the width is appropriately widened to enhance the oxygen supply per pulse. If only the change in the control quantity is large but the rate of change is small, the width is slightly reduced to avoid over-aeration. The width is limited to 10s-40s to ensure effective oxygen supply without wasting energy. The superposition of correction values ​​allows for real-time optimization of PID parameters, adapting to DO fluctuations. Incremental calculations prevent sudden changes in control values, meeting the requirements for fast response. The pulse modulator converts intensity and urgency into specific parameters, achieving precise aeration.

[0037] To avoid sudden changes in pulse parameters that could damage the frequency converter, the classification control unit 220 uses a pulse smoothing transition mechanism to set the maximum single change amplitude of each parameter when the pulse parameters change. It judges the urgency of the adjustment by the rate of change of the control quantity and dynamically adjusts the maximum allowable sudden change amplitude according to the urgency of the adjustment. When pulse parameters (intensity, interval, width) change, the core of the smooth transition mechanism is amplitude limiting and gradual change: by setting the maximum single change amplitude of each parameter (avoiding sudden changes), and at the same time dynamically adjusting the gradual change speed according to the urgency of the adjustment (ensuring timely response), both the equipment is protected and the DO is stabilized. Based on the equipment's capacity and DO stability requirements, a limit is set for the amplitude of a single adjustment of pulse intensity, interval, and width, serving as the bottom line for abrupt change constraints. When the difference between the newly calculated pulse parameter (target value) and the current operating parameter exceeds the maximum allowable abrupt change amplitude, the parameter does not jump directly to the target value but is gradually adjusted according to the maximum amplitude until the target value is reached. If the difference does not exceed the upper limit, the parameter is updated directly. The urgency of adjustment is determined by the rate of change of the control quantity, and the maximum allowable abrupt change is dynamically switched. When the absolute value of the rate of change of the control quantity is large (high urgency, requiring rapid response): the maximum amplitude during emergency is enabled to speed up the transition and avoid control lag; when the absolute value of the rate of change of the control quantity is small (low urgency, requiring stability as the main focus): the normal maximum amplitude is enabled to strictly limit abrupt changes and prioritize smoothing.

[0038] When the intelligent analysis module 200 takes countermeasures for wastewater with low biodegradability and wastewater with high biodegradability and outputs pulse parameters, the pulse execution module 300 receives the pulse parameters and controls the fan speed and start-stop cycle. In addition, the pulse execution module 300 is used to convert pulse control parameters into control commands, control the speed and start-stop cycle of the blower according to the control commands, monitor the status data of the aeration pipeline network in real time, extract the dynamic response characteristics within each pulse cycle, calculate the health index, and perform feedforward compensation on the output control quantity according to the health index.

[0039] The pulse execution module 300 includes an aeration execution unit 310 that adjusts the output frequency of the frequency converter according to the pulse intensity command, controls the speed of the blower, precisely controls the aeration intensity, and controls the start-stop cycle of the blower under a specific time sequence according to the pulse width and pulse interval commands to perform pulse aeration. The pulse intensity command (e.g., 70%) is a percentage signal that needs to be converted into a frequency command for the inverter. The intensity percentage is linearly correlated with the frequency range of the inverter. After receiving the frequency command (35Hz), the inverter outputs the corresponding AC frequency to drive the blower motor to operate at this frequency. The speed is proportional to the frequency (at 35Hz, the speed = rated speed × 70%), thereby precisely controlling the aeration intensity (the higher the speed, the stronger the aeration). Based on pulse width and pulse interval commands, the start-stop cycle of the blower is controlled under a specific time sequence. The pulse width command is the duration for which the blower runs continuously at the current intensity. Upon receiving the command, the frequency converter starts the blower and begins timing, maintaining a 35Hz output for 35 seconds. The interval command (e.g., 40 seconds) is the stop time between two runs. After the 35-second run ends, the frequency converter stops outputting, the blower stops, and timing begins, maintaining a stopped state for 40 seconds. After a 40-second stop, the cycle of 35 seconds run + 40 seconds stop repeats, achieving a pulse aeration rhythm of strong aeration, aeration stop, and strong aeration. If the pulse parameters are 70% intensity, 35s width, and 40s interval, the frequency converter first outputs 35Hz (70% intensity), the fan runs for 35s (aeration), then stops outputting 0Hz, the fan stops for 40s (no aeration), and repeats the above process to form a pulse rhythm of 35s aeration + 40s stop, which can accurately control oxygen supply according to intensity and control the aeration interval according to time sequence. Traditional aeration requires continuous operation for 24 hours, with the blower consuming energy under high load for extended periods. Pulse aeration, on the other hand, cycles through aeration and shutdown, starting only when needed (such as when DO decreases), thus reducing blower operating time. For example, during low-load periods, traditional aeration still requires low-power operation, while pulse aeration can significantly reduce ineffective energy consumption by extending the interval and shortening the width, making it particularly suitable for scenarios with large day-night load fluctuations.

[0040] In order to better control the start and stop of the blower, the aeration execution unit 310 controls the slope and curve of the frequency change when the blower switches between aeration and aeration states, and controls the start and stop of the blower through the soft start and stop function of the frequency converter. To control the slope and curve of frequency change when the blower switches between aeration and aeration states, the start and stop of the blower are controlled by the soft start / stop function of the frequency converter. During the startup phase, set the frequency ramp rate (e.g., 5Hz / second), meaning starting from 0Hz, it increases uniformly by 5Hz per second until the target frequency corresponding to the pulse intensity (e.g., 35Hz) is reached. For example, if the target is 35Hz, it takes 7 seconds to increase from 0 to 35Hz, rather than an instantaneous jump. Prioritize using a linear curve (frequency increases uniformly over time) to ensure a smooth increase in speed. If the equipment is sensitive to startup current, an S-curve (slow initial acceleration, fast speed in the middle, and slow down again when approaching the target) can be selected to further reduce startup shock. During the stopping phase, set the frequency decrease slope (e.g., 3Hz / second), starting from the target frequency (35Hz), and uniformly decrease by 3Hz per second until 0Hz. For example, it takes about 12 seconds to decrease from 35Hz to 0Hz. To avoid a sudden drop in pipeline pressure and rapid drop in DO caused by a sudden shutdown, a linear curve is usually used. If more stable DO is required (e.g., for highly biodegradable wastewater), the decrease slope can be appropriately reduced (e.g., 2Hz / second), and the deceleration time can be extended to allow the DO to fall back slowly. By setting the acceleration and deceleration times using a frequency converter, the rate of frequency change can be directly defined, ultimately reducing the starting current from 3-5 times the rated value to less than 1.5 times. When the machine stops, the back electromotive force of the motor is greatly reduced, avoiding damage to the windings and bearings. The DO will not rise or fall suddenly due to sudden start and stop of aeration, resulting in a more stable microbial environment.

[0041] Among them, the monitoring feedback unit 320 monitors the real-time current of the blower motor in real time, extracts the dynamic response characteristics such as pressure rise slope, dissolved oxygen response delay and pressure drop half-life, and converts the degree of deviation of the characteristics from the normal range into health loss, and scores it using the health index. Monitoring the real-time current of the blower motor directly reflects the blower load, which is strongly related to pipeline pressure, aeration efficiency, and sealing performance. The pressure rise slope, dissolved oxygen response delay, and pressure drop half-life characteristics can be indirectly obtained through current changes. By leveraging the correlation between blower motor current and characteristics, the normal ranges (including median, upper limit, and lower limit) of three characteristics are pre-defined as a benchmark for quantifying deviations. These normal ranges are determined by continuous monitoring data during system commissioning. Based on real-time monitored characteristic values ​​(indirectly extracted from current), the degree of deviation from the normal range is quantified using a deviation coefficient, as shown in the following formula: ; in, This is the deviation coefficient. For real-time monitoring of feature values, This is the median value within the normal range for this feature. This represents the maximum value within the normal range of this feature. This is the minimum value within the normal range of this feature.

[0042] The deviation coefficient is greater than 0. The larger the value, the more serious the deviation from the normal range (0 means no deviation, 1 means the deviation is equal to the width of the normal range, and greater than 1 means serious deviation). The health score of a single feature (Hᵢ, i=1,2,3 corresponding to the three features respectively) is negatively correlated with the deviation coefficient, with a value ranging from 0 to 1 (1 for healthy, 0 for severely abnormal), as shown in the following formula: ; in, For a single health index, This is the deviation coefficient.

[0043] The three characteristics (pressure rise slope, DO response delay, and pressure fall half-life) correspond to the unobstructedness, aeration efficiency, and sealing of the pipeline network, respectively. If any one of these characteristics is severely abnormal (low score), the overall health of the pipeline network will be affected. Therefore, the minimum value among the individual scores of the three characteristics is directly taken as the final health index. If the score of a certain characteristic is extremely low, the overall health is directly 0, and the most serious problem is exposed first. If all three characteristics are healthy, the minimum value among the individual scores of the three characteristics is taken as the final health index, reflecting the weakest link at present. The lowest score among the three individual scores is taken as the overall health of the pipeline network, completely eliminating the need for weight settings and greatly simplifying the calculation (only three individual scores need to be calculated and the minimum value is taken). Even non-professional maintenance personnel can operate it quickly. It can also expose the most serious problems first and avoid high scores masking low scores after weighting.

[0044] In order to better correct the pulse parameters, the monitoring feedback unit 320 divides the pipeline network status level according to the health index, matches exclusive compensation coefficients for different levels of pipeline network status, establishes a correlation model between the health index and the compensation coefficient, and calculates the compensation coefficient in real time through the pipeline network health index to perform targeted correction of the pulse parameters output by the fuzzy PID parameter optimizer. The pipeline network status is classified according to the health index, and a special compensation coefficient is matched for each level (corresponding to the correction direction to offset pipeline network problems). The pulse parameters are then dynamically adjusted using the coefficient to ensure that even if there are problems such as blockage or leakage in the pipeline network, the aeration requirements can still be met. Using the health index (0-100 points) as input, and classifying the previous status (healthy, warning, fault), analyze the core pipeline problems (such as blockage, leakage) under each status, and design targeted compensation coefficients for pulse intensity, interval, and width (positive and negative coefficients represent the correction direction, and absolute values ​​represent the correction magnitude). A positive strength coefficient indicates that blockages or leaks in the pipeline will result in a lower actual oxygen supply than the commanded amount, requiring an increase in strength to compensate. A negative interval coefficient indicates that leakage will cause a rapid drop in pressure, requiring a shorter interval to compensate for the loss. A positive width coefficient: This ensures that the total oxygen supply meets the standard by adjusting the strength and spacing, and avoids excessive correction that could lead to energy waste. The correction logic is: original parameter × (1 + compensation coefficient) (intensity / width) or original parameter × (1 - absolute value of compensation coefficient) (interval; since the interval coefficient is negative, the interval is shortened by subtraction). At the same time, the original pulse parameter safety constraints (intensity 20%-80%, interval 30s-120s, width 10s-40s) are retained to avoid exceeding the limits. The calculated health index (0-100 points, which may be decimals) is rounded to the nearest integer to obtain an integer score, which facilitates interval matching. Based on the preset intervals (80-100 points, 60-79 points, 40-59 points, 20-39 points, 0-19 points) in which the integer score falls, the corresponding pipeline problem level is determined. The controller has a pre-stored compensation coefficient table that corresponds one-to-one with the above intervals, and the intensity, interval, and width coefficients of the current interval are directly extracted. The steps for determining the compensation coefficient table are as follows: Based on the oxygen supply loss rate caused by pipeline problems (blockage / leakage), the approximate range of coefficients is first defined. In a pilot-scale device consistent with the actual system, pipeline problems corresponding to different health levels (such as minor anomalies, moderate blockages, etc.) are simulated. The basic values ​​of pulse parameters are fixed, and the effects of multiple coefficient combinations are tested. The coefficients are fine-tuned according to the actual operating conditions of the target wastewater treatment plant (such as blower power and pipeline length) to ensure that they are adapted to the site requirements. Finally, the determined coefficients are organized into a table according to the health level range and stored in the controller to form a compensation coefficient table that can be directly called. The correlation model directly looks up the coefficients in the table according to the health interval, without the need for complex algorithms. Operation and maintenance personnel can operate quickly. When the health of the pipeline changes, the compensation coefficients automatically switch without manual intervention. After correction, a mandatory check of safety constraints is performed to avoid equipment impact or DO fluctuations caused by compensation.

[0045] In order to better dynamically adjust the correction amount of the pulse parameters, the monitoring feedback unit 320 dynamically adjusts the correction amount of the pulse parameters according to the deviation between the influent load and dissolved oxygen. By quantifying the influent load, the reference range of pulse parameter correction is defined. Within the reference range, the specific correction magnitude is calculated according to the magnitude and trend of the dissolved oxygen deviation. First, determine the pulse baseline value based on the influent load (to control energy consumption), and then calculate the correction amount based on the dissolved oxygen deviation (to ensure oxygen supply). Through these two steps of adjustment, the pulse parameters can be adapted to the load demand and compensate for DO fluctuations in real time. First, determine the basic values ​​according to the influent load: use low intensity, long interval, and short width for low load (to save energy), medium load use medium parameters, and high load use high intensity, short interval, and wide width (to ensure basic oxygen supply). Next, let's look at the dissolved oxygen deviation adjustment: If the DO is higher than the set value (positive deviation), reduce the intensity, increase the interval, and decrease the width (less oxygen supply); if the DO is lower than the set value (negative deviation), increase the intensity, decrease the interval, and increase the width (more oxygen supply); if the deviation is small, no adjustment is needed. Calculate the final parameters: multiply the base value by (1 + correction factor), and then check whether it is within the equipment's safe range (e.g., strength 20%-80%). If there are no problems, proceed. For example: if the influent COD is high (high load), the base value is 65% intensity, 40s interval, and 32s width; if the real-time DO is low (negative deviation), increase the intensity by 15%, shorten the interval by 15%, and increase the width by 20%, resulting in an intensity of 74.75%, an interval of 34s, and a width of 38.4s, which is sufficient for oxygenation without waste. Every adjustment has a clear objective: the baseline value is set according to the load to directly avoid excessive aeration and wasted electricity at low loads; the adjustment is adjusted according to DO to accurately solve the problem of insufficient oxygen supply and DO deficiency at high loads. There are no ambiguities in the logic, and the effect of reduced energy consumption and stable DO can be seen intuitively after execution, without the need for repeated trial and error.

[0046] In summary, the working principle of this solution is as follows: This intelligent aeration control system for biological wastewater treatment collects five-day biochemical oxygen demand, chemical oxygen demand, and dissolved oxygen concentration through the water quality classification module 100. It then compares the biodegradability ratio with a preset threshold to classify water quality types, thus avoiding the problem of poor adaptability caused by using a uniform control strategy for high and low biodegradability wastewater. The intelligent analysis module 200 infers the PID parameter correction amount by analyzing dissolved oxygen deviation and change rate. It optimizes the fuzzy rule base with a genetic algorithm to improve the intelligence and accuracy of parameter adjustment, avoiding the limitations of manual experience rules. For wastewater with low biodegradability, it directly associates the PID correction amount with the pulse parameter through the dominant factor mapping method to simplify the calculation. For wastewater with high biodegradability, it uses incremental PID to calculate the control amount to adapt to its rapid dissolved oxygen fluctuation characteristics, and takes different measures for different water qualities to improve the system's adaptability. The pulse execution module 300 adjusts the frequency of the inverter and the start-stop cycle of the blower according to the pulse parameters. Combined with the soft start-stop function, it avoids equipment impact and extends equipment life. At the same time, it ensures precise control of aeration intensity and timing, reduces ineffective aeration, monitors the status of the aeration pipeline network in real time, extracts dynamic characteristics to calculate the health index, and corrects the control quantity through feedforward compensation to offset the impact of pipeline abnormalities on oxygen supply. It dynamically adjusts the pulse parameters in combination with the deviation of influent load and dissolved oxygen, which avoids the energy waste of high aeration under low load and prevents insufficient dissolved oxygen under high load and low aeration, thus achieving a balance between energy saving and treatment effect.

[0047] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent aeration control system for biological wastewater treatment, characterized in that, The system includes: The water quality classification module (100) is used to collect water quality data at the inlet, including five-day biochemical oxygen demand, chemical oxygen demand and dissolved oxygen concentration in the aeration tank. The ratio of five-day biochemical oxygen demand to chemical oxygen demand is compared with a preset threshold to classify the inlet water quality into high biodegradable wastewater and low biodegradable wastewater. The intelligent analysis module (200) is used to intelligently adjust the parameters of the PID controller based on the dynamic deviation between the measured dissolved oxygen value and the set value through fuzzy reasoning. If it is low biodegradable wastewater, the PID parameter correction amount is converted into optimized pulse control parameters through the dominant factor mapping method. If it is high biodegradable wastewater, the control amount is calculated based on the PID parameter correction amount through the incremental PID algorithm, and then converted into pulse control parameters through the pulse modulator. The pulse execution module (300) is used to convert pulse control parameters into control commands, control the speed and start-stop cycle of the blower according to the control commands, monitor the status data of the aeration pipeline network in real time, extract the dynamic response characteristics within each pulse cycle, calculate the health index, and perform feedforward compensation on the output control quantity according to the health index.

2. The intelligent aeration control system for biological wastewater treatment according to claim 1, characterized in that: The preset threshold in the water quality classification module (100) is determined based on the historical operating data of the sewage treatment plant by correlation analysis of the ratio of five-day biochemical oxygen demand to chemical oxygen demand of influent and the inflection point of organic matter removal rate, or by measuring the critical point of the change of oxygen consumption rate of activated sludge with the ratio of five-day biochemical oxygen demand to chemical oxygen demand through microbial respiration rate experiments.

3. The intelligent aeration control system for biological wastewater treatment according to claim 1, characterized in that: The intelligent analysis module (200) includes: The fuzzy control unit (210) is used to calculate the deviation between the dissolved oxygen concentration in the aeration tank and the set value and the rate of change of the deviation, input the data to the fuzzy machine, perform inference based on the fuzzy rule base, and output the correction amount of the PID parameters, including the correction amount of the proportional coefficient, the correction amount of the integral coefficient and the correction amount of the derivative coefficient. The classification control unit (220) targets low biodegradability wastewater and uses the physical semantic association of pulse parameters based on the PID parameter correction amount. The proportional coefficient correction amount represents the instantaneous control force requirement and is mapped to the pulse intensity; the integral coefficient correction amount represents the anti-integral saturation requirement and is mapped to the pulse interval; and the derivative coefficient correction amount represents the damping and prediction requirements and is mapped to the pulse width. A linear mapping function is used to obtain the pulse parameters. For wastewater with high biodegradability, the PID parameter correction is superimposed on the initial PID parameter. The incremental PID algorithm is used to calculate the real-time change of the control quantity and the rate of change of the control quantity. The pulse modulator reflects the output intensity based on the total control quantity, and the rate of change of the control quantity reflects the urgency of the regulation, and outputs pulse parameters.

4. The intelligent aeration control system for biological wastewater treatment according to claim 3, characterized in that: The fuzzy control unit (210) encodes each rule in the fuzzy rule base as a binary chromosome, performs selection, crossover and mutation operations on the chromosomes through a genetic algorithm, searches the global optimal rule set, and automatically adjusts the fuzzy rule base.

5. The intelligent aeration control system for biological wastewater treatment according to claim 3, characterized in that: When the classification and control unit (220) maps the PID parameter correction amount to the pulse parameter, it formulates the quantitative constraint boundary of the pulse parameter based on the physical limitations of the aeration tank, the survival requirements of microorganisms, and the safety boundary of the equipment. Among them, the pulse intensity corresponds to the output power of the aeration equipment, and the constraint is derived from the equipment safety and the microbial oxygen supply baseline; the pulse interval corresponds to the time interval between two aerations, and the constraint is derived from the equipment life and dissolved oxygen response speed; the pulse width corresponds to the duration of a single aeration, and the constraint is derived from the aeration effectiveness and energy consumption control.

6. The intelligent aeration control system for biological wastewater treatment according to claim 3, characterized in that: When the pulse parameters change, the classification control unit (220) sets the maximum single change amplitude of each parameter through the pulse smoothing transition mechanism, judges the urgency of the adjustment by the rate of change of the control quantity, and dynamically adjusts the maximum allowable sudden change amplitude according to the urgency of the adjustment.

7. The intelligent aeration control system for biological wastewater treatment according to claim 1, characterized in that: The pulse execution module (300) includes: The aeration execution unit (310) adjusts the output frequency of the frequency converter according to the pulse intensity command, controls the speed of the blower, accurately controls the aeration intensity, and controls the start-stop cycle of the blower under a specific time sequence according to the pulse width and pulse interval commands to perform pulse aeration. The monitoring feedback unit (320) monitors the real-time current of the blower motor, extracts the dynamic response characteristics such as pressure rise slope, dissolved oxygen response delay and pressure drop half-life, and converts the degree of deviation of the characteristics from the normal range into health loss, and scores it using the health index.

8. The intelligent aeration control system for biological wastewater treatment according to claim 7, characterized in that: The aeration execution unit (310) controls the start and stop of the blower by controlling the slope and curve of the frequency change when the blower switches between aeration and aeration states, through the soft start and stop function of the frequency converter.

9. The intelligent aeration control system for biological wastewater treatment according to claim 7, characterized in that: The monitoring feedback unit (320) classifies the pipeline network status level according to the health index, matches exclusive compensation coefficients for different levels of pipeline network status, establishes a correlation model between the health index and the compensation coefficient, calculates the compensation coefficient in real time through the pipeline network health index, and performs targeted correction on the pulse parameters output by the fuzzy PID parameter optimizer.

10. The intelligent aeration control system for biological wastewater treatment according to claim 9, characterized in that: The monitoring feedback unit (320) dynamically adjusts the correction amount of the pulse parameter according to the deviation between the influent load and dissolved oxygen. By quantifying the influent load, the reference range for pulse parameter correction is defined. Within the reference range, the specific correction magnitude is calculated based on the magnitude and trend of the dissolved oxygen deviation.

Images