Intelligent control method, system and program product for sewage treatment

Abstract

The application discloses a sewage treatment intelligent regulation and control method, system and program product, relates to the technical field of sewage treatment, and comprises the following steps: constructing a multi-modal water quality sensing network, acquiring sewage water quality parameter information and overall treatment target numerical information; generating sewage treatment process parameters according to the equipment operation state and the overall treatment target numerical information, wherein the treatment units in the sewage treatment system are distributed in multiple regions; triggering process regulation and control instructions according to the sewage treatment process parameters and the overall treatment target numerical information; acquiring partition treatment effect detection information, comparing the partition treatment effect detection information with corresponding index values, and obtaining partition difference values; if the partition difference values exceed a preset change threshold range, triggering partition adaptive regulation and control update instructions; acquiring partition adjustment parameter information according to the partition adaptive regulation and control update instructions; and triggering adjustment instructions. The application provides an intelligent regulation and control method based on multi-dimensional perception and adaptive decision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewage treatment, in particular to a sewage treatment intelligent control method, system and program product. BACKGROUND

[0002] In the existing sewage treatment system, aeration control generally relies on manual experience to set fixed parameters, and the problem of uneven distribution of dissolved oxygen concentration in the aeration tank has existed for a long time. For example, the traditional aeration system uses a single dissolved oxygen sensor to monitor the global state, but due to the large spatial span of the aeration tank and the complex water flow disturbance, the sensor data is difficult to reflect the oxygen supply and demand contradiction in the local area. When the influent water quality fluctuates, the aeration quantity adjustment has obvious hysteresis, which leads to excessive dissolved oxygen concentration (>5mg / L) in some areas, causing energy waste, while the dissolved oxygen concentration in other areas is insufficient (<1mg / L), inhibiting microbial activity, and directly affecting the biological treatment efficiency.

[0003] The dissolved oxygen concentration in the existing aeration tank of the sewage treatment system presents a significant spatial gradient distribution, but the existing control strategy does not consider the oxygen transfer difference in the longitudinal (water depth direction) and transverse (tank cross-section) directions, resulting in coexistence of oxygen supersaturation near the aeration head and anoxic zone at the far end.

[0004] In view of the above related technologies, an intelligent control method based on multi-dimensional perception and adaptive decision is urgently needed. SUMMARY

[0005] In order to provide an intelligent control method based on multi-dimensional perception and adaptive decision, the present application provides a sewage treatment intelligent control method, system and program product.

[0006] In the first aspect, the application aims to achieve the following technical solutions:

[0007] The sewage treatment intelligent control method comprises:

[0008] A multi-modal water quality sensing network is constructed to obtain water quality parameter information of the sewage, and according to the water quality parameter information of the sewage, overall treatment target numerical information is obtained;

[0009] The device running state is obtained, and according to the device running state and the overall treatment target numerical information, sewage treatment process parameters are generated, wherein the treatment units in the sewage treatment system are distributed in multiple regions;

[0010] According to the sewage treatment process parameters and the overall treatment target numerical information, process control instructions are triggered;

[0011] The partition processing effect detection information is obtained, and the partition processing effect detection information is compared with the corresponding index value in the overall treatment target numerical information to obtain a partition difference value.

[0012] The partition difference value is compared with a preset change threshold range, and if the partition difference value exceeds the preset change threshold range, a partition adaptive regulation update instruction is triggered;

[0013] According to the partition adaptive regulation update instruction, partition adjustment parameter information is obtained;

[0014] According to the partition adjustment parameter information, an adjustment instruction is triggered.

[0015] By adopting the above technical solution, a sewage treatment intelligent regulation system capable of accurately sensing and fusing multi-dimensional water quality parameters and monitoring real-time evaluation and adaptive adjustment of partition processing efficiency of the sewage treatment system is obtained. Specifically, by constructing a multi-modal water quality sensing network, multi-dimensional sewage water quality parameter information such as sewage flow, pollutant concentration, pH value, and suspended solids content is synchronously obtained, thereby breaking through the limitation of traditional single-parameter monitoring. Based on sewage treatment process parameters including process control parameters of processing units in each region of the sewage treatment system, coordinated regulation of processing units in different regions is realized, and the problem of system efficiency loss caused by isolated adjustment of parameters in the traditional process is solved. The application further compares the partition processing effect detection information with the target value dynamically, establishes a hierarchical response mechanism based on the deviation value, and automatically triggers a parameter adjustment instruction set when the difference value exceeds the preset threshold range, thereby realizing dynamic regulation on demand from traditional fixed parameter operation, improving the response speed of the system to water quality fluctuations, and solving the problem that the existing aeration system does not consider the oxygen transfer difference in the water depth direction (longitudinal direction) and the cross-section direction (transverse direction) of the tank. The present application introduces a partition processing strategy, which can adjust the oxygen supply according to the actual oxygen demand of different regions, alleviate the phenomenon of oxygen supersaturation in the near-aeration head region and coexistence of the far-end anoxic region, and improve the uniformity of dissolved oxygen distribution in the entire aeration tank. Through accurate control of the dissolved oxygen concentration in each region, unnecessary excessive aeration is avoided, the energy consumption of key equipment such as fans is reduced, which helps to achieve the goal of energy saving and consumption reduction, and improves the overall energy efficiency of the sewage treatment plant. The application continuously collects partition processing effect change information and partition adjustment parameter information, provides data support for optimization of subsequent regulation strategies, has iterative upgrading capability, and is suitable for long-term stable operation under complex working conditions.

[0016] In a preferred example of the present application, the multi-modal water quality sensing network is constructed to obtain sewage water quality parameter information, and overall processing target value information is obtained according to the sewage water quality parameter information, specifically including:

[0017] A distributed water quality sensor array is deployed, which includes a dissolved oxygen sensor, a pH / ORP composite electrode, and a multi-spectral turbidity meter;

[0018] extracting pH value, chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and suspended matter concentration from the distributed water quality sensor array to obtain sewage water quality parameter information;

[0019] According to the sewage water quality parameter information, the overall treatment target numerical information corresponding to the sewage treatment plant effluent standard is determined, and the overall treatment target numerical information includes COD removal rate target value, ammonia nitrogen removal rate target value, total phosphorus removal rate target value, and suspended matter removal rate target value.

[0020] By adopting the above technical solution, real-time monitoring of multiple key parameters (such as pH value, chemical oxygen demand COD, ammonia nitrogen content, etc.) in sewage is realized, and the accuracy and multi-source of data acquisition are improved; according to the extracted sewage water quality parameter information, the overall treatment target numerical information corresponding to the sewage treatment plant effluent standard is determined, ensuring that the sewage treatment process has a clear direction and standard to follow, thereby effectively improving the effect and efficiency of sewage treatment.

[0021] In a preferred example of the present application: according to the device operating state and the overall treatment target numerical information, the sewage treatment process parameters are generated, specifically including:

[0022] According to the device operating state, the bottom layer operating parameters are extracted, including aeration blower current, reflux pump frequency, and dosing pump stroke;

[0023] Based on a fuzzy logic reasoning system, the overall treatment target numerical information is converted into process parameter constraint conditions;

[0024] By using a multi-objective optimization algorithm, the optimal solution set of aeration intensity gradient distribution and dosing rate spatiotemporal configuration is solved, and the sewage treatment process parameters including longitudinal dissolved oxygen concentration gradient in the aeration tank and transverse medicament dosing concentration difference are generated.

[0025] By adopting the above technical solution, quantitative evaluation of equipment performance degradation is realized;

[0026] In a preferred example of the present application: the partition processing effect detection information is obtained, the partition processing effect detection information is compared with the corresponding index value in the overall treatment target numerical information, and the partition difference value is obtained, specifically including:

[0027] The COD concentration, ammonia nitrogen concentration, total phosphorus concentration, and suspended matter concentration detection values of the effluent of each partition are obtained;

[0028] The difference between the water quality index of the effluent of each partition and the corresponding overall treatment target numerical value is calculated to obtain the COD difference value, ammonia nitrogen difference value, total phosphorus difference value, and suspended matter difference value;

[0029] The COD difference value, the ammonia nitrogen difference value, the total phosphorus difference value and the suspended matter difference value are weighted and fused to obtain a comprehensive partition difference value.

[0030] By adopting the technical solution, a partition water quality monitoring network and a difference quantification evaluation system are constructed, each partition of the sewage treatment system is finely monitored, the water quality indexes (such as COD concentration, ammonia nitrogen concentration, etc.) of each partition are detected and compared with the overall treatment target value, the treatment difference value of each partition is calculated, and subsequent accurate regulation and control of the local area are facilitated. The comprehensive partition difference value obtained by weighted fusion of each single difference value can more comprehensively reflect the operation status of the entire sewage treatment system.

[0031] In a preferred example of the present application: the partition difference value is compared with a preset change threshold range, if the partition difference value exceeds the preset change threshold range, a partition adaptive regulation update instruction is triggered; according to the partition adaptive regulation update instruction, partition adjustment parameter information is obtained, specifically including:

[0032] A preset dynamic change threshold of each index is obtained, the dynamic change threshold includes a COD allowable fluctuation interval, an ammonia nitrogen allowable fluctuation interval, a total phosphorus allowable fluctuation interval and a suspended matter allowable fluctuation interval;

[0033] The comprehensive partition difference value is compared with the dynamic change threshold of each index, if the comprehensive partition difference value exceeds the allowable fluctuation interval of any index, it is determined that adaptive regulation is needed, and a partition adaptive regulation update instruction is triggered;

[0034] According to the partition adaptive regulation update instruction, the equipment operation parameters of the current partition and the water quality change trend information in the recent period are obtained;

[0035] Based on the current equipment operation parameters, the water quality change trend information and a preset adaptive adjustment rule, partition adjustment parameter information is calculated and generated by a machine learning algorithm, the partition adjustment parameter information includes an aeration intensity adjustment value, a reagent dosage adjustment value and a sludge return flow adjustment value.

[0036] By adopting the technical solution, the sewage treatment system has adaptive regulation and control capability and dynamic adjustment strategy, the equipment operation parameters of the current partition and the water quality change trend information in the recent period are automatically obtained by triggering the partition adaptive regulation update instruction, the partition adjustment parameter information is calculated and generated by the machine learning algorithm, and self-regulation and optimization of the sewage treatment system are realized. Based on the current equipment operation parameters, the water quality change trend and the preset adaptive adjustment rule, key parameters such as aeration intensity, reagent dosage and sludge return flow are dynamically adjusted, the ability of the system to cope with complex working conditions is enhanced, and the stability and reliability of the treatment effect are improved.

[0037] The application in a preferred example: the total treatment target value information is obtained according to the sewage quality parameter information, and the total treatment target value information further comprises:

[0038] According to the influent quantity, influent quality index and aeration tank volume of the sewage treatment system, the first oxygen demand of the aeration tank is calculated through the ASM2D model; according to the sludge concentration, reflux ratio and oxygen demand calculation coefficient of the sewage treatment system, the second oxygen demand of the sludge reflux system is calculated;

[0039] The total oxygen demand and the aeration gap amount are generated by combining the first oxygen demand and the second oxygen demand;

[0040] According to the aeration gap amount and the aeration efficiency curve, the theoretical aeration demand is obtained; the ratio of the first oxygen demand to the second oxygen demand is calculated to generate the oxygen demand distribution ratio;

[0041] Based on the historical difference between the theoretical aeration demand and the actual aeration demand, the aeration efficiency correction coefficient is obtained;

[0042] According to the oxygen demand distribution ratio and the aeration efficiency correction coefficient, the aeration amount-dissolved oxygen response relationship is generated;

[0043] According to the aeration amount-dissolved oxygen response relationship and the aeration gap amount, the aeration control table is generated.

[0044] By adopting the above technical scheme, the first oxygen demand and the second oxygen demand are calculated by using the ASM2D model according to the influent quantity, influent quality index and aeration tank volume, the aeration efficiency correction coefficient is obtained based on the historical difference of the actual aeration demand, the aeration amount-dissolved oxygen response relationship and the aeration control table are generated, so that the oxygen demand management is more scientific and reasonable; in order to improve the aeration efficiency, the total oxygen demand and the aeration gap amount are calculated, the theoretical aeration demand is obtained by combining the aeration efficiency curve, and the oxygen demand distribution ratio is considered, so that the aeration process is effectively controlled, the aeration efficiency is improved, and the energy consumption is reduced.

[0045] The application in a preferred example: the first oxygen demand The calculation formula comprises:

[0046] Wherein, is the influent quantity; is the influent BOD5 concentration; is the oxygen equivalent coefficient related to nitrogen oxidation; is the influent ammonia nitrogen concentration; is the effluent nitrate nitrogen concentration; is the volatile suspended solid concentration; is the endogenous respiration coefficient; is the aerobic microorganism decay coefficient; is a sludge yield coefficient; is an endogenous respiration rate constant.

[0047] By adopting the above technical solution, the first oxygen demand of the aeration tank is accurately calculated based on the detailed ASM2D model formula; the accurate first oxygen demand calculation helps to more accurately control the oxygen supply in the aeration process, avoiding the problems of microbial activity inhibition or energy waste caused by insufficient or excessive oxygen supply.

[0048] In a preferred example of the present application, the second oxygen demand is calculated according to the following formula:

[0049] wherein, is a residual sludge discharge amount; is a residual sludge MLSS concentration; is a sludge return ratio; is a return sludge MLSS concentration; is an oxygen demand calculation correction coefficient.

[0050] By adopting the above technical solution, the calculation formula of the second oxygen demand is defined in detail, including the residual sludge discharge amount, the residual sludge MLSS concentration, the sludge return ratio and other factors, thereby realizing comprehensive evaluation of the oxygen demand in the sludge return system.

[0051] In a second aspect, the application aims to achieve the following technical solutions:

[0052] The intelligent control system for wastewater treatment comprises:

[0053] A data acquisition and processing module is configured to construct a multi-modal water quality sensing network, acquire wastewater quality parameter information, and acquire overall treatment target numerical information according to the wastewater quality parameter information.

[0054] A process parameter generation module is configured to acquire equipment operating states, generate wastewater treatment process parameters according to the equipment operating states and the overall treatment target numerical information, wherein the treatment units in the wastewater treatment system are distributed in multiple regions.

[0055] A control instruction generation module is configured to trigger process control instructions according to the wastewater treatment process parameters and the overall treatment target numerical information.

[0056] A partition detection and evaluation module is configured to acquire partition treatment effect detection information, compare the partition treatment effect detection information with corresponding index values in the overall treatment target numerical information, and obtain partition difference values.

[0057] The actuator control module is configured to compare the partition difference value with a preset variation threshold range, trigger a partition adaptive control update instruction if the partition difference value exceeds the preset variation threshold range, obtain partition adjustment parameter information according to the partition adaptive control update instruction, and trigger an adjustment instruction according to the partition adjustment parameter information.

[0058] In a third aspect, the application aims to achieve the following technical solutions:

[0059] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of the intelligent sewage treatment control method described above.

[0060] In summary, the present application includes at least one of the following beneficial technical effects:

[0061] 1. By constructing a multi-modal water quality sensing network, real-time monitoring of sewage water quality parameters is achieved, and the overall treatment target value is determined. Combined with the equipment operation state, accurate sewage treatment process parameters are generated, and corresponding control instructions are triggered to optimize the treatment process. Through detection and evaluation of the partition treatment effect, the partition difference value is calculated and compared with the preset threshold, adaptive control update is realized, and the treatment effect of each region is ensured to reach the optimal standard, effectively improving the sewage treatment efficiency and effluent quality, and reducing energy consumption.

[0062] 2. A complete intelligent sewage treatment control system is provided, including data acquisition and processing module, process parameter generation module, control instruction generation module, partition detection and evaluation module, and actuator control module, etc. components, realizing the whole process automation management from water quality monitoring to process control. Through the collaborative work of each module, the system can dynamically adjust the process parameters according to the real-time water quality data and equipment operation state, has strong adaptive ability and flexibility, can effectively cope with the challenges brought by the fluctuation of influent water quality, and ensure that the effluent water quality meets the standard.

[0063] 3. The present application not only considers various parameters involved in the sludge reflux process, but also introduces an oxygen demand calculation correction coefficient, making the oxygen demand management of the whole system more refined, which helps to optimize resource utilization, reduce unnecessary energy consumption, and improve overall operation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 is a flowchart of the intelligent sewage treatment control method in an embodiment of the present application;

[0065] Figure 2 is a flowchart of step S1 in the intelligent sewage treatment control method in an embodiment of the present application. DETAILED DESCRIPTION

[0066] The application will be further described in detail below with reference to the accompanying drawings.

[0067] In an embodiment, as shown in Figure 1 The application discloses an intelligent sewage treatment control method, which specifically comprises the following steps:

[0068] S1: Construct a multi-modal water quality sensor network to obtain water quality parameter information, and obtain overall treatment target numerical information based on the water quality parameter information.

[0069] In this embodiment, the multi-modal water quality sensor network integrates a distributed monitoring system of multiple types of sensors (physical, chemical, optical) to synchronously collect water quality parameters; the overall treatment target numerical value is a key index removal rate target value set based on the effluent standard of a sewage treatment plant (such as the “Discharge Standard of Pollutants for Municipal Wastewater Treatment Plant” GB 18918).

[0070] Specifically, step S1 comprises:

[0071] S11: Deploy a distributed water quality sensor array, which comprises a dissolved oxygen sensor, a pH / ORP composite electrode, and a multi-spectral turbidity meter.

[0072] In this embodiment, the distributed water quality sensor array constitutes a multi-modal water quality sensor network, which covers key nodes such as the influent end, the aeration tank, and the secondary sedimentation tank of the sewage treatment system, and is used to synchronously collect physical, chemical, and biological water quality parameters.

[0073] Specifically, sensors are installed at the influent end, the aeration tank, and the secondary sedimentation tank of the sewage treatment system: the influent end is provided with a dissolved oxygen sensor, a pH / ORP composite electrode, and a multi-spectral turbidity meter, the pH / ORP composite electrode is composed of a glass electrode and a platinum electrode, and is installed at the opposite ends of the aeration tank (with a distance ≥ 5 m), contacts the water flow through a flow cell, and avoids bubble adhesion; the multi-spectral turbidity meter selects an LED light source with a wavelength range of 400-800 nm, is installed on the side wall of the aeration tank (0.5 m away from the bottom of the tank), and the light beam direction is perpendicular to the water flow direction; the aeration tank is provided with a dissolved oxygen sensor (installed along the longitudinal direction at 0.5 m, 1.0 m, and 1.5 m below the water surface to monitor the longitudinal gradient), and a sludge concentration sensor (MLSS); the secondary sedimentation tank is provided with a suspended solids concentration sensor (SS, laser scattering principle) and a total phosphorus sensor, the sensors are connected to an edge computing gateway through an RS485 bus or LoRa wireless communication, and the sampling frequency is set as needed, and a temperature sensor can also be installed in the sewage treatment system as needed.

[0074] S12: Extract the pH value, chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and suspended solids concentration from the distributed water quality sensor array to obtain the water quality parameter information.

[0075] In this embodiment, the sensor output is filtered and denoised by the edge computing gateway, and the core water quality indicators are extracted after data preprocessing such as outlier rejection: pH value, chemical oxygen demand (COD), ammonia nitrogen content (NH3-N), total phosphorus content (TP, unit: mg / L), suspended solids concentration (SS, unit: mg / L), and water temperature.

[0076] Specifically, the COD value is directly measured by a multi-parameter water quality analyzer (such as HACH DR3900) or converted based on TOC (total organic carbon) (COD = 2.66 x TOC + 0.2); the calculation of ammonia nitrogen content adopts the Nessler reagent spectrophotometric method, and the absorbance is measured at a wavelength of 420 nm by a cuvette; the calculation of total phosphorus content adopts the ammonium molybdate spectrophotometric method, and the absorbance at 660 nm wavelength is measured after acid digestion; the suspended solids are estimated by the weighing method (filtering and drying by a filter membrane) or the turbidity-SS empirical formula (SS = 100 x turbidity^0.85).

[0077] S13: According to the sewage water quality parameter information, the overall treatment target numerical information corresponding to the effluent standard of the sewage treatment plant is determined, and the overall treatment target numerical information includes COD removal rate target value, ammonia nitrogen removal rate target value, total phosphorus removal rate target value and suspended solids removal rate target value.

[0078] In this embodiment, the effluent standard refers to the pollutant discharge limit value of the sewage treatment plant specified by the state or the place (such as the “Urban Sewage Treatment Plant Pollutant Discharge Standard” GB 18918-2002 Level A standard); the removal rate target value is the minimum removal efficiency requirement of the sewage treatment system for each pollutant.

[0079] Specifically, according to the GB 18918-2002 Level A standard, the default removal rate target value is set: wherein, the COD removal rate is ≥90% (such as when the influent COD is ≤500 mg / L, the effluent is ≤50 mg / L); the ammonia nitrogen removal rate is ≥95% (such as when the influent NH3-N is ≤35 mg / L, the effluent is ≤1.5 mg / L); the total phosphorus removal rate is ≥85% (such as when the influent TP is ≤8 mg / L, the effluent is ≤1.0 mg / L); and the suspended solids removal rate is ≥90% (such as when the influent SS is ≤300 mg / L, the effluent is ≤30 mg / L).

[0080] Further, a dynamic adjustment mechanism for the removal rate target value is set: such as when the influent water quality suddenly changes (such as COD > 800 mg / L): the target value is adjusted to COD removal rate ≥85%, ammonia nitrogen removal rate ≥90% through manual intervention. Such as when the low load condition in the rainy season, the target value is reduced (such as COD removal rate ≥80%).

[0081] S2: Obtain the equipment operating status, and generate wastewater treatment process parameters based on the equipment operating status and overall treatment target numerical information. The treatment units in the wastewater treatment system are distributed in multiple areas.

[0082] In this embodiment, equipment operating status refers to the real-time operating parameters of each device in the wastewater treatment system, including the operating parameters of aeration equipment (such as blower current and air volume), dosing system (such as dosing pump frequency and dosage), and sludge return system (such as return pump speed and sludge concentration). Wastewater treatment process parameters refer to controllable variables that can directly affect the wastewater treatment effect, such as aeration intensity (mgO2 / L·h), dosing rate (kg / h), and return ratio (%). The wastewater treatment system comprises an aeration unit, a chemical dosing unit, and a sludge return unit. The operating status of each unit includes: the blower current (A), aeration head opening (°, feedback via electric actuator), and dissolved oxygen concentration (DO, mg / L) of the aeration unit; the metering pump stroke (mm, model MILTON ROY 330), dosing frequency (Hz, inverter output), and chemical concentration (%) of the chemical dosing unit; and the return pump speed (rpm), return sludge concentration (MLSS, g / L, sharing sensor data with step S1), and sludge age (SRT, d, calculated based on sludge production) of the sludge return unit.

[0083] Specifically, step S2 includes:

[0084] S21: Extract the underlying operating parameters based on the equipment's operating status. The underlying operating parameters include the aeration blower current, the return pump frequency, and the dosing pump stroke.

[0085] In this embodiment, parameters with different dimensions are normalized.

[0086] S22: Based on a fuzzy logic reasoning system, the overall processing target numerical information is transformed into process parameter constraints.

[0087] In this embodiment, the fuzzy logic reasoning system uses fuzzy set theory to transform fuzzy input linguistic variables (such as "high" and "low") into precise control outputs; the process parameter constraints are based on effluent standards and equipment performance limitations (such as DO≥2 mg / L, MLSS≤4000 mg / L).

[0088] For example, the input variables are fuzzified, where the input variables are such as the target COD removal rate ( ,%), real-time dissolved oxygen concentration (%) (mg / L) and sludge concentration ( (mg / L), set the corresponding membership function: where Concentrations of 85% or less are considered low; concentrations greater than 85% but less than or equal to 90% are considered medium; and concentrations exceeding 90% are considered high. Concentrations less than 1 mg / L are considered low concentrations, concentrations greater than or equal to 1 mg / L and less than or equal to 3 mg / L are considered medium concentrations, and concentrations exceeding 3 mg / L are considered high concentrations. A concentration less than 1000 mg / L is considered low concentration; a concentration greater than or equal to 1000 mg / L and less than or equal to 4000 mg / L is considered medium concentration; and a concentration exceeding 4000 mg / L is considered high concentration.

[0089] Configure the fuzzy rule base for the fuzzy logic reasoning system, for example:

[0090] IF is low AND The aeration intensity is reduced by 20% (weight 0.8) when the aeration intensity is high.

[0091] IF is high AND The reflux ratio in THEN is increased by 15% (weight 0.7).

[0092] IF The nitrification liquor reflux ratio is increased to 300% (mandatory rule).

[0093] The maximum-membership method is used to select the rule with the highest priority, and the centroid method is used to calculate the actual value of the fuzzy output. ,in, Let x be the membership function, the integration interval be the physical range of the device parameters, and x be the device parameter index.

[0094] S23: Solve the optimal solution set of aeration intensity gradient distribution and dosing rate spatiotemporal configuration through a multi-objective optimization algorithm to generate wastewater treatment process parameters including the longitudinal dissolved oxygen concentration gradient of the aeration tank and the lateral difference in reagent dosing concentration.

[0095] In this embodiment, the multi-objective optimization algorithm can simultaneously optimize multiple conflicting objective functions (such as minimizing energy consumption and maximizing removal rate); the aeration intensity gradient distribution refers to the dissolved oxygen concentration distribution along the water depth direction in the aeration tank (such as longitudinal 0.5-2.5 mg / L / m); the spatiotemporal configuration of the dosing rate refers to the spatiotemporal difference in the amount of reagent added in different areas (such as lateral concentration difference ≤5%).

[0096] Specifically, an objective function based on minimizing total energy consumption is defined. and the objective function to maximize pollutants ,in P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example

[0097] Further, NSGA-II (non-dominated sorting genetic algorithm) is used, wherein the population size is 200, the iteration number is 300, the crossover probability is 0.9, the aeration intensity (mgO2 / L·h), the dosing rate (kg / h) and the reflux ratio (%) are directly used as gene fragments for encoding calculation, when multi-objective optimization is performed, 100 groups of process parameter combinations can be randomly generated, the individual fitness is calculated according to the objective function and the constraint condition, and 50 groups of non-dominated solutions in the Pareto front solution set are screened out.

[0098] The longitudinal distribution of the aerobic zone with gradient distribution of aeration intensity is P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example The control formula is as follows P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example P is the power of each device, unit: kW; for example

[0099] S3: According to the process parameter and the overall treatment target numerical information, a process control instruction is triggered.

[0100] In the embodiment, the process control instruction includes a structured instruction of device type, control parameter, execution time, target area (such as an aerobic zone), aeration intensity, execution time and the like information, which is used to drive the field device action of the sewage treatment system, is in a JSON format, and is distinguished by an associated instruction ID.

[0101] Specifically, the sewage treatment process parameters and the overall treatment target values are used to determine the equipment that needs to be adjusted, such as if the COD removal rate does not reach the target (measured 85% < target 90%) and the MLSS > 3500 mg / L, the "enhanced biodegradation" instruction is triggered (increase aeration + extend sludge age). For example, if the NH3-N removal rate meets the standard (96%) but the TP exceeds the standard (1.2 mg / L > target 1.0 mg / L), the "chemical phosphorus removal" instruction is triggered (add polyaluminum chloride PAC, dosage = 50 mg / L x inflow).

[0102] S4: Obtain the partition processing effect detection information, compare the partition processing effect detection information with the corresponding index values in the overall treatment target value information, and obtain the partition difference value.

[0103] In this embodiment, the partition processing effect detection information refers to the real-time effluent quality data of each independent sub-region (such as anaerobic zone, anoxic zone, and aerobic zone) in the sewage treatment system. The partition detection network is constructed, and the sewage treatment system is divided into anaerobic zone (influent end to the end of anaerobic reaction tank), anoxic zone (end of anaerobic zone to end of anoxic reaction tank), and aerobic zone (end of anoxic zone to end of aerobic reaction tank). The anaerobic zone can be further divided into anaerobic zone 1 and anaerobic zone 2 according to the longitudinal or transverse direction. The present application divides the longitudinal dissolved oxygen concentration gradient along the longitudinal direction when analyzing the aeration unit, and divides the transverse direction when analyzing the dosing unit, so as to facilitate fine monitoring and analysis of each sewage treatment region. The partition difference value refers to the deviation degree of the measured water quality index and the target value (such as COD difference value = (measured COD - target COD) / target COD x 100%).

[0104] In this embodiment, step S4 specifically includes:

[0105] S41: Obtain the COD concentration, ammonia nitrogen concentration, total phosphorus concentration, and suspended solids concentration detection values of the effluent of each partition.

[0106] Specifically, independent sampling points are set in each region of the anaerobic zone, anoxic zone, and aerobic zone, for example, 1 point in the anaerobic zone, 2 symmetric points in the anoxic zone, and 3 longitudinal gradient points in the aerobic zone. The number of independent sampling points can be configured according to the actual size of the sewage treatment system.

[0107] S42: Calculate the difference between the water quality index of the effluent of each partition and the corresponding overall treatment target value, and obtain the COD difference value, ammonia nitrogen difference value, total phosphorus difference value, and suspended solids difference value.

[0108] Specifically, the target value is dynamically generated according to the water quality and discharge standard, for example, when the influent COD is less than or equal to 500 mg / L, the effluent COD is less than or equal to 50 mg / L; if the influent COD is greater than 500 mg / L, the effluent COD is less than or equal to 10% of the influent value. When the influent NH3-N is less than or equal to 35 mg / L, the effluent NH3-N is less than or equal to 1.5 mg / L; if the influent NH3-N is greater than 35 mg / L, the effluent NH3-N is less than or equal to 2 mg / L.

[0109] If the measured COD of a certain partition is 60 mg / L and the target is 50 mg / L, then ΔCOD = (60-50) / 50 x 100% = +20%, which increases by 20%; if the measured NH3-N of a certain partition is 1.0 mg / L and the target is 1.5 mg / L, then ΔNH3-N = (1.0-1.5) / 1.5 x 100% = -33.3%, which decreases by 33.3%.

[0110] S43: Weighted fusion of COD difference value, ammonia nitrogen difference value, total phosphorus difference value and suspended solids difference value to obtain a comprehensive partition difference value.

[0111] Specifically, the weighted fusion refers to assigning weights according to the importance of each water quality index, and comprehensively calculating a single difference value; the comprehensive partition difference value is used to quantify the overall deviation degree of the partition processing effect and the target (dimensionless).

[0112] For example, the weight distribution is as follows: based on the weight coefficient of the index on the environmental hazard and the processing difficulty (example):

[0113] The weight of COD is 0.4 (high economic cost, easy to exceed the standard); the weight of NH3-N is 0.3 (high toxicity, needs to be controlled first); the weight of total phosphorus is 0.2 (easy to cause eutrophication); the weight of suspended solids is 0.1 (affects the sensory index).

[0114] The comprehensive difference value calculation formula is as follows: wherein, is the weight of the i-th index; i is the index index; is the normalized difference value after standardization, and the standardization processing adopts a min-max standardization processing method.

[0115] S5: Range comparison of the partition difference value and the preset change threshold value, if the partition difference value exceeds the preset change threshold value range, a partition adaptive control update instruction is triggered.

[0116] In this embodiment, the change threshold value range is the allowed partition difference value fluctuation interval; the partition adaptive control update instruction is a parameter adjustment instruction (such as increasing the aeration amount, adjusting the reflux ratio) triggered when the difference value exceeds the threshold value.

[0117] Specifically, step S5 includes:

[0118] S51: Obtain preset dynamic change thresholds of each index, and the dynamic change thresholds include a COD allowed fluctuation range, an ammonia nitrogen allowed fluctuation range, a total phosphorus allowed fluctuation range, and a suspended solids allowed fluctuation range.

[0119] Specifically, the dynamic change threshold is an upper and lower limit allowing each water quality index to fluctuate within a set range, which is dynamically adjusted according to process stability requirements and water quality fluctuation characteristics. For example, the COD allowed fluctuation range is [45 mg / L, 55 mg / L] (corresponding to a target value of 50 mg / L, with an allowed fluctuation of ±10%). The ammonia nitrogen allowed fluctuation range is [1.2 mg / L, 1.8 mg / L] (corresponding to a target value of 1.5 mg / L, with an allowed fluctuation of ±20%). The total phosphorus allowed fluctuation range is [0.8 mg / L, 1.2 mg / L] (corresponding to a target value of 1.0 mg / L, with an allowed fluctuation of ±20%). The suspended solids allowed fluctuation range is [25 mg / L, 35 mg / L] (corresponding to a target value of 30 mg / L, with an allowed fluctuation of ±16.7%).

[0120] Further, the dynamic change threshold can be based on analyzing the fluctuation range of each index in the past 30 days, taking the 95% confidence interval as the benchmark threshold, and then adjusting the threshold according to the seasonal changes of the influent water quality (such as the rainy season / dry season) (for example, the COD threshold is relaxed to ±15% in the rainy season).

[0121] S52: Range comparison is performed between the comprehensive zoning difference value and the dynamic change thresholds of each index. If the comprehensive zoning difference value exceeds the allowed fluctuation range of any index, it is determined that adaptive regulation is needed, and a zoning adaptive regulation update instruction is triggered.

[0122] Specifically, it is determined whether the comprehensive difference value exceeds the threshold of each index (such as COD, ammonia nitrogen, total phosphorus, and SS respectively). If the comprehensive zoning difference value exceeds the allowed fluctuation range of any index, it is determined that adaptive regulation is triggered. Each index can also set a forced triggering condition, such as immediately triggering emergency regulation if any index exceeds the safety threshold (such as ammonia nitrogen > 2.0 mg / L).

[0123] Further, the zoning adaptive regulation update instruction also includes multiple hierarchical types, such as level I regulation, adjusting a single device parameter (such as aeration intensity ±5%); level II regulation, adjusting multiple devices (such as aeration intensity increasing by 10% and reflux ratio increasing by 5%); and level III regulation, starting emergency measures (such as adding carbon source, switching to a different type of reagent).

[0124] S53: According to the zoning adaptive regulation update instruction, obtain the device operating parameters of the current zone and the water quality change trend information in the recent period.

[0125] In this embodiment, the device operating parameter refers to a real-time parameter that directly affects the treatment effect; the near-term water quality change trend information refers to the change direction and rate of the water quality index in the near term (e.g., the COD concentration increases by 0.5 mg / L / min in the past 1 hour). The sliding window method (window size = 30 minutes) is used to fit the water quality change curve, predict the index fluctuation in the next 10 minutes, and identify water quality mutations (e.g., the ammonia nitrogen concentration increases by 20% in 5 minutes) through the LSTM neural network to detect abnormal water quality changes in a timely manner.

[0126] S54: Based on the current device operating parameter, water quality change trend information, and pre-set adaptive adjustment rules, the partition adjustment parameter information is calculated and generated through a machine learning algorithm, including aeration intensity adjustment value, reagent dosage adjustment value, and sludge return flow adjustment value.

[0127] In this embodiment, the adaptive adjustment rule is a pre-set rule library based on historical control experience (“aeration intensity is increased by 10% when DO < 2 mg / L”); the machine learning algorithm uses the reinforcement learning PPO algorithm to optimize the control strategy through trial and error.

[0128] Specifically, the historical control records (including input parameters, control actions, and effect feedback) are obtained, and based on the optimization objectives of minimum energy consumption and maximum removal rate, and the constraint conditions of DO ≥ 2 mg / L, MLSS ≤ 4000 mg / L, and sludge age ≥ 5 days, the training optimization is performed, and the model robustness is tested through cross-validation (K-fold = 5) to ensure that the prediction error is < 5%. Among them, the partition adjustment parameter information such as “aeration intensity adjustment value”: “+8%”; “reagent pump frequency adjustment value”: “-3%”; “return ratio adjustment value”: “+2%”; “execution priority”: “level II”.

[0129] S6: According to the partition adaptive control update instruction, the partition adjustment parameter information is obtained.

[0130] In this embodiment, the partition adjustment parameter information is the specific adjustment amount (e.g., aeration intensity is increased by 10%, reagent pump stroke is increased by 5%) for a specific partition (e.g., aerobic zone 1#).

[0131] Specifically, after receiving the partition adaptive regulation update instruction, the target indicators to be regulated (such as COD, ammonia nitrogen) are identified, the execution priority of the regulation instruction is determined (such as safety class instruction is prior to energy saving class instruction), and the regulation range is defined (such as defining that the adjustment amplitude of aeration intensity is not more than ±10%). The latest device operation parameters of the current partition are obtained, including aeration equipment (such as fan current, air volume, aeration intensity (DO concentration)), dosing system (such as dosing pump frequency, dosing amount), sludge return system (such as return pump speed, sludge concentration), wherein the PLC of the sewage treatment system directly reads the device operation state (such as frequency converter frequency, valve opening), and the sensor real-time monitors the water quality parameters (such as dissolved oxygen, ammonia nitrogen concentration).

[0132] Then, the aeration intensity adjustment value, the dosing amount adjustment value and the return ratio adjustment value are calculated based on the machine learning model of reinforcement learning:

[0133] The aeration intensity adjustment value is , wherein, is a DO adjustment coefficient; is a target dissolved oxygen concentration; is an actual dissolved oxygen concentration; and the DO change rate is the DO concentration change rate in a specified time period (such as in the last one hour).

[0134] The dosing amount adjustment value is , wherein, , is a dosing agent adding coefficient, which is obtained by regression model training.

[0135] The return ratio adjustment value is , wherein, is the return sludge amount; is the influent amount; is a coefficient dynamically adjusted according to the MLSS concentration, used for correcting the influence of the return ratio on the sludge age; When >1, the MLSS concentration is lower than the target value, and the return ratio needs to be increased to prolong the sludge age; When <1, the MLSS concentration is higher than the target value, and the return ratio needs to be reduced to shorten the sludge age. When MLSS>4000 mg / L, the value is too high, and the risk of sludge aging increases, is automatically adjusted to be low (such as is negative, the return sludge amount is reduced, otherwise, such as MLSS<2000 mg / L, the treatment capacity is reduced, is automatically adjusted to be high (such as =1.2).

[0136] S7: Trigger the adjustment instruction according to the partition adjustment parameter information.

[0137] In this embodiment, the adjustment instruction contains the control signal of device address, control parameter (such as frequency, opening degree), execution time and other information.

[0138] Specifically, the adjustment parameter is converted into a device control instruction (taking an aeration fan as an example):

[0139] The aeration intensity adjustment value is increased by 10% (the current value is 2.0 mgO2 / L·h, and the target value is 2.2 mgO2 / L·h), and the aeration fan frequency is increased from 40 Hz to 44 Hz (the frequency is linearly related to the aeration intensity: O2∝ frequency).

[0140] The instruction priority queue is generated:

[0141] High priority: safety-related instructions (such as forced aeration when DO < 1 mg / L).

[0142] Medium priority: process optimization instructions (such as adjusting the reflux ratio when ammonia nitrogen exceeds the standard).

[0143] Low priority: energy-saving instructions (such as reducing aeration during low load).

[0144] The instructions are issued to the field PLC (Programmable Logic Controller) through the Modbus TCP protocol, the adjustment instructions are executed, and the effects after adjustment are monitored in real time:

[0145] For example, the water quality data (COD, NH3-N, etc.) collected 30 minutes after adjustment is calculated to obtain a new partition difference value Δintegrated. If Δintegrated decreases by ≥80% (such as from -25% to -5%), it is recorded as an effective control case, and the control rule library is updated. If Δintegrated still exceeds the threshold value (such as from -25% to -20%), the expert system diagnosis is started (to check sensor faults, equipment blockage, etc.).

[0146] In an embodiment, as shown in FIG. 1, in step S1, according to the sewage quality parameter information, the overall treatment target value information is obtained, which further includes: Figure 2

[0147] S101: According to the influent quantity, influent quality index and aeration tank volume of the sewage treatment system, the first oxygen demand of the aeration tank is calculated through the ASM2D model; according to the sludge concentration, reflux ratio and oxygen demand coefficient of the sewage treatment system, the second oxygen demand of the sludge reflux system is calculated.

[0148] In this embodiment, the ASM2D model is a mathematical model of activated sludge system, which is used to simulate the biological denitrification and phosphorus removal process, including biochemical reactions such as carbon oxidation, nitrification, denitrification and phosphorus absorption; the first oxygen demand refers to the oxygen demand for the degradation of organic matter and ammonia nitrogen by microorganisms in the aeration tank; and the second oxygen demand refers to the oxygen consumption for endogenous respiration in the sludge reflux system. ​

[0149] Specifically, the first oxygen demand The calculation formula includes:

[0150] Wherein, is the influent amount; is the influent BOD5 concentration; is the oxygen equivalent coefficient related to nitrogen oxidation; is the influent ammonia nitrogen concentration; is the effluent nitrate nitrogen concentration; is the volatile suspended solid concentration; is the endogenous respiration coefficient; is the attenuation coefficient of aerobic microorganisms; is the sludge yield coefficient; is the endogenous respiration rate constant.

[0151] Specifically, the second oxygen demand The calculation formula includes:

[0152] Wherein, is the residual sludge discharge amount; is the residual sludge MLSS concentration; is the sludge return ratio; is the return sludge MLSS concentration; is the oxygen demand calculation correction coefficient.

[0153] S102: Combine the first oxygen demand and the second oxygen demand to generate the total oxygen demand and the aeration gap amount.

[0154] In this embodiment, the total oxygen demand includes the first oxygen demand and the second oxygen demand, and the sum of the first oxygen demand and the second oxygen demand is used in calculation; the aeration gap amount refers to the difference between the total oxygen demand and the actual aeration amount.

[0155] S103: Obtain the theoretical aeration amount according to the aeration gap amount and the aeration efficiency curve; calculate the ratio of the first oxygen demand to the second oxygen demand to generate the oxygen demand distribution ratio.

[0156] Specifically, the aeration efficiency curve refers to the corresponding relationship curve of the aeration amount (kgO2 / h) and the dissolved oxygen concentration (mg / L), which is fitted through experiments or empirical data. The oxygen demand distribution ratio is the distribution ratio of the oxygen demand along the longitudinal direction (water depth direction) of the aeration tank (such as 30% at the bottom, 50% in the middle, and 20% at the top).

[0157] Specifically, according to the aeration efficiency curve, the theoretical aeration amount required to meet the aeration gap amount is found. The oxygen demand distribution ratio can be determined by ASM2D model simulation or historical data analysis to determine the oxygen demand ratio of different regions of the aeration tank, for example, the longitudinal distribution of the aerobic zone: the bottom (0.5 m) accounts for 30% of the oxygen demand, the middle (1.5 m) accounts for 50%, and the top (2.5 m) accounts for 20%.

[0158] S104: Obtain the aeration efficiency correction coefficient based on the historical difference between the theoretical aeration amount and the actual aeration amount.

[0159] Specifically, the aeration efficiency correction coefficient is a coefficient for correcting the deviation between the theoretical aeration amount and the actual aeration amount, which is dynamically adjusted based on historical data.

[0160] For example, the difference between the theoretical aeration amount ( ) and the actual aeration amount ( ) in the past 30 days is calculated, and the aeration efficiency correction coefficient = , where is the date number identifier, is the theoretical aeration amount (kgO2 / d) on the th day; is the actual aeration amount (kgO2 / d) on the th day, for example, if the cumulative deviation is +10%, the correction coefficient = 1.1.

[0161] S105: Generate the aeration amount-dissolved oxygen response relationship according to the oxygen demand distribution ratio and the aeration efficiency correction coefficient.

[0162] In this embodiment, the aeration amount-dissolved oxygen response relationship is a mathematical model (such as a dynamic response curve) for describing the dynamic relationship between aeration amount and dissolved oxygen concentration.

[0163] Specifically, the DO data under different aeration amounts are collected by an online dissolved oxygen sensor, and a regression analysis is used to fit the response curve, for example , where A and B are fitting coefficients.

[0164] S106: Generate the aeration control table according to the aeration amount-dissolved oxygen response relationship and the aeration gap amount.

[0165] In the present embodiment, the aeration regulation control table includes the aeration gap amount, the theoretical required aeration amount, the corrected aeration amount, and the dissolved oxygen target value; according to the results of steps S103-S105, the control parameters corresponding to different aeration gap amounts are filled in advance, for example, if the aeration gap amount = 100 kgO2 / d, the corrected aeration amount = 275 kgO2 / d, and the dissolved oxygen target value = 2.0 mg / L. In actual application, the control table needs to be updated regularly (such as every week) according to the latest operation data to adapt to the complex water quality and working condition changes of the sewage treatment system.

[0166] It should be understood that the sequence numbers of the steps in the above embodiments do not mean the order of execution, and the execution order of the processes should be determined according to their functions and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0167] In an embodiment, a sewage treatment intelligent regulation system is provided, which corresponds to the sewage treatment intelligent regulation method in the above embodiments.

[0168] The sewage treatment intelligent regulation system comprises a data acquisition and processing module, a process parameter generation module, a regulation instruction generation module, a partition detection and evaluation module, and an execution mechanism regulation module. The detailed description of each functional module is as follows:

[0169] The data acquisition and processing module is used to construct a multi-modal water quality sensing network, acquire sewage water quality parameter information, and acquire overall treatment target value information according to the sewage water quality parameter information;

[0170] The process parameter generation module is used to acquire equipment operating states, generate sewage treatment process parameters according to the equipment operating states and the overall treatment target value information, wherein the treatment units in the sewage treatment system are distributed in multiple regions;

[0171] The regulation instruction generation module is used to trigger process regulation instructions according to the sewage treatment process parameters and the overall treatment target value information;

[0172] The partition detection and evaluation module is used to acquire partition treatment effect detection information, compare the partition treatment effect detection information with the corresponding index values in the overall treatment target value information, and obtain partition difference values;

[0173] The execution mechanism regulation module is used to compare the partition difference values with a preset change threshold range, trigger a partition adaptive regulation update instruction if the partition difference values exceed the preset change threshold range, acquire partition adjustment parameter information according to the partition adaptive regulation update instruction, and trigger adjustment instructions according to the partition adjustment parameter information.

[0174] Optionally, the data acquisition and processing module comprises:

[0175] a first oxygen demand calculation unit configured to calculate a first oxygen demand of the aeration tank according to an influent quantity, an influent quality index of the wastewater treatment system and a volume of the aeration tank by means of the ASM2D model;

[0176] a second oxygen demand calculation unit configured to calculate a second oxygen demand of the sludge return system according to a sludge concentration, a return ratio and an oxygen demand calculation coefficient of the wastewater treatment system;

[0177] a total oxygen demand and aeration gap calculation submodule configured to combine the first oxygen demand and the second oxygen demand to generate a total oxygen demand and an aeration gap quantity;

[0178] a distribution ratio generation submodule configured to obtain a theoretical aeration quantity according to the aeration gap quantity and an aeration efficiency curve, and generate an oxygen demand distribution ratio according to a ratio of the first oxygen demand and the second oxygen demand;

[0179] an efficiency correction coefficient obtaining submodule configured to obtain an aeration efficiency correction coefficient according to a historical difference between the theoretical aeration quantity and an actual aeration quantity;

[0180] a response relationship generation submodule configured to generate an aeration quantity-dissolved oxygen response relationship according to the oxygen demand distribution ratio and the aeration efficiency correction coefficient;

[0181] an aeration regulation control table generation submodule configured to generate an aeration regulation control table according to the aeration quantity-dissolved oxygen response relationship and the aeration gap quantity.

[0182] The specific limitations of the wastewater treatment intelligent regulation system can refer to the limitations of the wastewater treatment intelligent regulation method described above, which will not be repeated here; each module in the wastewater treatment intelligent regulation system described above can be realized by software, hardware and combinations thereof; each module described above can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0183] In one embodiment, in particular, the process described above with reference to the flowchart can be implemented as a computer software program according to embodiments of the present application. For example, embodiments of the present application include a computer program product including computer programs / instructions that, when executed by a processor, implement the steps of the wastewater treatment intelligent regulation method as described. In such embodiments, the computer program can be downloaded and installed from a network by a communication module, and / or installed from a detachable medium. When the computer program is executed by a central processing unit (CPU), various functions defined in the present application are performed.

[0184] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0185] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand; it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A smart control method for wastewater treatment, characterized in that, include: A multimodal water quality sensing network is constructed to obtain wastewater quality parameter information, and the overall treatment target value information is obtained based on the wastewater quality parameter information. The equipment operating status is obtained, and wastewater treatment process parameters are generated based on the equipment operating status and the overall treatment target value information, wherein the treatment units in the wastewater treatment system are distributed in multiple areas; Based on the wastewater treatment process parameters and the overall treatment target value information, a process control command is triggered. Obtain the partition processing effect detection information, compare the partition processing effect detection information with the corresponding index value in the overall processing target numerical information, and obtain the partition difference value; The partition difference value is compared with a preset change threshold. If the partition difference value exceeds the preset change threshold range, a partition adaptive adjustment update command is triggered. According to the partition adaptive adjustment update instruction, obtain the partition adjustment parameter information; Based on the partition adjustment parameter information, an adjustment command is triggered. The step of obtaining partition processing effect detection information, comparing the partition processing effect detection information with the corresponding indicator value in the overall processing target numerical information to obtain the partition difference value, specifically includes: Obtain the measured values ​​of COD concentration, ammonia nitrogen concentration, total phosphorus concentration, and suspended solids concentration in the effluent from each zone; Calculate the difference between the effluent water quality indicators of each zone and the corresponding overall treatment target values ​​to obtain the COD difference value, ammonia nitrogen difference value, total phosphorus difference value and suspended solids difference value; The COD difference value, ammonia nitrogen difference value, total phosphorus difference value and suspended solids difference value are weighted and fused to obtain the comprehensive zoning difference value; The step involves comparing the partition difference value with a preset change threshold. If the partition difference value exceeds the preset change threshold range, a partition adaptive adjustment update command is triggered. Based on the partition adaptive adjustment update command, partition adjustment parameter information is obtained, specifically including: Obtain preset dynamic change thresholds for each indicator, including allowable fluctuation ranges for COD, ammonia nitrogen, total phosphorus, and suspended solids; The comprehensive partition difference value is compared with the dynamic change threshold of each indicator. If the comprehensive partition difference value exceeds the allowable fluctuation range of any indicator, it is determined that adaptive adjustment is required, and the partition adaptive adjustment update instruction is triggered. According to the partition adaptive control update instruction, obtain the current partition's equipment operating parameters and recent water quality change trend information; Based on the current equipment operating parameters, the water quality change trend information, and the preset adaptive adjustment rules, the zonal adjustment parameter information is calculated and generated through machine learning algorithms. The zonal adjustment parameter information includes the aeration intensity adjustment value, the chemical dosage adjustment value, and the sludge return flow adjustment value.

2. The intelligent control method for wastewater treatment according to claim 1, characterized in that, The construction of a multimodal water quality sensing network to acquire wastewater quality parameter information, and the acquisition of overall treatment target numerical information based on the wastewater quality parameter information, specifically includes: Deploy a distributed water quality sensor array, which includes a dissolved oxygen sensor, a pH / ORP composite electrode, and a multispectral turbidimeter; The pH value, chemical oxygen demand, ammonia nitrogen content, total phosphorus content, and suspended solids concentration are extracted from the distributed water quality sensor array to obtain wastewater quality parameter information; Based on the wastewater quality parameters, the overall treatment target values ​​corresponding to the effluent standards of the wastewater treatment plant are determined. The overall treatment target values ​​include the target values ​​for COD removal rate, ammonia nitrogen removal rate, total phosphorus removal rate, and suspended solids removal rate.

3. The intelligent control method for wastewater treatment according to claim 2, characterized in that, The step of generating wastewater treatment process parameters based on the equipment operating status and the overall treatment target value information specifically includes: Based on the operating status of the equipment, the underlying operating parameters are extracted, including the aeration blower current, the return pump frequency, and the dosing pump stroke. Based on the fuzzy logic reasoning system, the overall processing target numerical information is transformed into process parameter constraints. The optimal solution set for the aeration intensity gradient distribution and the spatiotemporal configuration of the dosing rate is obtained by using a multi-objective optimization algorithm, generating wastewater treatment process parameters that include the longitudinal dissolved oxygen concentration gradient of the aeration tank and the lateral difference in the dosage of the reagents.

4. The intelligent control method for wastewater treatment according to claim 1, characterized in that, The step of obtaining the overall treatment target numerical information based on the wastewater quality parameters further includes: Based on the influent flow rate, influent water quality indicators, and aeration tank volume of the wastewater treatment system, the first oxygen demand of the aeration tank is calculated using the ASM2D model; based on the sludge concentration, return ratio, and oxygen demand calculation coefficient of the wastewater treatment system, the second oxygen demand of the sludge return system is calculated. Combining the first oxygen demand and the second oxygen demand, the total oxygen demand and the aeration deficit are generated; Based on the aeration deficit and aeration efficiency curve, the theoretical aeration requirement is obtained; the ratio of the first oxygen demand to the second oxygen demand is calculated to generate the oxygen demand distribution ratio. Based on the historical difference between the theoretical required aeration volume and the actual aeration volume, an aeration efficiency correction coefficient is obtained. Based on the oxygen demand distribution ratio and the aeration efficiency correction coefficient, an aeration rate-dissolved oxygen response relationship is generated. An aeration control comparison table is generated based on the aeration volume-dissolved oxygen response relationship and the aeration deficit.

5. The intelligent control method for wastewater treatment according to claim 4, characterized in that, The first oxygen demand Q aer Calculation formula include: Q aer =Q×S0×(K o,N ×N0-K o,N ×N e )+Q×S V ×(f d ×b H ×Y×N0-k d ×N e ) Q represents the influent flow rate; S0 represents the influent BOD5 concentration; K represents the influent concentration. o,N The oxygen equivalent coefficient related to nitrogen oxidation; N0 is the influent ammonia nitrogen concentration; N e The concentration of nitrate nitrogen in the effluent; S V f represents the concentration of volatile suspended solids. d b is the endogenous respiration coefficient; H Y is the aerobic microbial attenuation coefficient; K is the sludge yield coefficient; d is the intrinsic respiration rate constant.

6. The intelligent control method for wastewater treatment according to claim 4, characterized in that, The second oxygen demand O end The calculation formulas include: O end =(Q w ×MLSS w ×R+Q×MLSS r )×α Among them, Q w For excess sludge discharge; MLSS w MLSS concentration of excess sludge; Q is the influent flow rate; R is the sludge return ratio; MLSS r α represents the MLSS concentration of the returned sludge; α is the correction factor for calculating oxygen demand.

7. A wastewater treatment intelligent control system, characterized in that, The system includes: The data acquisition and processing module is used to construct a multimodal water quality sensing network, acquire wastewater quality parameter information, and acquire overall treatment target numerical information based on the wastewater quality parameter information. The process parameter generation module is used to obtain the equipment operating status and generate wastewater treatment process parameters based on the equipment operating status and the overall treatment target value information, wherein the treatment units in the wastewater treatment system are distributed in multiple areas. The control instruction generation module is used to trigger process control instructions based on the wastewater treatment process parameters and the overall treatment target value information; The partition detection and evaluation module is used to acquire partition processing effect detection information, compare the partition processing effect detection information with the corresponding index value in the overall processing target numerical information, and obtain the partition difference value. The actuator control module is used to compare the partition difference value with a preset change threshold range. If the partition difference value exceeds the preset change threshold range, a partition adaptive control update command is triggered. Based on the partition adaptive control update command, partition adjustment parameter information is obtained. Based on the partition adjustment parameter information, an adjustment command is triggered. The step of obtaining partition processing effect detection information, comparing the partition processing effect detection information with the corresponding indicator value in the overall processing target numerical information to obtain the partition difference value, specifically includes: Obtain the measured values ​​of COD concentration, ammonia nitrogen concentration, total phosphorus concentration, and suspended solids concentration in the effluent from each zone; Calculate the difference between the effluent water quality indicators of each zone and the corresponding overall treatment target values ​​to obtain the COD difference value, ammonia nitrogen difference value, total phosphorus difference value and suspended solids difference value; The COD difference value, ammonia nitrogen difference value, total phosphorus difference value and suspended solids difference value are weighted and fused to obtain the comprehensive zoning difference value; The step involves comparing the partition difference value with a preset change threshold. If the partition difference value exceeds the preset change threshold range, a partition adaptive adjustment update command is triggered. Based on the partition adaptive adjustment update command, partition adjustment parameter information is obtained, specifically including: Obtain preset dynamic change thresholds for each indicator, including allowable fluctuation ranges for COD, ammonia nitrogen, total phosphorus, and suspended solids; The comprehensive partition difference value is compared with the dynamic change threshold of each indicator. If the comprehensive partition difference value exceeds the allowable fluctuation range of any indicator, it is determined that adaptive adjustment is required, and the partition adaptive adjustment update instruction is triggered. According to the partition adaptive control update instruction, obtain the current partition's equipment operating parameters and recent water quality change trend information; Based on the current equipment operating parameters, the water quality change trend information, and the preset adaptive adjustment rules, the zonal adjustment parameter information is calculated and generated through machine learning algorithms. The zonal adjustment parameter information includes the aeration intensity adjustment value, the chemical dosage adjustment value, and the sludge return flow adjustment value.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the intelligent control method for wastewater treatment as described in any one of claims 1 to 6.

Images