A system for purifying waste water from cattle and sheep slaughtering

Abstract

The present application provides a kind of cattle and sheep slaughter waste water purification treatment system, including the following function modules: monitoring module is used to collect process node multi-source working condition parameters;Pollution equivalent calculation module, construct water quality pollution equivalent calculation model, output pollution equivalent calculation value;System resilience assessment module, dissolved oxygen and sludge concentration data are collected in biochemical treatment unit, calculate system resilience coefficient and dynamically generate early warning threshold;Risk early warning control module, based on pollution equivalent calculation value, calculate system risk index, compare with dynamic early warning threshold, trigger hierarchical early warning and execute corresponding control strategy, the present application realizes key index pre-judgment by pollution equivalent early calculation, generates early warning threshold using resilience coefficient to adapt to working condition fluctuation, and triggers differentiated regulation according to risk classification, realize the change from passive alarm to active early warning, improve system impact resistance and operating stability.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment technology, specifically to a wastewater purification and treatment system for beef and mutton slaughtering. Background Technology

[0002] Wastewater generated during the slaughtering and processing of beef and mutton is characterized by high organic matter concentration, high suspended solids content, high oil content, and drastic fluctuations in water quality and quantity. Among these, chemical oxygen demand (COD), suspended solids concentration, and oil concentration are key indicators of pollution load. If not properly treated, high-concentration organic wastewater discharged into natural water bodies will rapidly deplete dissolved oxygen, leading to black and odorous water and an imbalance in aquatic ecosystems. Oily substances tend to adhere and caking on the surfaces of pipe networks and equipment, reducing treatment efficiency and increasing operating energy consumption. With increasingly stringent national water pollutant discharge standards and continuously strengthened ecological protection requirements in key areas such as the Yellow River Basin, slaughterhouse wastewater treatment facilities should not only aim to meet end-of-pipe discharge limits but should also strengthen the ability to control the entire process and predict risks.

[0003] In the prior art, CN118754356A discloses a reuse control method and system for treating wastewater from biological products. This technology detects the wastewater source type of the wastewater to be reused; inputs the wastewater to be reused into a multi-stage purification device, and sets a water quality detection sensor at the outlet of a preset purification device; sets graded water quality, and detects whether the treated wastewater at the outlet of the corresponding purification device meets the water quality of the preset water type; and reuses the treated wastewater output from the downstream purification device to the upstream purification device for further treatment after passing through the corresponding water channel.

[0004] However, the existing technologies mentioned above rely on real-time online monitoring and fixed threshold alarm mechanisms. The monitoring nodes are scattered, the data utilization is low, and they can only passively respond to events that have already exceeded standards, lacking the ability to predict future water quality fluctuations. The monitoring and control of key pollutants such as chemical oxygen demand, suspended solids concentration, and oil concentration are fragmented, making it difficult to form a systematic risk assessment. The dissolved oxygen and sludge concentration control of the biochemical treatment unit adopt independent closed-loop regulation, without establishing a correlation between their deviation and the overall shock resistance capacity of the system, resulting in frequent false alarms or missed alarms of fixed warning thresholds under operating conditions. In addition, the existing technologies lack graded early warning and differentiated control strategies, and cannot automatically match and adapt control or emergency intervention instructions according to the risk level. Often, disposal measures are only initiated after the exceedance occurs, resulting in a slow system response, insufficient shock load resistance, and high levels of operating energy consumption and risk of exceeding emission standards.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a wastewater purification and treatment system for beef and mutton slaughtering to solve the problems mentioned in the background art. This invention constructs a water pollution equivalent estimation model, enabling time-series prediction of key pollutants such as chemical oxygen demand (COD), suspended solids concentration, and oil concentration, providing reliable data support for early warning. A system resilience assessment module calculates the system resilience coefficient, characterized by the deviation between dissolved oxygen and sludge concentration, in real time, and dynamically generates an adaptive early warning threshold based on this coefficient. This solves the problem that traditional fixed thresholds are difficult to adapt to fluctuations in operating conditions, improving the flexibility and accuracy of the early warning mechanism. A risk early warning and control module calculates the system risk index based on the pollution equivalent estimation value and compares it with the dynamic early warning threshold in a graded manner. This allows the system to identify whether it is in a stable operation period, a load adaptation period, or a shock overload period, and automatically triggers differentiated response strategies such as green normal operation, yellow adaptive control, or red emergency intervention. This achieves a shift from passive response to proactive prevention and control, reducing the risk of excessive emissions and improving the system's operational stability and shock resistance.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A wastewater purification and treatment system for beef and mutton slaughtering includes the following modules:

[0009] Monitoring module: Real-time acquisition of multi-source operating parameters of the process nodes of the beef and mutton slaughter wastewater purification and treatment system. The process nodes include the main inlet, the outlet of the slag removal and oil separation unit, the outlet of the air flotation unit, and the biochemical treatment unit. The multi-source operating parameters include flow parameters and key pollutants. The flow parameters are the instantaneous influent flow data of the main inlet. The key pollutants include chemical oxygen demand, suspended solids concentration, and oil concentration. The module also preprocesses the acquired multi-source operating parameters.

[0010] Pollution equivalent estimation module: Constructs a water pollution equivalent estimation model, using a time series estimation neural network, including an input layer, a hidden layer, and an output layer. The input layer receives preprocessed multi-source operating parameters, the hidden layer is responsible for data processing of the multi-source operating parameters, and the output layer is used to output the pollution equivalent estimation values ​​of key pollutants. The preprocessed multi-source operating parameters are input into the water pollution equivalent estimation model to obtain the pollution equivalent estimation values ​​of key pollutants.

[0011] System resilience assessment module: Real-time collection of dissolved oxygen concentration data and sludge concentration data from the biochemical treatment unit; preprocessing of the collected dissolved oxygen concentration data and sludge concentration data; calculation of the system resilience coefficient based on the preprocessed dissolved oxygen concentration data and sludge concentration data; and calculation of the dynamic early warning threshold based on the system resilience coefficient.

[0012] Risk warning and control module: Based on the pollution equivalent estimated value of key pollutants, calculate the system risk index, compare the system risk index with the dynamic warning threshold, trigger the corresponding graded warning according to the comparison result of the system risk index and the dynamic warning threshold, and execute the corresponding control strategy according to the triggered graded warning.

[0013] Furthermore, the method for preprocessing the collected multi-source operating parameters is as follows:

[0014] The preprocessing includes data cleaning, time-series alignment, and data standardization of multi-source operating parameters.

[0015] The data cleaning of multi-source operating condition parameters includes the identification and removal of outliers and duplicate data, and the handling of missing values. Specifically, statistical methods are used to identify outliers and duplicate data in the multi-source operating condition parameters, delete outliers and duplicate data in the multi-source operating condition parameters, and fill missing values ​​in the multi-source operating condition parameters with the historical mean, median or mode of the same type of parameters in the multi-source operating condition parameters.

[0016] The time-series alignment specifically involves using the time series of the instantaneous influent flow data at the main inlet as the reference time axis, and synchronizing the key pollutants collected from the outlets of the slag and oil separation unit, the air flotation unit, and the biochemical treatment unit to the reference time axis through timestamp matching and interpolation algorithms.

[0017] Furthermore, the data standardization process is as follows:

[0018] The min-max normalization method is used to scale the collected multi-source operating condition parameters to the [0,1] interval. The formula used is as follows:

[0019] in, These are multi-source operating condition parameters after normalization. These are the original multi-source operating parameters. It is the minimum value among the same type of parameters in multi-source operating condition parameters. It represents the maximum value among parameters of the same type in multi-source operating conditions.

[0020] Furthermore, the method used to construct the water pollution equivalent estimation model is as follows:

[0021] The water pollution equivalent estimation model adopts a time series neural network structure based on long short-term memory network, including an input layer, a hidden layer and an output layer. The input layer is used to receive preprocessed multi-source operating parameters. The hidden layer is used to extract features and model the time dependence of the input multi-source operating parameters. The output layer is used to output the pollution equivalent estimation value of the key pollutant in the m-th hour in the future, where m is a positive integer preset according to the hydraulic residence time and early warning response requirements of the purification system.

[0022] The water pollution equivalent estimation model selects mean squared error as the loss function. The loss function is calculated based on the output results and the true labels. The weight parameters of the model are iteratively optimized through the backpropagation algorithm and the gradient descent optimizer until the model loss converges to a predetermined range.

[0023] Furthermore, the method for preprocessing the collected dissolved oxygen concentration data and sludge concentration data is as follows:

[0024] The preprocessing includes data cleaning and data smoothing filtering of dissolved oxygen concentration data and sludge concentration data.

[0025] The data cleaning method involves using statistical methods to identify and remove outliers and duplicate data in the dissolved oxygen concentration data and sludge concentration data; for missing values ​​in the dissolved oxygen concentration data and sludge concentration data, linear interpolation with the same parameter in adjacent time periods is used to fill them in.

[0026] The data smoothing and filtering process employs a moving average method to smooth the dissolved oxygen concentration data and sludge concentration data after cleaning, in order to suppress measurement noise and short-term drastic fluctuations. The window length for smoothing is calibrated based on the hydraulic characteristics of the biochemical treatment unit and the changes in microbial activity.

[0027] Furthermore, the formula used to calculate the system toughness coefficient is as follows:

[0028]

[0029] in, for The system resilience coefficient at any given time;

[0030] For preprocessed Dissolved oxygen concentration data at any given time;

[0031] For preprocessed Real-time sludge concentration data;

[0032] and These are the optimal dissolved oxygen concentration setpoint and the optimal sludge concentration setpoint, determined based on the combined degradation efficiency of the target pollutants in the biochemical treatment unit and the activity of sludge microorganisms.

[0033] and These are the allowable fluctuation ranges of dissolved oxygen concentration and sludge concentration, determined based on process tolerance limits and historical operating data, respectively.

[0034] and These are the preset contribution weighting coefficients for dissolved oxygen deviation and sludge concentration deviation, respectively. and Based on the historical operating data of the biochemical treatment unit and the experience of domain experts, and in accordance with the requirements... ,as well as > .

[0035] Furthermore, the formula used to calculate the dynamic early warning threshold is as follows:

[0036]

[0037] in, for Dynamic warning thresholds at any given time;

[0038] The preset baseline warning threshold;

[0039] The value is a preset stability constant, and satisfies 0 < ≤1.

[0040] Furthermore, the formula used to calculate the system risk index is as follows:

[0041]

[0042] in, for The system risk index at any given moment;

[0043] , and These are the pollution equivalent estimated values ​​of chemical oxygen demand, suspended solids concentration, and oil concentration for the m-th hour in the future, output by the water pollution equivalent estimation model.

[0044] , and These are the emission standard limits corresponding to chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. The emission standard limits are determined comprehensively based on the national water pollutant emission standards and the requirements of the project's environmental impact assessment approval.

[0045] , and These are the preset contribution weighting coefficients for chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. , and Based on a comprehensive assessment of historical influent water quality characteristics, the ecotoxicological contribution of pollutants, and the experience of experts in the field, and meeting the following requirements... ,as well as > > .

[0046] Furthermore, the logic for triggering corresponding tiered early warnings based on the comparison results between the system risk index and the dynamic early warning threshold is as follows:

[0047] when < At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in a stable operating period;

[0048] when ≤ < At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in the load adaptation period;

[0049] when ≥ At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in the impact overload period;

[0050] in, The peak safety factor is preset based on the historical operating data of the wastewater purification and treatment system for beef and mutton slaughtering and the experience of experts in the field, and >1.

[0051] Furthermore, the execution logic for implementing the corresponding control strategy based on the triggered tiered early warning is as follows:

[0052] When the wastewater purification and treatment system for beef and mutton slaughtering is in a stable operating period, the system outputs a green normal signal and does not trigger control commands.

[0053] When the wastewater purification and treatment system for beef and mutton slaughtering is in the load adaptation period, the system outputs a yellow warning signal and executes adaptation and control commands.

[0054] When the wastewater purification and treatment system for beef and mutton slaughtering is in the impact overload period, the system outputs a red warning signal and executes emergency intervention and control commands.

[0055] Compared with the prior art, the beneficial effects of the present invention are:

[0056] This invention introduces time-series forecasting technology into the treatment of beef and mutton slaughterhouse wastewater by constructing a water pollution equivalent estimation model. This enables advanced estimation of key pollutants such as chemical oxygen demand (COD), suspended solids concentration, and oil concentration, overcoming the limitations of traditional monitoring systems that rely solely on real-time feedback and exhibit delayed responses to water quality fluctuations. The system resilience assessment module calculates the deviation of dissolved oxygen and sludge concentration from their optimal states in real time and dynamically generates adaptive early warning thresholds, solving the problem of false alarms and missed alarms caused by traditional fixed thresholds under fluctuating operating conditions. Furthermore, the system integrates pollution equivalent estimation values ​​and dynamic thresholds to construct a graded risk index. Based on whether the system is in a stable operation period, load adaptation period, or shock overload period, it automatically triggers a three-level response strategy: green normal, yellow adaptive control, or red emergency intervention. This promotes the transformation of slaughterhouse wastewater treatment from passive alarm to proactive early warning, improving the system's resistance to shock loads and operational stability, reducing the risk of excessive emissions and operational energy consumption, and providing more accurate and reliable technical support for wastewater treatment in the slaughtering industry. Attached Figure Description

[0057] Figure 1 A block diagram of a wastewater purification and treatment system for beef and mutton slaughtering;

[0058] Figure 2 This is a schematic diagram of the operation process of a wastewater purification and treatment system for beef and mutton slaughtering. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0060] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0061] Example:

[0062] Please see Figures 1-2 The present invention provides a technical solution:

[0063] A wastewater purification and treatment system for beef and mutton slaughtering includes the following modules:

[0064] Monitoring Module: Real-time acquisition of multi-source operating parameters of the process nodes of the beef and mutton slaughter wastewater purification system. The process nodes include the main inlet, the outlet of the slag removal and oil separation unit, the outlet of the air flotation unit, and the internal parts of the biochemical treatment unit. The multi-source operating parameters include flow parameters and key pollutants. The flow parameters are the instantaneous influent flow data of the main inlet, used to characterize the dynamic changes of the system's influent load. The key pollutants include chemical oxygen demand, suspended solids concentration, and oil concentration, which serve as core indicators reflecting the intensity of water pollution and treatment efficiency. The acquired multi-source operating parameters are preprocessed to ensure the integrity, consistency, and comparability of the data sequence, providing a reliable data foundation for subsequent analysis.

[0065] The method for preprocessing the collected multi-source operating parameters is as follows:

[0066] The preprocessing includes data cleaning, time-series alignment, and data standardization of multi-source operating parameters.

[0067] The data cleaning of multi-source operating parameters includes the identification and removal of outliers and duplicate data, and the handling of missing values. Specifically, statistical methods are used based on the distribution characteristics of each parameter to identify and delete values ​​that exceed the normal fluctuation range or significantly deviate from the process rules. For duplicate records that are completely identical or logically conflicting, deduplication is performed by comparing timestamps and data content to ensure the uniqueness and traceability of the parameter sequence. Based on the statistical characteristics of the same type of parameters in the historical operating cycle, combined with the data distribution pattern, the mean, median, or mode is selected to fill in missing values, so as to maintain the overall data distribution structure while minimizing the sample bias caused by missing data.

[0068] The time-series alignment specifically involves using the time series of instantaneous influent flow data from the main inlet as the reference time axis. This time series of instantaneous influent flow data from the main inlet reflects the overall load changes of the system and possesses continuity and process relevance. Key pollutants collected from the slag and oil separation unit outlet, the flotation unit outlet, and the biochemical treatment unit are strictly matched with the reference time axis based on their respective original timestamps. For data points whose timestamps do not completely overlap, an interpolation algorithm is used for position synchronization. The interpolation process is based on the actual measurement values ​​of adjacent moments before and after, constructing a continuous change curve along the time dimension to estimate the corresponding pollutant concentration at the reference time point, thereby achieving complete alignment of data from each process node under a unified time coordinate.

[0069] The data standardization process is as follows:

[0070] The min-max normalization method is used to scale the collected multi-source operating condition parameters to the [0,1] interval. The formula used is as follows:

[0071]

[0072] in, These are multi-source operating condition parameters after normalization. These are the original multi-source operating parameters. It is the minimum value among the same type of parameters in multi-source operating condition parameters. It represents the maximum value among parameters of the same type in multi-source operating conditions.

[0073] Pollution Equivalent Estimation Module: Constructs a water pollution equivalent estimation model using a time-series neural network, comprising an input layer, a hidden layer, and an output layer. The input layer receives preprocessed multi-source operating parameters, which encompass the time-series expression of flow characteristics and key pollutants at different process nodes, providing the water pollution equivalent estimation model with basic data reflecting the current state and historical trends of the system. The hidden layer is responsible for performing multi-level nonlinear transformations and feature extraction on the multi-source operating parameters, capturing the implicit coupling relationships between the parameters and their dynamic evolution over time, thereby expressing the intrinsic driving mechanism of water pollution load. The output layer outputs the pollution equivalent estimation values ​​of key pollutants by inputting the preprocessed multi-source operating parameters into the water pollution equivalent estimation model to obtain the pollution equivalent estimation values ​​of key pollutants.

[0074] The method used to construct the water pollution equivalent estimation model is as follows:

[0075] The water pollution equivalent estimation model adopts a time series neural network structure based on long short-term memory network, including an input layer, a hidden layer and an output layer. The input layer has input nodes that match the dimensions of the pre-processed multi-source operating parameters. Each node corresponds to the value of a certain type of parameter at the same timestamp on a process node. The input layer extracts the multi-source operating parameter sequence within a continuous historical period according to a fixed time window length. The instantaneous influent flow rate data, chemical oxygen demand, suspended solids concentration and oil concentration of the total influent at each time step are used as a set of input vectors and sent to the hidden layer along the time axis.

[0076] The hidden layer is composed of several long short-term memory units (LSMs) connected in time steps. Each LSM at each time step includes a forget gate, an input gate, an output gate, and a memory unit. Each gate structure has a trainable weight matrix and a bias term. At each time step, the hidden layer receives the current input vector and the hidden state and memory state passed from the previous time step. The forget gate calculates the forget weight based on the current input and the previous hidden state to determine the proportion of historical information to be discarded. The input gate calculates the candidate memory update quantity based on the same input and filters the content to be written. The output gate calculates the current hidden state based on the updated memory state. The memory unit is updated sequentially along the time steps, continuously transmitting the pollution load characteristics of the early input of the sequence to subsequent time steps, so that the model can capture the hysteresis response relationship between the influent flow rate and the concentration of pollutants migrating along the process unit. The number of stacked layers of LSMs in the hidden layer and the number of units in each layer are preset according to the length of the historical operating data sequence and the pollution load fluctuation period.

[0077] The output layer is a fully connected layer. Its input is the hidden state vector output by the hidden layer at the last time step. The number of neurons in the output layer is the same as the number of key pollutant types to be estimated. Each output neuron corresponds to the pollution equivalent estimated value of one of the pollutants, namely chemical oxygen demand, suspended solids concentration, or oil concentration. The output layer uses a linear activation function to map the high-dimensional features extracted by the hidden layer to a continuous numerical space and outputs the pollution equivalent estimated values ​​of chemical oxygen demand, suspended solids concentration, and oil concentration in the m-th hour in the future, where m is a positive integer preset according to the hydraulic residence time and early warning response requirements of the purification system.

[0078] During the training phase, the water pollution equivalent estimation model uses mean squared error as the loss function. The estimated value output by the water pollution equivalent estimation model is compared with the actual sample labels one by one. The prediction deviation is calculated and propagated back along the network structure. The gradient of each weight parameter is calculated layer by layer. The optimizer iteratively updates the weights according to the gradient descent principle, so that the loss function value continues to decrease during the training process. After multiple rounds of forward calculation and parameter correction, the model output gradually approaches the actual pollution load level, and the loss curve tends to be stable within the preset acceptable range.

[0079] System resilience assessment module: Real-time acquisition of dissolved oxygen concentration data and sludge concentration data of the biochemical treatment unit. These data are core operating parameters for measuring the microbial activity and treatment efficiency of the biochemical system. The acquired dissolved oxygen concentration data and sludge concentration data are preprocessed to eliminate data deviations caused by instantaneous fluctuations in sensors, transmission interference, and occasional anomalies, ensuring the accuracy and time consistency of the data entering subsequent calculation stages. Based on the preprocessed dissolved oxygen concentration data and sludge concentration data, the system resilience coefficient is calculated. The system resilience coefficient quantifies the degree of deviation between the current dissolved oxygen and sludge concentrations and the steady-state range of the process target, comprehensively reflecting the ability of the biochemical treatment unit to maintain treatment efficiency and restore balance when facing fluctuations in influent water quality and quantity. The higher the resilience coefficient, the stronger the system's resistance to shock loads and the more sufficient the operating margin. Based on the system resilience coefficient, a dynamic early warning threshold is calculated.

[0080] The method for preprocessing the collected dissolved oxygen concentration data and sludge concentration data is as follows:

[0081] The preprocessing includes data cleaning and data smoothing filtering of dissolved oxygen concentration data and sludge concentration data.

[0082] The data cleaning method employs statistical methods to identify and remove outliers and duplicate data in dissolved oxygen concentration and sludge concentration data. Outlier identification is based on a reasonable fluctuation range set according to the historical distribution characteristics of the parameters themselves; values ​​exceeding this range and not conforming to the abrupt changes in the process mechanism are deemed outliers and removed. Duplicate data is detected using a dual criterion of timestamps and numerical values, retaining only the first occurrence to avoid redundant information interfering with the temporal characteristics. For missing values ​​in dissolved oxygen and sludge concentration data, linear interpolation of the same parameter in adjacent time periods is used for filling. A linear function is constructed using the values ​​at adjacent valid sampling times as endpoints, and the filler value is calculated based on the relative position of the missing time within the time interval, thereby restoring data integrity while maintaining the overall trend of data change.

[0083] The data smoothing and filtering process employs a moving average method to smooth the dissolved oxygen concentration data and sludge concentration data after cleaning. This process aims to suppress short-term, drastic fluctuations caused by factors such as sensor noise, water flow turbulence, and instantaneous aeration fluctuations, thereby highlighting the true trends in dissolved oxygen and sludge concentration. The smoothing window length is calibrated based on the hydraulic characteristics of the biochemical treatment unit and the changes in microbial activity.

[0084] Meanwhile, to ensure the long-term stable operation of the biochemical unit, the system is equipped with automatic sludge removal machinery to clean the bottom of the sedimentation zone and sludge hopper according to a preset cycle, so as to avoid local anaerobic conditions and reduced toughness caused by sludge accumulation.

[0085] The formula used to calculate the system toughness coefficient is as follows:

[0086]

[0087] in, for The system resilience coefficient at any given time;

[0088] For preprocessed Dissolved oxygen concentration data at any given time;

[0089] For preprocessed Real-time sludge concentration data;

[0090] and These are the optimal dissolved oxygen concentration setpoint and the optimal sludge concentration setpoint, determined based on the combined degradation efficiency of the target pollutants in the biochemical treatment unit and the activity of sludge microorganisms.

[0091] and These are the allowable fluctuation ranges of dissolved oxygen concentration and sludge concentration, determined based on process tolerance limits and historical operating data, respectively.

[0092] and These are the preset contribution weighting coefficients for dissolved oxygen deviation and sludge concentration deviation, respectively. and Based on the historical operating data of the biochemical treatment unit and the experience of domain experts, and in accordance with the requirements... Since dissolved oxygen concentration directly regulates the activity of the aerobic microbial respiratory chain, deviations from it not only rapidly affect the rate of organic matter degradation and the nitrification process, but also often exhibit a lag in aeration response during recovery regulation, resulting in a more significant marginal weakening effect on system resilience. > ;

[0093] and These represent the absolute deviations of the current dissolved oxygen concentration and sludge concentration from the optimal steady-state range of the process, respectively. The larger the deviation, the more serious the deviation between the actual operating state of the biochemical system and the ideal operating condition, and the higher the risk of inhibited microbial activity, excessive aeration, sludge loss, or bulking.

[0094] The formula used to calculate the dynamic early warning threshold is as follows:

[0095]

[0096] in, for Dynamic warning thresholds at any given time;

[0097] The preset baseline warning threshold is determined based on the historical operating data of the beef and mutton slaughter wastewater purification and treatment system, the stringency of emission standards, and the redundancy of process design.

[0098] The value is a preset stability constant, and satisfies 0 < ≤1;

[0099] when When the dissolved oxygen and sludge concentrations approach 0, it indicates that the current dissolved oxygen and sludge concentrations in the biological treatment unit are closer to the optimal setpoints, the microbial activity is within the ideal range, and the system has a strong buffering capacity against fluctuations in influent water quality and quantity. Correspondingly, the system can tolerate relatively higher expected pollution load values ​​without triggering an early warning, avoiding frequent false alarms due to overly stringent thresholds when the system is robust; when The higher the concentration, the more it indicates that the dissolved oxygen or sludge concentration deviates from the optimal range, microbial activity is inhibited, and the system's resilience to shocks is on the verge of being lost. The corresponding reduction means that the system is extremely sensitive to any additional increase in pollution load, thereby tightening the early warning threshold in advance and gaining a longer response time window for emergency control.

[0100] Risk warning and control module: Based on the pollution equivalent estimated value of key pollutants, calculate the system risk index, compare the system risk index with the dynamic warning threshold, trigger the corresponding graded warning according to the comparison result of the system risk index and the dynamic warning threshold, and execute the corresponding control strategy according to the triggered graded warning.

[0101] The formula used to calculate the system risk index is as follows:

[0102]

[0103] in, for The system risk index at any given moment;

[0104] , and These are the pollution equivalent estimated values ​​of chemical oxygen demand, suspended solids concentration, and oil concentration for the m-th hour in the future, output by the water pollution equivalent estimation model.

[0105] , and These are the emission standard limits corresponding to chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. The emission standard limits are determined comprehensively based on the national water pollutant emission standards and the requirements of the project's environmental impact assessment approval.

[0106] , and These are the preset contribution weighting coefficients for chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. , and Based on a comprehensive assessment of historical influent water quality characteristics, the ecotoxicological contribution of pollutants, and the experience of experts in the field, and meeting the following requirements... Because chemical oxygen demand (COD) is the core indicator of organic load, it fluctuates greatly, has a delayed degradation response, and causes the most severe ecological impact after exceeding the standard, making it the dominant risk source for excessive emissions in the system. Suspended solids and residual fine particles continuously adhere to the sludge surface, reducing mass transfer efficiency and sludge activity. Oils have been efficiently removed by oil-water separation and flotation, resulting in extremely low residual concentrations and infrequent daily fluctuations, thus having a limited impact on increasing system risk. Therefore... > > ;

[0107] when The larger the ratio, the closer the expected concentration of chemical oxygen demand is to the upper limit of the emission standard, leading to a higher systemic risk index. The corresponding increase; when The smaller the ratio, the more it indicates that organic pollutants have sufficient degradation capacity and buffer space in the biochemical unit, and the lower the system risk index. The corresponding decrease.

[0108] The logic for triggering corresponding tiered early warnings based on the comparison results between the system risk index and the dynamic early warning threshold is as follows:

[0109] when < When the pollution load is below the warning threshold, it indicates that the treatment capacity of each process is sufficient to absorb the impact of incoming water, the system is operating smoothly and has sufficient buffer capacity, and can maintain compliant discharge and normal operation without additional intervention. Therefore, the wastewater purification and treatment system for beef and mutton slaughtering is in a stable operating period.

[0110] when ≤ < When the pollution load is at a certain level, it indicates that the expected value of the pollution load has reached the warning line and the system has begun to bear significant pressure, but it is still within the adjustable range. It is determined that the wastewater purification and treatment system for beef and mutton slaughtering is in the load adaptation period.

[0111] when ≥ When the pollution load is significantly exceeded, it indicates that the system's capacity is nearing or has reached its limit, and the operational stability is under significant threat. Therefore, the wastewater purification and treatment system for beef and mutton slaughtering is in a period of shock overload.

[0112] in, The peak safety factor is preset based on the historical operating data of the wastewater purification and treatment system for beef and mutton slaughtering and the experience of experts in the field, and >1.

[0113] The execution logic for implementing the corresponding control strategy based on the triggered tiered early warning is as follows:

[0114] When the wastewater purification and treatment system for beef and mutton slaughtering is in a stable operating period, the system outputs a green normal signal and does not trigger control commands.

[0115] When the wastewater purification and treatment system for beef and mutton slaughtering is in the load adaptation period, the system outputs a yellow warning signal and executes adaptation and control instructions. The adaptation and control instructions include dynamically fine-tuning the operating parameters of the wastewater purification and treatment system for beef and mutton slaughtering, optimizing the coordination and matching of each unit, and appropriately moving the control nodes forward, so as to smooth load fluctuations and prevent risk escalation.

[0116] When the wastewater purification and treatment system for beef and mutton slaughtering is under shock overload, the system outputs a red warning signal and executes emergency intervention and control commands. The emergency intervention and control commands include forcibly reducing the treatment load, activating the system's protective regulation, rapidly diverting the excessive pollution load, and automatically adding high-efficiency activated carbon and special water pollution prevention and control agents to enhance the rapid adsorption and toxicity buffering of the shock load, so as to curb the spread of risk and ensure the safe operation of the system.

[0117] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0118] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0119] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A wastewater purification and treatment system for beef and mutton slaughtering, characterized in that, Includes the following modules: Monitoring module: Real-time acquisition of multi-source operating parameters of the process nodes of the beef and mutton slaughter wastewater purification and treatment system. The process nodes include the main inlet, the outlet of the slag removal and oil separation unit, the outlet of the air flotation unit, and the biochemical treatment unit. The multi-source operating parameters include flow parameters and key pollutants. The flow parameters are the instantaneous influent flow data of the main inlet. The key pollutants include chemical oxygen demand, suspended solids concentration, and oil concentration. The module also preprocesses the acquired multi-source operating parameters. Pollution equivalent estimation module: Constructs a water pollution equivalent estimation model, using a time series estimation neural network, including an input layer, a hidden layer, and an output layer. The input layer receives preprocessed multi-source operating parameters, the hidden layer is responsible for data processing of the multi-source operating parameters, and the output layer is used to output the pollution equivalent estimation values ​​of key pollutants. The preprocessed multi-source operating parameters are input into the water pollution equivalent estimation model to obtain the pollution equivalent estimation values ​​of key pollutants. System resilience assessment module: Real-time collection of dissolved oxygen concentration data and sludge concentration data from the biochemical treatment unit; preprocessing of the collected dissolved oxygen concentration data and sludge concentration data; calculation of the system resilience coefficient based on the preprocessed dissolved oxygen concentration data and sludge concentration data; and calculation of the dynamic early warning threshold based on the system resilience coefficient. Risk warning and control module: Based on the pollution equivalent estimated value of key pollutants, calculate the system risk index, compare the system risk index with the dynamic warning threshold, trigger the corresponding graded warning according to the comparison result of the system risk index and the dynamic warning threshold, and execute the corresponding control strategy according to the triggered graded warning.

2. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 1, characterized in that: The method for preprocessing the collected multi-source operating parameters is as follows: The preprocessing includes data cleaning, time-series alignment, and data standardization of multi-source operating parameters. The data cleaning of multi-source operating condition parameters includes the identification and removal of outliers and duplicate data, and the handling of missing values. Specifically, statistical methods are used to identify outliers and duplicate data in the multi-source operating condition parameters, delete outliers and duplicate data in the multi-source operating condition parameters, and fill missing values ​​in the multi-source operating condition parameters with the historical mean, median or mode of the same type of parameters in the multi-source operating condition parameters. The time-series alignment specifically involves using the time series of the instantaneous influent flow data at the main inlet as the reference time axis, and synchronizing the key pollutants collected from the outlets of the slag and oil separation unit, the air flotation unit, and the biochemical treatment unit to the reference time axis through timestamp matching and interpolation algorithms.

3. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 2, characterized in that: The data standardization process is as follows: The min-max normalization method is used to scale the collected multi-source operating condition parameters to the [0,1] interval. The formula used is as follows: in, These are multi-source operating condition parameters after normalization. These are the original multi-source operating parameters. It is the minimum value among the same type of parameters in multi-source operating condition parameters. It represents the maximum value among parameters of the same type in multi-source operating conditions.

4. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 1, characterized in that: The method used to construct the water pollution equivalent estimation model is as follows: The water pollution equivalent estimation model adopts a time series neural network structure based on long short-term memory network, including an input layer, a hidden layer and an output layer. The input layer is used to receive preprocessed multi-source operating parameters. The hidden layer is used to extract features and model the time dependence of the input multi-source operating parameters. The output layer is used to output the pollution equivalent estimation value of the key pollutant in the m-th hour in the future, where m is a positive integer preset according to the hydraulic residence time and early warning response requirements of the purification system. The water pollution equivalent estimation model selects mean squared error as the loss function. The loss function is calculated based on the output results and the true labels. The weight parameters of the model are iteratively optimized through the backpropagation algorithm and the gradient descent optimizer until the model loss converges to a predetermined range.

5. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 1, characterized in that: The method for preprocessing the collected dissolved oxygen concentration data and sludge concentration data is as follows: The preprocessing includes data cleaning and data smoothing filtering of dissolved oxygen concentration data and sludge concentration data. The data cleaning method involves using statistical methods to identify and remove outliers and duplicate data in the dissolved oxygen concentration data and sludge concentration data; for missing values ​​in the dissolved oxygen concentration data and sludge concentration data, linear interpolation with the same parameter in adjacent time periods is used to fill them in. The data smoothing and filtering process employs a moving average method to smooth the dissolved oxygen concentration data and sludge concentration data after cleaning, in order to suppress measurement noise and short-term drastic fluctuations. The window length for smoothing is calibrated based on the hydraulic characteristics of the biochemical treatment unit and the changes in microbial activity.

6. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 5, characterized in that: The formula used to calculate the system toughness coefficient is as follows: in, for The system resilience coefficient at any given time; For preprocessed Dissolved oxygen concentration data at any given time; For preprocessed Real-time sludge concentration data; and These are the optimal dissolved oxygen concentration setpoint and the optimal sludge concentration setpoint, determined based on the combined degradation efficiency of the target pollutants in the biochemical treatment unit and the activity of sludge microorganisms. and These are the allowable fluctuation ranges of dissolved oxygen concentration and sludge concentration, determined based on process tolerance limits and historical operating data, respectively. and These are the preset contribution weighting coefficients for dissolved oxygen deviation and sludge concentration deviation, respectively. and Based on the historical operating data of the biochemical treatment unit and the experience of domain experts, and in accordance with the requirements... ,as well as > .

7. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 6, characterized in that: The formula used to calculate the dynamic early warning threshold is as follows: in, for Dynamic warning thresholds at any given time; The preset baseline warning threshold; The value is a preset stability constant, and satisfies 0 < 0. ≤1.

8. The wastewater purification and treatment system for beef and mutton slaughtering according to claim 1, characterized in that: The formula used to calculate the system risk index is as follows: in, for The system risk index at any given moment; , and These are the pollution equivalent estimated values ​​of chemical oxygen demand, suspended solids concentration, and oil concentration for the m-th hour in the future, output by the water pollution equivalent estimation model. , and These are the emission standard limits corresponding to chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. The emission standard limits are determined comprehensively based on the national water pollutant emission standards and the requirements of the project's environmental impact assessment approval. , and These are the preset contribution weighting coefficients for chemical oxygen demand, suspended solids concentration, and oil concentration, respectively. , and Based on a comprehensive assessment of historical influent water quality characteristics, the ecotoxicological contribution of pollutants, and the experience of experts in the field, and meeting the following requirements... ,as well as > > .

9. A wastewater purification and treatment system for beef and mutton slaughtering according to claim 8, characterized in that: The logic for triggering corresponding tiered early warnings based on the comparison results between the system risk index and the dynamic early warning threshold is as follows: when < At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in a stable operating period; when ≤ < At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in the load adaptation period; when ≥ At that time, it was determined that the wastewater purification and treatment system for beef and mutton slaughtering was in the impact overload period; in, The peak safety factor is preset based on the historical operating data of the wastewater purification and treatment system for beef and mutton slaughtering and the experience of experts in the field, and >

1.

10. A wastewater purification and treatment system for beef and mutton slaughtering according to claim 9, characterized in that: The execution logic for implementing the corresponding control strategy based on the triggered tiered early warning is as follows: When the wastewater purification and treatment system for beef and mutton slaughtering is in a stable operating period, the system outputs a green normal signal and does not trigger control commands. When the wastewater purification and treatment system for beef and mutton slaughtering is in the load adaptation period, the system outputs a yellow warning signal and executes adaptation and control commands. When the wastewater purification and treatment system for beef and mutton slaughtering is in the impact overload period, the system outputs a red warning signal and executes emergency intervention and control commands.

Images