Livestock breeding pollution control method and system based on internet of things

Abstract

The present application relates to the technical field of Internet of Things environment monitoring and automatic control of sewage treatment, in particular to a livestock and poultry breeding pollution control method and system based on Internet of Things, comprising: obtaining water inlet data, biochemical pool state data and equipment state data; determining sensor state score according to network communication state data and deviation related data of dissolved oxygen measurement value relative to offline calibration reference value; when the score is lower than the preset score threshold, stopping using the dissolved oxygen measurement value of the current control period to participate in control calculation, and determining the dissolved oxygen estimated value according to the oxidation-reduction potential, mixed liquid suspended solid concentration and biochemical pool water temperature; determining the water inlet impact load coefficient according to the water inlet data, and determining the microbial activity coefficient according to the biochemical pool water temperature; generating control instructions and issuing them to the edge control node according to the above, adjusting the aeration equipment frequency and the reflux pump start-stop sequence; the present application can maintain control continuity when data distortion occurs.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) environmental monitoring and wastewater treatment automatic control technology, specifically to an IoT-based method and system for controlling pollution in livestock and poultry farming. Background Technology

[0002] Existing livestock and poultry breeding wastewater treatment systems include sensors, controllers, and aeration return devices installed at the inlet, biological tank, and actuator side; In related technologies, in order to control the treatment process of aquaculture wastewater, the operating parameters of the aeration equipment can be directly adjusted by using the real-time measurement value of the dissolved oxygen sensor; or, the return pump and the aeration equipment can be linked and controlled by manually set fixed time sequence rules to maintain the treatment state of the biological tank. When the farm environment is corrosive, communication links fluctuate, and the influent load changes beyond the preset fluctuation threshold, the former control method relies more heavily on sensor measurements than the preset ratio, while the latter control method has a preset delay in its response to influent shocks and changes in microbial activity. Therefore, the livestock and poultry farming pollution control methods in related technologies are difficult to balance control stability and operational adaptability. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for controlling pollution in livestock and poultry farming based on the Internet of Things, which reduces the reliance on sensor measurements in corrosive environments and fluctuating communication links, and effectively improves the adaptability to water influx and changes in microbial activity, thereby balancing the stability of system control and adaptability to operating conditions.

[0004] The objective of this invention can be achieved through the following technical solutions: Acquire influent data, biochemical tank status data, and equipment status data during the aquaculture wastewater treatment process. Biochemical tank status data includes dissolved oxygen measurement, oxidation-reduction potential, mixed liquor suspended solids concentration, and biochemical tank water temperature. The sensor status score is determined based on the network communication status data in the equipment status data and the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is no longer used in the control calculation, and the estimated dissolved oxygen value is determined based on the redox potential, the concentration of suspended solids in the mixed liquor, and the water temperature of the biological treatment tank; when the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is used as the estimated dissolved oxygen value. The influent impact load coefficient is determined based on the influent data, and the microbial activity coefficient is determined based on the degree of deviation of the biological tank water temperature from the preset active reference water temperature. Based on the dissolved oxygen estimate, influent shock load coefficient, and microbial activity coefficient, the dissolved oxygen target range, aeration rate coefficient, and return flow sequence rule are determined, and then combined to obtain control commands. Control commands are sent to edge control nodes, which then adjust the frequency of aeration equipment and the start / stop sequence of the return pump according to the control commands.

[0005] Preferably, based on the network packet loss rate and heartbeat delay time in the device status data, and the deviation of the dissolved oxygen measurement value in the biochemical pool status data and offline calibration data from the dissolved oxygen sensor offline calibration reference value, packet loss score, delay score and deviation score are determined respectively, and sensor status score is determined based on the weighted sum of each score; When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from being used in the control calculation. The rate of change of the oxidation-reduction potential in the current control cycle is calculated, and the predicted dissolved oxygen value is determined based on the estimated dissolved oxygen value, rate of change, biological tank water temperature and mixed liquor suspended solids concentration obtained in the previous control cycle. The initial dissolved oxygen estimate is set as the average of the dissolved oxygen estimates within the historical control period. The dissolved oxygen prediction is then filtered and corrected based on the reference correction value obtained from offline calibration to obtain the dissolved oxygen estimate.

[0006] Preferably, the influent data includes influent flow rate, influent turbidity, and influent conductivity; the average influent flow rate, average influent turbidity, and average influent conductivity over the past 24 hours are obtained, and the relative deviations of influent flow rate, influent turbidity, and influent conductivity relative to the average influent flow rate, respectively, are determined, and the influent shock load coefficient is determined based on the weighted sum of the relative deviations.

[0007] Preferably, the preset dissolved oxygen control range is corrected based on the influent shock load coefficient and the microbial activity coefficient to obtain the dissolved oxygen target range; The aeration rate coefficient is obtained by correcting the basic aeration rate parameter based on the influent shock load coefficient, microbial activity coefficient and sensor status score. Determine the start-up and stop-down durations of the reflux pump, and determine the reflux timing rule based on the start-up and stop-down durations; combine the dissolved oxygen target range, aeration rate coefficient, and reflux timing rule to obtain the control command.

[0008] Preferably, after determining the influent shock load coefficient based on the influent data and the microbial activity coefficient based on the biochemical tank water temperature, the backflow control priority for comparison and judgment is also determined based on the weighted combination value of the influent shock load coefficient and the microbial activity coefficient. Determine if the backflow control priority is greater than the preset priority threshold; if the backflow control priority is not greater than the preset priority threshold, keep the currently executed backflow timing rule unchanged. The influent impact load coefficient is multiplied by the preset start-up duration gain constant to obtain the start-up extension time. The reciprocal of the microbial activity coefficient is multiplied by the preset shutdown reduction constant to obtain the shutdown reduction time. The updated duration is limited to the range of the preset minimum duration and the preset maximum duration. The updated reflux pump start-up duration and reflux pump shutdown duration are written into the reflux timing rules.

[0009] Preferably, after the control edge control node adjusts the frequency of the aeration equipment and the start-stop sequence of the return pump according to the control command, it also determines whether the number of times the edge control node has not received a control command for a consecutive time has reached a preset disconnection threshold. If the number of consecutive times no control commands are received does not reach the preset disconnection threshold, the current online control logic of the edge control node is maintained; if the number of consecutive times no control commands are received reaches the preset disconnection threshold, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on the locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

[0010] A pollution control system for livestock and poultry farming based on the Internet of Things includes a sensor unit, a communication unit, an edge control node, an aeration device and a return pump. The system also includes an acquisition module, an evaluation and calculation module, a coefficient calculation module, an instruction generation module and a control distribution module. The acquisition module connects the sensor unit and the communication unit to acquire influent data, biochemical tank status data, and equipment status data during the aquaculture wastewater treatment process. The evaluation calculation module is connected to the acquisition module. It is used to determine the sensor status score based on the network communication status data in the equipment status data and the relevant data on the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from participating in the control calculation. The dissolved oxygen estimated value is determined based on the redox potential, the concentration of suspended solids in the mixed liquor and the water temperature of the biological tank. When the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is used as the dissolved oxygen estimated value. The coefficient calculation module is connected to the acquisition module and the evaluation calculation module. It is used to determine the influent shock load coefficient based on the influent data and to determine the microbial activity coefficient based on the degree of deviation of the biological tank water temperature from the preset active reference water temperature. The instruction generation module is connected to the evaluation calculation module and the coefficient calculation module. It is used to determine the target range of dissolved oxygen, the aeration adjustment rate coefficient and the return flow sequence rule based on the dissolved oxygen estimate, the influent shock load coefficient and the microbial activity coefficient, and combine them to obtain control instructions. The control sending module connects the communication unit and the edge control node, and is used to send control commands to the edge control node to control the edge control node to adjust the frequency of the aeration equipment and the start-stop sequence of the return pump.

[0011] Preferably, the system also includes a network disconnection protection module, which is connected to the communication unit and the edge control node, and is used to determine whether the number of times the edge control node has not received control commands for a series of consecutive times has reached a preset disconnection threshold. If the number of consecutive times without receiving control commands does not reach the preset disconnection threshold, the current online control logic of the edge control node is maintained. When the number of consecutive times without receiving control commands reaches the preset disconnection threshold, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on the locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

[0012] The beneficial effects of this invention are: 1. The pollution control method for livestock and poultry farming based on the Internet of Things in this invention acquires influent data, biochemical tank status data, and equipment status data. It determines sensor status scores based on network communication status data and the deviation of dissolved oxygen measurement values ​​from offline calibration reference values. When the score is lower than a preset score threshold, the dissolved oxygen measurement value of the current control cycle is no longer used in the control calculation. Instead, the dissolved oxygen estimated value is determined based on the oxidation-reduction potential, the concentration of suspended solids in the mixed liquor, and the water temperature of the biochemical tank. This method can isolate distorted data under corrosive environments and communication link fluctuations in farms, reduce the interference of sudden changes or loss of original measurement values ​​on aeration control, and improve the stability and data reliability of the control process. 2. The IoT-based livestock and poultry farming pollution control method of this invention determines the dissolved oxygen prediction value by combining the dissolved oxygen estimate value, oxidation-reduction potential change rate, biological tank water temperature and mixed liquor suspended solids concentration when the sensor status score is lower than the preset score threshold. The method also uses the reference correction value obtained from offline calibration to filter and correct the dissolved oxygen prediction value to obtain the dissolved oxygen estimate value. This ensures the continuity of the control cycle operation, smooths data jumps, and avoids control interruption due to a single sensor malfunction, thereby improving the continuous operation capability of the wastewater treatment process under complex working conditions. 3. The pollution control method for livestock and poultry farming based on the Internet of Things in this invention determines the influent impact load coefficient based on the influent flow rate, influent turbidity, and influent conductivity, and determines the microbial activity coefficient based on the deviation of the biochemical tank water temperature from the preset active reference water temperature. This method can characterize the load intensity about to enter the biochemical tank in advance when concentrated flushing of livestock and poultry pens causes short-term high flow rate and high concentration influent impact, and at the same time distinguish the treatment capacity reduction caused by changes in microbial activity. This makes the control strategy no longer solely dependent on dissolved oxygen feedback in the biochemical tank, and improves the adaptability to influent impact and activity changes. Attached Figure Description

[0013] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A flowchart of a method for controlling livestock and poultry breeding pollution based on the Internet of Things (IoT) of the present invention; Figure 2 This is a structural block diagram of the livestock and poultry breeding pollution control system in the embodiments of this application. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figure 1 A method for controlling pollution in livestock and poultry farming based on the Internet of Things includes: acquiring influent data, biochemical tank status data and equipment status data during the treatment of livestock and poultry wastewater, wherein the biochemical tank status data includes dissolved oxygen measurement value, oxidation-reduction potential, mixed liquor suspended solids concentration and biochemical tank water temperature; The sensor status score is determined based on the network communication status data in the equipment status data and the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is no longer used in the control calculation, and the estimated dissolved oxygen value is determined based on the redox potential, the concentration of suspended solids in the mixed liquor and the water temperature of the biological treatment tank. When the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value will be used as the dissolved oxygen estimate value. The influent impact load coefficient is determined based on the influent data, and the microbial activity coefficient is determined based on the degree of deviation of the biological tank water temperature from the preset active reference water temperature. Based on the dissolved oxygen estimate, influent shock load coefficient, and microbial activity coefficient, the dissolved oxygen target range, aeration rate coefficient, and return flow sequence rule are determined, and then combined to obtain control commands. Control commands are sent to edge control nodes, which then adjust the frequency of aeration equipment and the start / stop sequence of the return pump according to the control commands. The sensor status score is determined based on the network communication status data in the equipment status data and the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen estimate value is determined. This includes: obtaining the network packet loss rate and heartbeat delay time from the equipment status data, and obtaining the deviation of the dissolved oxygen measurement value from the offline calibration reference value by combining the biochemical tank status data. Packet loss score, latency score, and deviation score are determined based on preset scoring intervals corresponding to network packet loss rate, heartbeat delay time, and deviation, respectively, with each score being a standardized score of the same dimension; The sensor status score is determined by a weighted sum of packet loss score, delay score, and deviation score, where each weight is a preset constant and the sum of the weights is 1. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from being used in the control calculation. The rate of change of the oxidation-reduction potential in the current control cycle is calculated, and the predicted dissolved oxygen value is determined based on the estimated dissolved oxygen value, rate of change, biological tank water temperature and mixed liquor suspended solids concentration obtained in the previous control cycle. The initial dissolved oxygen estimate is set as the average of the dissolved oxygen estimates over the historical control period. The predicted dissolved oxygen value is then filtered and corrected based on the reference correction value obtained from offline calibration to obtain the final dissolved oxygen estimate. The estimated dissolved oxygen value satisfies the following formula: , in, This is an estimated value for dissolved oxygen. The preset filter coefficients are greater than 0 and less than 1. This is the predicted value for dissolved oxygen. For reference correction values; The influent data includes influent flow rate, influent turbidity, and influent conductivity; the influent shock load coefficient is determined based on the influent data, including obtaining the average influent flow rate, average influent turbidity, and average influent conductivity over the past 24 hours; The relative deviations of the influent flow rate relative to the mean influent flow rate, the relative deviations of the influent turbidity relative to the mean influent turbidity, and the relative deviations of the influent conductivity relative to the mean influent conductivity were determined respectively. The influent impact load coefficient is determined by the weighted sum of relative deviations, where each weight is a preset constant and the sum of the weights is 1.

[0016] This embodiment discloses a method for controlling pollution in livestock and poultry farming based on the Internet of Things. By collecting multi-source data from the biochemical tank and the water inlet, data features are reconstructed when the equipment status affects the reliability of the data, and control commands are generated by combining the water inlet load and temperature decay. This method was applied to the aerobic biological treatment system of wastewater in an intensive pig farm. The system acquires influent data such as influent flow rate, influent turbidity, and influent conductivity, as well as biological tank status data such as dissolved oxygen measurement, oxidation-reduction potential, mixed liquor suspended solids concentration, and biological tank water temperature by sensor nodes deployed at the influent end and the biological tank area. Meanwhile, the system extracts network communication status data such as network packet loss rate and heartbeat delay time through the communication link; in the underlying communication engineering implementation, the dissolved oxygen sensor and other underlying devices of the biochemical pool are connected to the field gateway through a serial communication bus using the remote terminal control system protocol, and the field gateway then interacts with the system communication link through the message queue telemetry transmission protocol or the industrial Ethernet communication protocol based on the transmission control protocol. To accurately obtain network status, the system communication module periodically sends out diagnostic heartbeat request messages with timestamps and incrementing sequence number fields. The system parses the timestamps in the response messages sent back by the edge nodes, calculates the difference between them and the current system time, and accurately calculates the heartbeat delay time. Meanwhile, by comparing the continuity of response message sequence numbers and the total number of messages that have not received a response within a set time window, such as 60 seconds, the network packet loss rate for the current control period is calculated. In a highly corrosive biological treatment tank environment, dissolved oxygen sensor probes are prone to adhesion or polarization, and the on-site communication network is susceptible to interference. If the control logic directly adopts the original measurement values, unexpected frequency fluctuations in the aeration equipment can easily occur due to sudden data changes or data loss. To this end, the system extracts the network packet loss rate and heartbeat delay time, and combines the deviation of dissolved oxygen measurement values ​​from offline calibration reference values, mapping them to packet loss scores, delay scores, and deviation scores of the same dimension. The specific mapping rules adopt the normalization logic of extremely small indicators. Taking the network packet loss rate as an example, the packet loss score is equal to the difference between the maximum tolerable packet loss rate and the current network packet loss rate, divided by the difference between the maximum tolerable packet loss rate and the ideal packet loss rate, and then multiplied by 100 to generate a percentage score. If the current network packet loss rate is greater than or equal to the maximum tolerable packet loss rate, the packet loss score is directly set to 0; the heartbeat delay time and deviation are converted using the same logic, and their scoring formula is as follows: , Among them, the maximum tolerance value for heartbeat delay time is, for example, preset to be 1. The ideal value is set as Regarding the deviation, its maximum tolerable deviation is, for example, a preset upper limit value. The ideal deviation is usually set as ; The sensor status score is generated by weighting and summing the scores according to the preset weights. When generating the sensor status score, the preset weights are allocated according to the different degrees of influence of deviation features and delay features on biochemical reactions: deviation score accounts for 0.5 weight, packet loss score accounts for 0.3 weight, and delay score accounts for 0.2 weight. When the sensor status score is lower than the preset score threshold (the preset score threshold is set to, for example, 1), When the system determines that the current measurement data is distorted, it immediately stops including the dissolved oxygen measurement value of the current cycle in the control calculation. At this point, the system extracts the estimated dissolved oxygen value from the previous control cycle, and combines it with the current redox potential change rate, biological tank water temperature, and mixed liquor suspended solids concentration to calculate the predicted dissolved oxygen value. The specific logic is as follows: The product of the suspended solids concentration in the mixed liquor and the corresponding active factor at water temperature is extracted as the basic oxygen consumption characteristic. The active factor corresponding to water temperature is determined through a preset piecewise linear mapping rule. For example, when the water temperature in the biological treatment tank is... to When the temperature range is reached, the activity factor is equal to the preset temperature conversion coefficient, for example... And the dimensions are Multiply by the current water temperature and The difference is added to the dimensionless fundamental factor of 0.5 to ensure dimensional consistency; When the rate of change of redox potential is negative, the system is in a state of rapid oxygen consumption, and the absolute value of this rate of change is used as the dynamic oxygen consumption compensation. When the rate of change of redox potential is greater than or equal to zero, the dynamic oxygen consumption compensation value is taken as... ; To eliminate dimensional conflicts and match the physical decay timescale, the system introduces a baseline oxygen consumption conversion coefficient determined based on offline static oxygen consumption experiments, for example, a value of 0.008 with dimensions of _____. and the dimensionless The dynamic oxygen conversion coefficient, for example, takes the value of The baseline oxygen consumption characteristic and dynamic oxygen consumption compensation are multiplied by the aforementioned conversion coefficients to convert them into oxygen consumption rate components of a unified dimension. These are then multiplied by the duration of the current control cycle, e.g., 5 minutes, to map them uniformly to concentration decay amounts of the same dimension. Then, the terms are added together according to preset weights to obtain the comprehensive oxygen consumption reduction term; In the specific quantitative combination logic, based on the difference in characteristics between substrate overload and activity decay, the first preset weight of the influent shock load coefficient is, for example, 0.7, and the second preset weight of the variable reflecting activity is, for example, 0.3. The two are added together to obtain the comprehensive oxygen consumption decay term. The dissolved oxygen estimated value of the previous control cycle is subtracted from the comprehensive oxygen consumption decay term to obtain the dissolved oxygen predicted value. If the dissolved oxygen estimate for the previous period was... Based on the above parameters, conversion coefficients, and cycle duration, the comprehensive oxygen consumption reduction term obtained through quantitative calculation is: The predicted value for this period is... To ensure the continuity of control cycle calculations, the system sets the initial dissolved oxygen estimate to the average of the dissolved oxygen estimates over historical control cycles. To smooth out data jumps, the system filters and corrects the dissolved oxygen prediction value based on the reference correction value obtained from offline calibration to obtain the dissolved oxygen estimate. The reference correction value is the difference between the average measured value of the system's dissolved oxygen sensor and the manually measured value of a standard portable dissolved oxygen meter under steady-state conditions with no influent and constant aeration. The specific logic of its filter correction satisfies the following: the dissolved oxygen estimate is equal to the product of the filter coefficient and the dissolved oxygen prediction, plus 1 minus the product of the difference between the filter coefficient and the reference correction value, where the filter coefficient is a preset constant greater than 0 and less than 1. If the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is directly used as the dissolved oxygen estimate value to ensure that the control operation is not interrupted due to abnormal data. In addition, concentrated flushing of livestock and poultry pens can cause short-term high-flow-rate and high-concentration influent shocks; if only the dissolved oxygen feedback in the biological treatment tank is relied upon for regulation, the oxygen supply will lag behind the increase in oxygen consumption rate; therefore, the system extracts the average influent flow rate, average influent turbidity and average influent conductivity over the past 24 hours, and calculates the relative deviations of the current influent flow rate, influent turbidity and influent conductivity relative to the above averages. The calculation logic for relative deviation is to subtract the absolute value of the corresponding mean from the current measured value, and then divide by the mean; these relative deviations are weighted and summed according to preset weights to generate the influent impact load coefficient. When allocating weights, based on the different impact characteristics of flow rate, turbidity, and conductivity on hydraulic residence time, the preset weights for the relative deviations of influent flow rate, influent turbidity, and influent conductivity are set to 0.5, 0.3, and 0.2, respectively. This coefficient characterizes the load impact intensity about to enter the biological treatment tank, providing a quantitative basis for pre-regulation. Simultaneously, the system calculates the deviation of the biological treatment tank water temperature from the preset active baseline water temperature to determine the microbial activity coefficient, distinguishing between water oxygenation limitation and microbial activity limitation. The specific mapping rule is as follows: If the current water temperature in the biochemical tank is lower than the preset activity reference water temperature, for example... Then the microbial activity coefficient is equal to the dimensionless preset temperature decay constant, for example... The deviation is raised to the power of , and to ensure that the exponent in the physical formula is dimensionless, the deviation is strictly defined as the difference between the current biochemical tank water temperature and the preset active reference water temperature, divided by the reference temperature difference constant, i.e., dimensionless. And for example, the value... After dimensionless processing, if the current water temperature of the biological treatment tank is greater than or equal to the preset activity reference water temperature, the microbial activity coefficient is taken as 1.0. Finally, the system generates control commands based on the dissolved oxygen estimate, influent shock load coefficient, and microbial activity coefficient, and sends them to the edge control nodes to drive the aeration equipment and return pump to perform corresponding actions.

[0017] In some possible implementations, the dissolved oxygen target range, aeration rate coefficient, and return flow sequence rule are determined based on the dissolved oxygen estimate, influent shock load coefficient, and microbial activity coefficient, respectively, and combined to obtain control instructions, including: correcting the preset dissolved oxygen control range based on the influent shock load coefficient and microbial activity coefficient to obtain the dissolved oxygen target range; The aeration rate coefficient is obtained by correcting the basic aeration rate parameter based on the influent shock load coefficient, microbial activity coefficient and sensor status score. Determine the start-up and stop-down durations of the reflux pump, and determine the reflux timing rules based on the start-up and stop-down durations; The control command is obtained by combining the target dissolved oxygen range, the aeration rate coefficient, and the reflux timing rule. After determining the influent shock load coefficient based on the influent data and the microbial activity coefficient based on the biochemical tank water temperature, the process also includes: determining the recirculation control priority based on the weighted combination value of the influent shock load coefficient and the microbial activity coefficient, wherein the combination weight is a preset constant and the sum of the weights is 1; and determining whether the recirculation control priority is greater than the preset priority threshold. If not, the currently executed reflux timing rule remains unchanged; if yes, the influent impact load coefficient is multiplied by the preset start-up duration gain constant to obtain the start-up extension time, the reciprocal of the microbial activity coefficient is multiplied by the preset shutdown reduction constant to obtain the shutdown reduction time, and the updated duration is limited to the range of the preset minimum duration and the preset maximum duration. The updated reflux pump start-up duration and reflux pump shutdown duration are written into the reflux timing rule. After the control edge control node adjusts the frequency of the aeration equipment and the start-stop sequence of the return pump according to the control command, it also includes: determining whether the number of times the edge control node has not received a control command for a consecutive time has reached the preset disconnection threshold. If the condition is not met, the current online control logic of the edge control node is maintained; if the condition is met, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on the locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

[0018] The system generates executable control commands for edge nodes based on the extracted feature coefficients, and uses these control commands to maintain system operation under preset operating conditions and network anomalies. After obtaining the influent shock load coefficient and microbial activity coefficient, the intensive pig farm wastewater treatment system dynamically corrects the preset basic dissolved oxygen control range to generate the dissolved oxygen target range. Based on this, the system combines sensor status scores to correct the basic aeration rate parameter and obtain the aeration rate coefficient. The specific correction logic is as follows: the preset basic dissolved oxygen control range has an initial lower limit and an initial upper limit; when the influent shock load coefficient is greater than 0 and the load increases, the system adds the dimensionless influent shock load coefficient and the preset concentration gain parameter (e.g., a value) to the initial lower limit and initial upper limit respectively. And the dimensions are The product of these two terms enables the overall upward shift of the control interval and ensures that the dimensions on both sides of the equation are homogeneous. Meanwhile, if the microbial activity coefficient is less than 1, it indicates that the microbial oxygen consumption capacity is reduced. The system then uses the microbial activity coefficient as a proportional factor to reduce the width of the control interval and generate the final dissolved oxygen target interval. The specific logic for reducing the width of the control interval is as follows: keep the lower limit of the target interval after the overall upward shift unchanged, multiply the difference between the upper limit and the lower limit of the target interval after the upward shift by the microbial activity coefficient, and add the product to the lower limit of the target interval as the final upper limit of the target interval. In addition, the quantitative update logic of the aeration rate coefficient is as follows: the aeration rate coefficient is equal to the sum of the basic aeration rate parameter in units of Hz / (mg / L) multiplied by the preset basic bias constant, such as 1.0, and the influent shock load coefficient, and then multiplied by the quotient of the sensor status score divided by 100. The basic adjustment rate under steady-state conditions is set based on the basic bias constant, and the adjustment step size of the aeration frequency is limited based on the sensor status score. Simultaneously, the system calculates the start-up and shutdown durations of the reflux pump to form a reflux timing rule; the aforementioned dissolved oxygen target range, aeration rate coefficient, and reflux timing rule are combined and packaged, and then distributed to the edge control nodes on site. After parsing the instruction, the edge control node compares the target range with the local estimate and constructs a proportional control logic that considers physical response hysteresis to adjust the frequency of the aeration equipment. Specifically, the mapping relationship is as follows: when the local estimate is lower than the lower limit of the dissolved oxygen target range, the difference between the lower limit and the local estimate is multiplied by a unit of magnitude... The aeration rate adjustment coefficient is used as the frequency increment to increase the frequency of the current aeration equipment; When the local estimated value is higher than the upper limit of the dissolved oxygen target range, the difference between the local estimated value and the upper limit is multiplied by the aeration adjustment rate coefficient as the frequency reduction, thereby reducing the frequency of the current aeration equipment. After performing the above frequency increase / decrease actions, the edge control node will initiate a hysteresis lock-in period, such as 15 minutes, that matches the oxygen transfer timescale of the water body. During this lock-in period, the calculation of frequency increase / decrease based on the deviation difference will be paused and only the current frequency of the aeration equipment will be maintained to prevent the system control from diverging due to the lag in the dissolved oxygen response of a large volume of water body. When not in a hysteresis lockout period and the local estimate is within the dissolved oxygen target range, the current aeration equipment frequency should also be maintained. Meanwhile, the edge nodes drive the reflux pump according to the time sequence rules; when the influent impact load coefficient and the biological tank water temperature meet the preset high load and low temperature threshold conditions, the influent impact load coefficient and the microbial activity coefficient are weighted and combined to calculate the reflux control priority. In the combined logic for determining the priority of backflow control, the first preset weight of the influent impact load coefficient is 0.7 according to preset rules, and the second preset weight of the variable reflecting activity is 0.3. Meanwhile, since the lower the microbial activity, the greater the need for reflux dilution to alleviate toxicity inhibition, the system uses the difference between 1 and the microbial activity coefficient as the positive activity limit index. The product of the influent shock load coefficient and 0.7 is added to the product of the positive activity limit index and 0.3 to obtain the reflux control priority. When the priority is greater than the preset priority threshold, the system determines that the current working condition requires priority to buffer and dilute through hydraulic recirculation. At this time, the system extends the duration of the recirculation pump based on the influent impact load coefficient according to the preset ratio and shortens the duration of the recirculation pump shutdown based on the microbial activity coefficient. In the specific quantitative execution, in order to maintain the consistency of dimensions in the timing logic, the system multiplies the influent impact load coefficient with the preset start-up duration gain constant in minutes to generate the start-up extension time, and adds it to the current return pump start-up duration. Similarly, the downtime reduction time is generated by multiplying the reciprocal of the microbial activity coefficient by the preset downtime reduction constant in minutes, and then subtracted from the current downtime duration of the reflux pump. For example, if the current reflux pump is on for 30 minutes, and the extension time calculated based on the inlet water impact load coefficient is 10 minutes, then the updated value is 40 minutes. To prevent device overload, the updated time parameters are limited to the minimum and maximum durations allowed by the physical device, such as the maximum single on-time not exceeding 50 minutes, and are written into the backflow timing rules; if the priority does not reach the threshold, the current backflow timing rules will remain unchanged. In addition, since edge control nodes are deployed in harsh field environments, if the communication link with the main control room is unexpectedly interrupted, the regular nodes will stop due to the loss of control commands. To cope with this boundary situation, the software timer inside the edge control node, based on the embedded real-time operating system, maintains a communication counter task. At the engineering implementation level, each time the edge node's communication interface successfully receives, parses, and verifies a valid data packet containing control instructions (e.g., if the cyclic redundancy check is successful), a communication interruption or event callback is triggered, and the communication counter is cleared to zero. If the clearing operation is not triggered within a timed communication detection cycle, the timer overflow interrupt increments the communication counter by 1 to determine whether the number of consecutive times no control command has been received has reached the preset disconnection threshold. The preset disconnection threshold must ensure that its corresponding time span is strictly greater than the set hysteresis lockout period, for example, a value of 4 times. If the communication is normal and the counter has not reached the threshold, the edge node maintains the current online control logic. If the disconnection threshold is reached, the edge node automatically switches to an offline operation strategy. During the underlying switching process of the control state machine, the edge node flips the master control flag from remote collaborative mode to local closed-loop mode and temporarily blocks the external communication interface from writing access to the control parameter register to prevent erroneous instructions from overwriting it. In this state, the edge node accesses local non-volatile memory such as electrically erasable programmable read-only memory or backup random access memory area protected by mutex lock, locks and retrieves the dissolved oxygen target range and aeration regulation rate coefficient from the control instruction successfully written once in the local cache. The edge node independently calculates the dissolved oxygen estimate using the biochemical tank status data continuously collected by local sensors, and continues to perform closed-loop adjustment of the aeration equipment frequency based on the locked target range and rate parameters through the aforementioned proportional control logic of the product of the deviation difference and the rate parameters. If the disconnection lasts longer than the preset safe time limit, such as 2 hours based on the average sludge age of the aerobic tank, the edge node will automatically and smoothly transition the aeration frequency to the preset safe minimum frequency, such as 30% of the inverter's power frequency, to maintain the basic sludge-water mixing state and prevent over-aeration caused by long-term operation under a certain extreme working condition strategy, until communication is restored and the global commands are resynchronized.

[0019] Please see Figure 2 A pollution control system for livestock and poultry farming based on the Internet of Things (IoT) includes a sensor unit, a communication unit, an edge control node, an aeration device, and a return pump; the system also includes: The acquisition module connects the sensor unit and the communication unit to acquire influent data, biochemical tank status data and equipment status data during the aquaculture wastewater treatment process. The biochemical tank status data includes dissolved oxygen measurement value, oxidation-reduction potential, mixed liquor suspended solids concentration and biochemical tank water temperature. The evaluation calculation module, connected to the acquisition module, is used to determine the sensor status score based on the network communication status data in the equipment status data and the relevant data on the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from participating in the control calculation. The dissolved oxygen estimated value is determined based on the redox potential, the concentration of suspended solids in the mixed liquor and the water temperature of the biological treatment tank. When the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is used as the dissolved oxygen estimated value. The coefficient calculation module, connected to the acquisition module and the evaluation calculation module, is used to determine the influent shock load coefficient based on the influent data and to determine the microbial activity coefficient based on the degree of deviation of the biological tank water temperature from the preset active reference water temperature. The instruction generation module, connected to the evaluation calculation module and the coefficient calculation module, is used to determine the target range of dissolved oxygen, the aeration adjustment rate coefficient and the return flow sequence rule based on the dissolved oxygen estimate, the influent shock load coefficient and the microbial activity coefficient, and combine them to obtain control instructions. The control sending module connects the communication unit and the edge control node, and is used to send control commands to the edge control node to control the edge control node to adjust the frequency of the aeration equipment and the start-stop sequence of the return pump; In some possible implementations, the system also includes a network disconnection protection module, which is connected to the communication unit and the edge control node, and is used to determine whether the number of times the edge control node has not received control commands for a series of consecutive times has reached a preset disconnection threshold. If the condition is not met, the current online control logic of the edge control node is maintained; if the condition is met, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on the locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

[0020] The hardware components of the intensive pig farm wastewater treatment system include sensor units, communication units, edge control nodes, aeration equipment, and return pumps. At the software level, the acquisition module collects influent data, biochemical tank status data, and equipment status data from the sensor units through the communication unit. After receiving the above data, the evaluation and calculation module performs the calculation of sensor status score; when the score is lower than the preset threshold, it isolates the abnormal dissolved oxygen measurement value and calculates the estimated dissolved oxygen value based on parameters such as biological tank water temperature and redox potential to ensure the reliability of the data source for subsequent calculation modules; The coefficient calculation module calculates the influent shock load coefficient and the microbial activity coefficient based on the influent data and the biological tank water temperature. The instruction generation module combines the above estimated values ​​and coefficients to generate a comprehensive control instruction that includes the dissolved oxygen target range, aeration rate coefficient and return timing rules. The control and distribution module serializes the instructions according to a preset data structure, such as encapsulating them into a standard object simplified spectrum data format string or a preset hexadecimal instruction frame, and transmits them to the edge control node through an industrial routing communication unit that includes fourth-generation or fifth-generation mobile communication technology. The specific control command payload contains explicit control fields, such as the target range lower limit field, upper limit field, rate parameter field, and return timing field composed of integer data of on / off duration, to perform fine-grained scheduling of field equipment; In complex industrial electromagnetic environments, occasional communication link disconnections can cause edge control nodes to become isolated and lose the scheduling function of the host computer. To address this, the system is equipped with a network disconnection protection module on the edge control node side. This module continuously monitors the command reception status, and when the number of consecutive times no commands are received reaches a preset disconnection threshold, the network disconnection protection module automatically takes over control. In the underlying system architecture, the network outage protection module blocks the external network receiving task's access to the executor driver by changing the arbitration flag bit of the internal control bus or switching the control task pointer. At this time, the network outage protection module extracts the received control command parameters from the local storage medium with the power stored on it, and combines them with the dissolved oxygen estimate calculated by the edge control node based on the locally acquired biochemical pool status data to maintain the adjustment of the aeration equipment frequency. When the network communication status data meets the preset disconnection conditions, the network disconnection protection module triggers the offline operation strategy and executes the local closed-loop control logic; when it is determined that the communication link has been restored, the network disconnection protection module removes the control permission interception and resumes the execution of online collaborative instructions.

Claims

1. A method for controlling pollution in livestock and poultry farming based on the Internet of Things, characterized in that, include: Acquire influent data, biochemical tank status data, and equipment status data during the aquaculture wastewater treatment process. The biochemical tank status data includes dissolved oxygen measurement, oxidation-reduction potential, mixed liquor suspended solids concentration, and biochemical tank water temperature. The sensor status score is determined based on the network communication status data in the device status data and the deviation data of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is no longer used in the control calculation, and the estimated dissolved oxygen value is determined based on the redox potential, the concentration of suspended solids in the mixed liquid, and the water temperature of the biological tank. When the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is used as the dissolved oxygen estimate value; The influent impact load coefficient is determined based on the influent data, and the microbial activity coefficient is determined based on the degree of deviation of the biochemical tank water temperature from the preset active reference water temperature. Based on the dissolved oxygen estimate, the influent shock load coefficient, and the microbial activity coefficient, the dissolved oxygen target range, the aeration rate coefficient, and the return flow sequence rule are determined respectively, and combined to obtain control commands; The control command is sent to the edge control node, which then adjusts the frequency of the aeration equipment and the start / stop sequence of the return pump according to the control command.

2. The method for controlling livestock and poultry farming pollution based on the Internet of Things according to claim 1, characterized in that, A sensor status score is determined based on network communication status data in the device status data and deviation data of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than a preset score threshold, an estimated dissolved oxygen value is determined, including: The network packet loss rate and heartbeat delay time are obtained from the device status data, and the deviation of the dissolved oxygen measurement value from the offline calibration reference value is obtained by combining the biochemical pool status data. The packet loss score, latency score, and deviation score are determined based on the network packet loss rate, the heartbeat delay time, and the preset scoring interval corresponding to the deviation, respectively, wherein each score is a standardized score of the same dimension; The sensor state score is determined by a weighted sum of the packet loss score, the delay score, and the deviation score, wherein each weight is a preset constant and the sum of the weights is 1. When the sensor status score is lower than the preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from being used in the control calculation. The rate of change of the redox potential in the current control cycle is calculated, and the predicted dissolved oxygen value is determined based on the obtained dissolved oxygen estimate value of the previous control cycle, the rate of change, the water temperature of the biological tank, and the concentration of suspended solids in the mixed liquor. The initial dissolved oxygen estimate is set as the average of the dissolved oxygen estimates over the historical control period. The predicted dissolved oxygen value is then filtered and corrected based on the reference correction value obtained from offline calibration to obtain the final dissolved oxygen estimate. The estimated dissolved oxygen value satisfies the following formula: ， in, The estimated value of dissolved oxygen. The preset filter coefficients are greater than 0 and less than 1. The predicted dissolved oxygen value, This is the reference correction value.

3. The method for controlling livestock and poultry farming pollution based on the Internet of Things according to claim 1, characterized in that, The influent data includes influent flow rate, influent turbidity, and influent conductivity; the influent shock load coefficient is determined based on the influent data, including: Obtain the average value of the influent flow rate, the average value of the influent turbidity, and the average value of the influent conductivity over the past 24 hours; The relative deviations of the influent flow rate relative to the mean of the influent flow rate, the relative deviations of the influent turbidity relative to the mean of the influent turbidity, and the relative deviations of the influent conductivity relative to the mean of the influent conductivity are determined respectively. The influent impact load coefficient is determined based on the weighted sum of the relative deviations, wherein each weight is a preset constant and the sum of the weights is 1.

4. The method for controlling livestock and poultry farming pollution based on the Internet of Things according to claim 1, characterized in that, Based on the dissolved oxygen estimate, the influent shock load coefficient, and the microbial activity coefficient, the dissolved oxygen target range, aeration rate coefficient, and return flow sequence rules are determined, and combined to obtain control commands, including: The target dissolved oxygen range is obtained by correcting the preset dissolved oxygen control range based on the influent shock load coefficient and the microbial activity coefficient. The aeration rate coefficient is obtained by correcting the basic aeration rate parameter based on the influent shock load coefficient, the microbial activity coefficient, and the sensor status score. Determine the start-up duration and stop-down duration of the reflux pump, and determine the reflux timing rule based on the start-up duration and the stop-down duration; The control command is obtained by combining the dissolved oxygen target range, the aeration rate coefficient, and the reflux timing rule.

5. The method for controlling livestock and poultry farming pollution based on the Internet of Things according to claim 1, characterized in that, After determining the influent shock load coefficient based on the influent data and the microbial activity coefficient based on the biological tank water temperature, the process further includes: The priority of backflow control is determined based on the weighted combination of the influent shock load coefficient and the microbial activity coefficient, wherein the combination weight is a preset constant and the sum of the weights is 1; Determine whether the reflux control priority is greater than a preset priority threshold; If not, then the currently executed reflow timing rules remain unchanged; If so, the influent impact load coefficient is multiplied by the preset start-up duration gain constant to obtain the start-up extension time, the reciprocal of the microbial activity coefficient is multiplied by the preset shutdown reduction constant to obtain the shutdown reduction time, and the updated duration is limited to the range of the preset minimum duration and the preset maximum duration. The updated reflux pump start-up duration and reflux pump shutdown duration are written into the reflux timing rule.

6. The method for controlling livestock and poultry farming pollution based on the Internet of Things according to claim 1, characterized in that, After controlling the edge control node to adjust the aeration equipment frequency and the start / stop sequence of the return pump according to the control command, the method further includes: Determine whether the number of consecutive times the edge control node has not received the control command has reached a preset disconnection threshold; If the condition is not met, the current online control logic of the edge control node is maintained. If this is achieved, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

7. An Internet of Things (IoT)-based livestock and poultry farming pollution control system, applied to implement the IoT-based livestock and poultry farming pollution control method according to any one of claims 1 to 6, characterized in that, Includes sensor units, communication units, edge control nodes, aeration equipment, and return pumps; The system also includes: The acquisition module, connected to the sensor unit and the communication unit, is used to acquire influent data, biochemical tank status data and equipment status data during the aquaculture wastewater treatment process. The biochemical tank status data includes dissolved oxygen measurement value, oxidation-reduction potential, mixed liquor suspended solids concentration and biochemical tank water temperature. An evaluation and calculation module, connected to the acquisition module, is used to determine a sensor status score based on network communication status data in the device status data and related data on the deviation of the dissolved oxygen measurement value from the offline calibration reference value. When the sensor status score is lower than a preset score threshold, the dissolved oxygen measurement value of the current control cycle is stopped from participating in the control calculation. A dissolved oxygen estimate value is determined based on the redox potential, the concentration of suspended solids in the mixed liquor, and the water temperature of the biological treatment tank. When the sensor status score is not lower than the preset score threshold, the dissolved oxygen measurement value is used as the dissolved oxygen estimate value. The coefficient calculation module, connected to the acquisition module and the evaluation calculation module, is used to determine the influent shock load coefficient based on the influent data, and to determine the microbial activity coefficient based on the degree of deviation of the biochemical tank water temperature from the preset active reference water temperature. The instruction generation module, connected to the evaluation calculation module and the coefficient calculation module, is used to determine the dissolved oxygen target range, aeration adjustment rate coefficient and return flow sequence rule based on the dissolved oxygen estimate, the influent shock load coefficient and the microbial activity coefficient, and combine them to obtain control instructions. The control sending module connects the communication unit and the edge control node, and is used to send the control commands to the edge control node to control the edge control node to adjust the frequency of the aeration equipment and the start-stop sequence of the return pump.

8. A livestock and poultry farming pollution control system based on the Internet of Things according to claim 7, characterized in that, The system also includes a network disconnection protection module, which is connected to the communication unit and the edge control node, and is used to determine whether the number of times the edge control node has not received the control command for a consecutive period of time has reached a preset disconnection threshold. If the condition is not met, the current online control logic of the edge control node is maintained. If this is achieved, the offline operation strategy is triggered, and the edge control node is controlled to execute the last received control command. Based on the dissolved oxygen target range and aeration rate coefficient in the last received control command, and combined with the dissolved oxygen estimate calculated by the edge control node based on locally acquired biochemical tank status data, the closed-loop regulation of the aeration equipment frequency is maintained.

Images