A breeding environment sewage treatment and data hosting system based on the Internet of Things

Abstract

The present application relates to the field of Internet of Things and aquaculture wastewater treatment control technology, in particular to a breeding environment wastewater treatment and data hosting system based on Internet of Things, comprising: a cloud scheduling layer, an edge control layer and a field device layer, containing sensors, fans and the like; data processing, collecting equipment and network data to generate time sequence characteristics; compensation and pressure resistance: determining probe drift compensation parameters and impact resistance targets according to characteristics; cooperative control, issuing control rules, and the edge end accurately adjusts the frequency converter of the fan in combination with real-time dissolved oxygen; network protection, when the network is disconnected or has high delay, automatically executing local degradation keep-alive instructions; closed-loop optimization, automatically updating the evaluation parameters according to the standard water discharge record and operation data; the present application can provide a unified data basis for subsequent impact resistance working condition target calculation and edge control execution.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) and aquaculture wastewater treatment and control technology, specifically to an IoT-based aquaculture environment wastewater treatment and data management system. Background Technology

[0002] As livestock farm wastewater treatment facilities develop towards automation and precision, aeration regulation and operation management under centralized flushing conditions have become important application requirements in the field of wastewater treatment control. In order to achieve stable operation of the livestock wastewater treatment process, it is particularly important to coordinate the processing of influent fluctuation identification, dissolved oxygen feedback correction, edge execution control, and continuous operation under abnormal communication conditions.

[0003] In the process of wastewater treatment in livestock farms, concentrated flushing of the pens can cause the liquid level to rise beyond the set threshold within a preset time threshold, resulting in a water volume surge. If the aeration is still adjusted according to the conventional aeration target, there is a risk of delayed aeration response, insufficient oxygen supply, or excessive energy consumption. At the same time, after long-term operation, the dissolved oxygen probes on site are affected by pollution and drift, so the dissolved oxygen data collected cannot accurately reflect the aeration status of the tank, thus affecting the accuracy of the control results. In the process of cloud scheduling and edge execution, the existing wastewater treatment control system may also be affected by factors such as increased link latency, communication failure, and missing data in some areas, which may cause the control rules to be unstable or the on-site equipment to be unable to adjust its operating conditions in a timely manner, making it difficult to balance effluent compliance, control continuity, and operating energy consumption.

[0004] Therefore, it is crucial to unify the processing of liquid level data, dissolved oxygen data, equipment operation data, and network status data during the aquaculture wastewater treatment process to form a time-series data foundation that can be used for drift judgment, impact identification, and control execution. Based on this foundation, it is essential to achieve dynamic adjustment of target dissolved oxygen under centralized flushing conditions, probe drift compensation, edge-side local control, and degraded operation under network outage conditions. At the same time, it is also crucial to continuously correct control parameters based on the results of periodic operation to ensure the continuous and stable operation of the aquaculture wastewater treatment system. Summary of the Invention

[0005] The purpose of this invention is to provide an IoT-based wastewater treatment and data management system for aquaculture environments, which avoids the situation where traditional aeration regulation often lags behind changes in influent, and avoids the uncertainty of control caused by qualitative judgment based solely on dissolved oxygen probe drift due to pollution. Furthermore, it can safely degrade control during network link anomalies or disconnections to maintain basic aeration capacity, so as to cope with the water volume shock load caused by concentrated flushing of aquaculture farms and realize pre-aeration adjustment in advance.

[0006] The objective of this invention can be achieved through the following technical solution: It includes a cloud scheduling layer, an edge control layer, and a field device layer communicating with it; the field device layer includes a dissolved oxygen probe, a liquid level sensor, a blower frequency converter, a booster pump, and flushing valves; the system is equipped with:

[0007] The data acquisition and processing module is used to acquire equipment acquisition data, including real-time raw dissolved oxygen data, equipment operating condition data, and network status data, and process them to obtain time-series characteristic data.

[0008] The probe drift compensation module is used to determine dynamic compensation parameters based on time-series characteristic data and equipment operating condition data;

[0009] The impact load judgment module is used to determine the impact resistance control target based on time-series characteristic data;

[0010] The control rule distribution module is used to generate and distribute control rule files based on dynamic compensation parameters and shock resistance control objectives.

[0011] The edge control execution module is used to determine the final control quantity based on the control rule file and the real-time raw dissolved oxygen, and to control the blower frequency converter to execute it;

[0012] The network outage degradation operation module is used to execute a locally fixed set of degradation maintenance instructions when the network connection is lost or the link latency exceeds the limit;

[0013] The operation assessment and parameter update module is used to update the system assessment parameters based on the system operation records and effluent compliance data.

[0014] Optionally, the data acquisition and processing module is used for:

[0015] Acquire liquid level data collected by liquid level sensors in the field equipment layer to determine the rate of liquid level rise; acquire raw dissolved oxygen data collected by dissolved oxygen probes;

[0016] Acquire the current frequency of the blower, the start / stop status of the booster pump, the opening signal of the flushing valve, and network link latency data;

[0017] The cleaning step difference is determined by the change in dissolved oxygen probe readings before and after probe cleaning, and the corresponding cleaning time markers are recorded.

[0018] The acquired liquid level data, raw dissolved oxygen data, equipment operating data, liquid level rise rate, network link latency data, cleaning step difference, and cleaning time markers are processed for noise reduction, missing data completion, and time synchronization to obtain time-series characteristic data.

[0019] Optionally, the probe drift compensation module is used for:

[0020] Historical cleaning step difference data is extracted from time-series feature data and compared with the current frequency of the blower to determine whether the dissolved oxygen probe is in a drift state.

[0021] When it is determined that the state is in a drift state, the basic deviation rate is determined based on the time difference between the current time and the most recent cleaning time mark, as well as the correlation between the cleaning step difference and the original dissolved oxygen.

[0022] The dynamic compensation coefficient is determined by combining the basic deviation rate with the preset initial penalty coefficient or the penalty coefficient of the historical control cycle, and the cleaning time mark, cleaning step difference, cleaning cycle and the penalty coefficient of the previous cycle are combined into dynamic compensation parameters.

[0023] Optionally, the impact load determination module is used for:

[0024] Continuously monitor the flush valve opening signal and liquid level rise rate in the time-series characteristic data; when the flush valve opening signal is active and the liquid level rise rate is greater than or equal to the preset threshold, it is determined that a water surge has occurred.

[0025] The target temporary dissolved oxygen increment is determined based on the liquid level change during flushing and the system's preset load conversion rate.

[0026] The timing of pre-aeration start-up and the target range for shock resistance are determined based on the target temporary dissolved oxygen increment and the upper limit frequency of the current operating condition, and combined with the preset operating condition exit conditions to form the shock resistance control target.

[0027] Optionally, the control rule distribution module is used for:

[0028] Based on the start-up time and exit conditions of pre-aeration, the working condition flow rules for normal working condition, pre-aeration working condition, shock-resistant working condition and recovery working condition are constructed, and switching conditions and minimum maintenance timeout conditions are configured for each working condition.

[0029] Each operating condition is mapped to the target dissolved oxygen range, and the compensation parameter version number, control algorithm parameter group, and upper and lower limits of the blower frequency converter are configured.

[0030] The operating condition flow rules, target dissolved oxygen range, and control parameters are encapsulated into a control rule file and distributed to the edge control layer.

[0031] Optionally, the edge control execution module is used for:

[0032] The control rule file is analyzed to determine the target dissolved oxygen value and dynamic compensation parameters corresponding to the current operating condition; the dynamic compensation parameters are used to compensate the real-time raw dissolved oxygen to obtain the actual feedback quantity;

[0033] The control increment is determined based on the deviation between the target dissolved oxygen value and the actual feedback value. The control increment is then combined with the historical control value from the previous control cycle and the upper and lower limits of the blower frequency converter for amplitude limiting. The final control value is then written into the blower frequency converter for execution.

[0034] Optionally, the network outage degradation operation module is used for:

[0035] Continuously monitor the number of consecutive network disconnections and link latency; when the number of consecutive network disconnections or link latency is greater than or equal to the set threshold and continues to time out, it is determined that the network communication is interrupted.

[0036] Freeze the current control rule file and effective target dissolved oxygen value, and call the pre-fixed degradation maintenance instruction set. Based on the range to which the liquid level data belongs, directly output the preset low frequency value to the blower frequency converter for safe degradation control. After the network is restored, upload the equipment execution record during the communication interruption and wait for the restoration of normal control.

[0037] Optionally, the performance evaluation and parameter update module is used for:

[0038] After the shock treatment cycle is completed, the cycle evaluation data is formed by summarizing the working condition switching records, actual feedback data, effluent compliance data, and total energy consumption data from equipment power metering.

[0039] The operation score is determined based on the total energy consumption, the level of ammonia nitrogen exceeding the standard in the effluent, and the frequency of operating condition switching; the penalty coefficient and load conversion rate are updated based on the operation score and written back to the cloud scheduling layer.

[0040] The beneficial effects of this invention are:

[0041] 1) This invention records the difference in probe readings before and after probe cleaning to form a cleaning step difference. By comparing the changes in multiple consecutive cleaning cycles with the blower frequency change rate, it can distinguish the measurement drift caused by probe surface contamination from the actual aeration capacity change of the equipment. Furthermore, it determines the basic deviation rate based on the correlation between the most recent cleaning time, the cleaning step difference, and the original dissolved oxygen. Combined with the penalty coefficient of the previous cycle, it obtains a dynamic compensation coefficient, thereby making the dissolved oxygen feedback close to the real aeration state, improving the accuracy of aeration control, and reducing the probability of misadjustment caused by probe contamination distortion.

[0042] 2) This invention continuously monitors the opening signal of the flushing valve and the rate of liquid level rise. When preset conditions are met, it determines the water volume impact and determines the target temporary dissolved oxygen increment based on the liquid level change and load conversion rate during flushing. Furthermore, it combines the current operating condition limit upper limit frequency to determine the pre-aeration start time and the impact resistance target range. This enables the system to enter the aeration adjustment stage in advance before the concentrated flushing arrives, avoiding the response lag problem caused by using conventional aeration targets. This enhances the system's ability to cope with short-term water inflow impacts and takes into account both oxygen supply demand and physical saturation upper limit constraints.

[0043] 3) This invention unifies dynamic compensation parameters and shock resistance control targets into a control rule file, and constructs the flow relationship between normal operating conditions, pre-aeration operating conditions, shock resistance operating conditions, and recovery operating conditions within it. At the same time, it configures target dissolved oxygen range, switching conditions, minimum maintenance timeout conditions, incremental proportional-integral-derivative algorithm parameter groups, and upper and lower limits of blower frequency conversion for each operating condition. This enables the edge control layer to directly parse the current operating condition based on the local rule file and complete compensation feedback, deviation calculation, control increment acquisition, and frequency limiting execution. This reduces the dependence on cloud-based direct control quantities issued cycle by cycle and improves the real-time performance, continuity, and executability of field control. Attached Figure Description

[0044] 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:

[0045] Figure 1 This is a module architecture diagram of an IoT-based wastewater treatment and data management system for aquaculture environments, as described in an embodiment of this application. Detailed Implementation

[0046] 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.

[0047] Please see Figure 1 A wastewater treatment and data-driven management system for aquaculture environments based on the Internet of Things (IoT) includes a cloud scheduling layer, an edge control layer, and a field device layer communicating with them. The field device layer includes dissolved oxygen probes, liquid level sensors, blower frequency converters, booster pumps, and flushing valves. The system is equipped with:

[0048] The data acquisition and processing module is used to acquire equipment acquisition data, including real-time raw dissolved oxygen data, equipment operating condition data, and network status data, and process them to obtain time-series characteristic data.

[0049] The probe drift compensation module is used to determine dynamic compensation parameters based on time-series characteristic data and equipment operating condition data;

[0050] The impact load judgment module is used to determine the impact resistance control target based on time-series characteristic data;

[0051] The control rule distribution module is used to generate and distribute control rule files based on dynamic compensation parameters and shock resistance control objectives.

[0052] The edge control execution module is used to determine the final control quantity based on the control rule file and the real-time raw dissolved oxygen, and to control the blower frequency converter to execute it;

[0053] The network outage degradation operation module is used to execute a locally fixed set of degradation maintenance instructions when the network connection is lost or the link latency exceeds the limit;

[0054] The operation assessment and parameter update module is used to update the system assessment parameters based on the system operation records and effluent compliance data;

[0055] The data acquisition and processing module is used to: acquire liquid level data collected by the liquid level sensor in the field equipment layer to determine the liquid level rise rate; and acquire raw dissolved oxygen data collected by the dissolved oxygen probe.

[0056] Acquire the current frequency of the blower, the start / stop status of the booster pump, the opening signal of the flushing valve, and network link latency data;

[0057] The cleaning step difference is determined by the change in dissolved oxygen probe readings before and after probe cleaning, and the corresponding cleaning time markers are recorded.

[0058] The acquired liquid level data, raw dissolved oxygen data, equipment operating data, liquid level rise rate, network link latency data, cleaning step difference, and cleaning time markers are processed for noise reduction, missing data completion, and time synchronization to obtain time-series characteristic data.

[0059] The probe drift compensation module is used to: extract historical cleaning step difference data from time-series feature data and compare it with the current frequency of the blower to determine whether the dissolved oxygen probe is in a drift state;

[0060] When it is determined that the state is in a drift state, the basic deviation rate is determined based on the time difference between the current time and the most recent cleaning time mark, as well as the correlation between the cleaning step difference and the original dissolved oxygen.

[0061] The dynamic compensation coefficient is determined by combining the basic deviation rate with the preset initial penalty coefficient or the penalty coefficient of the historical control cycle, and the cleaning time mark, cleaning step difference, cleaning cycle and the penalty coefficient of the previous cycle are combined into dynamic compensation parameters.

[0062] The edge control layer receives current data from the liquid level sensor, dissolved oxygen probe, blower frequency converter, lift pump and flushing valve according to the control cycle. The liquid level data is used to reflect changes in the influent, the dissolved oxygen data is used to characterize the aeration effect of the tank, the equipment status data is used to limit the current control capability, and the network status data is used to determine whether the remote scheduling link is available.

[0063] After the above data enters the data acquisition and processing module, it first undergoes noise reduction and time synchronization processing, and then is uniformly organized according to the same control cycle.

[0064] In the noise reduction process, a moving average filtering algorithm is used to filter out instantaneous spike anomaly data caused by on-site electromagnetic interference.

[0065] In time synchronization processing, a unified control cycle trigger time is used as a reference. For data points located within a preset allowable window before and after the reference time, the nearest valid data is extracted using the nearest interpolation method, so that all heterogeneous data points are aligned to the same timestamp in the time dimension.

[0066] For liquid level data, the liquid level rise rate is obtained by using the liquid level difference between adjacent time points; for probe cleaning events, the probe reading at the first preset time before probe cleaning starts and the probe reading at the second preset time after probe cleaning is completed are recorded. The difference between the probe reading at the second preset time after probe cleaning is completed and the probe reading at the first preset time before probe cleaning starts is obtained, and the difference is associated with and stored with the corresponding cleaning time marker.

[0067] The processed liquid level rise rate, raw dissolved oxygen data, blower current frequency, booster pump start / stop status, cleaning step difference, cleaning time marker, flushing valve opening signal and network link delay data are organized into feature data in the same time series for subsequent drift judgment, impact judgment and control rule generation.

[0068] Furthermore, if the liquid level or dissolved oxygen sampling is not successfully obtained within a certain control cycle, the data acquisition and processing module can read the stable record of the corresponding quantity in the previous effective cycle as the replacement value for the current cycle and add a missing measurement mark to avoid subsequent modules from interrupting calculations due to single-point missing measurement.

[0069] If no probe cleaning event occurs, no new cleaning step difference is generated in this cycle, but the most recent valid cleaning time stamp and its corresponding difference are retained for subsequent comparison.

[0070] If the network link latency data cannot be updated within the current period, the edge control layer is still allowed to use the most recently valid time-series feature data that has been organized locally to continue to participate in the calculation of the current period. At the same time, the link status of the current period is recorded as pending confirmation so that the subsequent network disconnection judgment module can continue to make a judgment based on the number of consecutive disconnections.

[0071] Based on the above data, the probe drift compensation module extracts the historical cleaning step difference from multiple consecutive cleaning cycles and compares it with the current frequency of the blower in the same range.

[0072] The system compares the reading recovery after probe cleaning with the operating status of the aeration equipment. If the cleaning step difference increases continuously while the current frequency change rate of the blower is within the preset fluctuation range, it indicates that there is no obvious correlation between the probe reading change and the aeration capacity of the equipment, and it can be determined that the measurement drift caused by probe surface contamination has occurred.

[0073] Furthermore, the system determines the basic deviation rate based on the time difference between the current time and the most recent cleaning time marker, combined with the correlation between the cleaning step difference and the original dissolved oxygen. Then, the basic deviation rate is multiplied by the penalty coefficient of the previous cycle to obtain the dynamic compensation coefficient of the current cycle.

[0074] Specifically, the preset fluctuation range is determined by extracting the natural fluctuation variance of the blower frequency within the same control cycle under historical stable operating conditions, and taking three times its standard deviation as the upper and lower fluctuation boundaries; the preset ratio is set to 10%.

[0075] The preset drift rate constant is obtained by recording the correlation between the probe's contamination aging time and the reading deviation in a standard test solution and then performing linear fitting. Assuming that the current time is 24 hours away from the last cleaning time, the system first calculates the step ratio, which is the ratio of the latest cleaning step difference to the original dissolved oxygen.

[0076] In conjunction with the preset drift rate constant of the above fitting, in order to avoid the division by zero error when the original dissolved oxygen approaches 0, which would cause numerical divergence, the system introduces a preset minimum positive real number compensation term, for example, set to 0.1 mg / L, during the calculation.

[0077] Meanwhile, since the dimension of the preset drift rate constant is the reciprocal of time, in order to prevent the deviation rate from going out of bounds due to the infinite growth of the time difference, the system applies a hard limit of, for example, 0.5 of the maximum physical deviation threshold to the calculation result.

[0078] After the above processing, the base deviation rate is equal to the product of the time difference, the preset drift rate constant, and the step ratio. If the product is greater than 0.5, it is truncated to 0.5. For example, the final base deviation rate is 0.05. The base deviation rate is calculated according to the following formula:

[0079]

[0080] in, The time difference marked by the cleaning time. As a preset drift rate constant, For the latest cleaning step difference, Original dissolved oxygen, To pre-define the minimum positive real number compensation term, The maximum physical deviation threshold. To obtain the smaller value of the function, Basic deviation rate;

[0081] The system is based on a mapping formula to convert the step change of the probe before and after cleaning into a basic deviation rate parameter, thereby reducing the control error caused by qualitative judgment alone.

[0082] Further read the penalty coefficient from the previous cycle. The initial value of this coefficient is 1.0. If the historical evaluation record shows that a miscompensation has occurred, it may be adjusted to 1.1, etc. Multiply the basic deviation rate of 0.05 by the penalty coefficient of 1.1 to obtain a dynamic compensation coefficient of 0.055.

[0083] The cleaning time stamp, cleaning step difference, cleaning cycle and penalty coefficient are combined into dynamic compensation parameters and written into the control rule association area to be issued, so that the edge control layer can directly call them in subsequent control cycles;

[0084] Furthermore, if any cycle in the continuous cleaning cycle lacks an effective cleaning step difference, the drift compensation module will not make a drift judgment directly, but will postpone the comparison until sufficient cycle data is obtained, in order to avoid misjudgment due to discontinuous samples.

[0085] If the rate of change of the blower frequency within the corresponding cycle is greater than the preset frequency change threshold, the cleaning step difference in this cycle will not be used as a basis for drift determination, but will only be retained as a historical reference value.

[0086] The system organizes the on-site sampling results into time-series feature data and transforms the differences before and after probe cleaning into dynamic compensation parameters, generating the data basis for target calculation and edge control execution under impact conditions;

[0087] The data acquisition and processing module is used to acquire liquid level data collected by the liquid level sensor in the field equipment layer and determine the liquid level rise rate based on the liquid level difference between adjacent times.

[0088] Acquire raw dissolved oxygen data collected by the dissolved oxygen probe; acquire the current frequency of the blower, the start / stop status of the booster pump, the opening signal of the flushing valve, and network link latency data;

[0089] The dissolved oxygen probe reading at the second preset time after probe cleaning is completed is subtracted from the dissolved oxygen probe reading at the first preset time before probe cleaning is started to obtain the cleaning step difference, and the corresponding cleaning time mark is recorded.

[0090] Denoising, missing data completion, and time synchronization are performed on liquid level rise rate, raw dissolved oxygen data, blower current frequency, booster pump start / stop status, cleaning step difference, cleaning time marker, flushing valve opening signal, and network link delay data to obtain time series characteristic data.

[0091] Among them, the probe drift compensation module is used to extract historical cleaning step difference data from time-series feature data and compare it with the current frequency of the blower in the equipment operating data;

[0092] When the increase in cleaning step difference for three consecutive cleaning cycles is greater than the preset ratio and the current frequency change rate of the blower is within the preset fluctuation range, the dissolved oxygen probe is determined to be in a drift state.

[0093] The basic deviation rate is determined based on the time difference between the current time and the most recent cleaning time marker, as well as the correlation between the cleaning step difference and the original dissolved oxygen. The basic deviation rate is multiplied by the penalty coefficient of the previous cycle to obtain the dynamic compensation coefficient. The cleaning time marker, cleaning step difference, cleaning cycle, and penalty coefficient are combined into dynamic compensation parameters.

[0094] In a preferred embodiment of the present invention, the impact load judgment module is used to: continuously monitor the flushing valve opening signal and the liquid level rise rate in the time-series characteristic data; and determine that a water impact has occurred when the flushing valve opening signal is effective and the liquid level rise rate is greater than or equal to a preset threshold.

[0095] The target temporary dissolved oxygen increment is determined based on the liquid level change during flushing and the system's preset load conversion rate.

[0096] The pre-aeration start time and shock resistance target range are determined based on the target temporary dissolved oxygen increment and the current operating condition limit frequency, and combined with the preset operating condition exit conditions to form the shock resistance control target.

[0097] The control rule distribution module is used to: construct the working condition flow rules for normal working condition, pre-aeration working condition, anti-shock working condition and recovery working condition based on the pre-aeration start time and exit conditions, and configure switching conditions and minimum maintenance timeout conditions for each working condition;

[0098] Each operating condition is mapped to the target dissolved oxygen range, and the compensation parameter version number, control algorithm parameter group, and upper and lower limits of the blower frequency converter are configured.

[0099] The operating condition flow rules, target dissolved oxygen range, and control parameters are encapsulated into a control rule file and distributed to the edge control layer;

[0100] The edge control execution module is used to: parse the control rule file to determine the target dissolved oxygen value and dynamic compensation parameters corresponding to the current operating condition; and use the dynamic compensation parameters to compensate the real-time raw dissolved oxygen to obtain the actual feedback amount.

[0101] The control increment is determined based on the deviation between the target dissolved oxygen value and the actual feedback value. The control increment is then combined with the historical control value from the previous control cycle and the upper and lower limits of the blower frequency converter for amplitude limiting. The final control value is then written into the blower frequency converter for execution.

[0102] This embodiment is applicable to the wastewater treatment scenario during centralized flushing of livestock pens in farms; after the flushing valve is opened, the rise in the wastewater network level will be greater than the preset level difference threshold within the continuous control cycle.

[0103] If conventional aeration targets are still used, aeration adjustments often lag behind changes in influent; the system converts flushing events, liquid level changes, and probe compensation results into control provisions that can be executed on the edge side.

[0104] The impact load judgment module continuously reads the flushing valve opening signal and liquid level rise rate from the timing characteristic data;

[0105] If only the flushing valve is open but the rate of liquid level rise does not reach the preset threshold, the system will maintain its original operating conditions and will not generate new anti-impact control targets, thus determining that an effective water impact condition has not yet been formed.

[0106] If the flushing valve opening signal is effective and the liquid level rise rate exceeds the preset threshold, it is determined that a water surge has occurred. The preset threshold is calibrated based on 1.5 to 2 times the historical maximum value of the instantaneous liquid level rise rate of the pipeline network during the normal non-flushing period of the aquaculture farm, which is used to effectively distinguish between local water fluctuations and global centralized flushing conditions.

[0107] The system calculates the target temporary dissolved oxygen increment by combining the liquid level change during flushing and the current load conversion rate. Based on this, and combined with the current operating condition limit frequency, it determines the period when aeration adjustment needs to be initiated in advance, thus obtaining the pre-aeration start time.

[0108] Simultaneously, a target dissolved oxygen range applicable to the shock resistance phase is generated based on the target temporary dissolved oxygen increment, and a shock resistance control target is formed in combination with the exit conditions; after the control target is formed, it is not directly written into the equipment, but is transferred to the control rule distribution module together with the aforementioned dynamic compensation parameters;

[0109] Specifically, if the liquid level change during the concentrated flushing period is 0.5m, the newly added impact water volume can be calculated by combining the known cross-sectional area of ​​the pool.

[0110] If the current load conversion rate is set to the mass of oxygen consumed per cubic meter of impact water, the system will then obtain the total oxygen demand. Dividing this by the total volume of the tank will convert it into the target temporary dissolved oxygen increment, for example, 1.5 mg / L.

[0111] Calculate the target temporary dissolved oxygen increment using the following formula:

[0112]

[0113] in, This represents the change in liquid level during concentrated flushing. The cross-sectional area of ​​the pool; This represents the current load conversion rate. This refers to the total volume of the pool. The target is a temporary increase in dissolved oxygen.

[0114] For example, assuming the change in liquid level The diameter is 0.5m, and the cross-sectional area of ​​the pool is... 100m 2 Current load conversion rate 30g / m 3 Total volume of the pool 1000m 3 Substituting the values ​​into the formula, the target temporary dissolved oxygen increment is calculated to be 1.5 mg / L.

[0115] After obtaining this increment, the system retrieves the maximum oxygen supply rate corresponding to the current operating condition's upper limit frequency, assuming it to be 0.5 mg / (L·h). The increment is then divided by the maximum oxygen supply rate to determine the time span required for the system to reach full load operation in advance. Extract the daily cyclical statistical flushing schedule from the historical operation database, and find the next most recent historical high-frequency flushing time node. Subtract the time span of early full-load operation Obtain the absolute timestamp and determine it as the start time of pre-aeration;

[0116] Meanwhile, the system uses the original conventional target dissolved oxygen center value plus the temporary dissolved oxygen increment of 1.5 mg / L as the initial new interval center value;

[0117] Since dissolved oxygen in water has a physical saturation limit under normal pressure, if the initial new range center value plus the allowable control deviation exceeds the physical saturation dissolved oxygen concentration limit at the current water temperature, for example, about 9.09 mg / L at normal temperature, the system will directly cut off the upper limit of the impact resistance target range to the saturation dissolved oxygen concentration limit and back-calculate and correct the center value downwards, so as to avoid driving the blower to run continuously without load at the limit frequency due to setting a physically unattainable target.

[0118] After passing the physical limit verification, the system uses this center value to offset the preset allowable control deviations above and below, such as ±0.3 mg / L, thereby clearly generating the target dissolved oxygen range applicable to the shock resistance stage;

[0119] Furthermore, if the flushing valve opening signal is effective and the rate of change of the liquid level rise rate is greater than the preset fluctuation threshold in multiple consecutive cycles, the system can review the results of consecutive cycles and only confirm the entry into the water flow impact state when the preset judgment conditions are met, so as to avoid frequent switching of operating conditions due to short-term disturbances.

[0120] If there is a missing measurement of the liquid level change, the target temporary dissolved oxygen increment will not be recalculated in this cycle. Instead, the judgment result corresponding to the previous effective cycle will be used, and the current operating condition will remain unchanged in the control rules. This maintains the continuity of the rule file and avoids the edge control layer receiving conflicting target intervals in the same flushing process.

[0121] After obtaining the dynamic compensation parameters and the shock resistance control target, the control rule distribution module will organize the two into a structured rule file.

[0122] The system constructs the flow relationship between normal working conditions, pre-aeration working conditions, shock-resistant working conditions and recovery working conditions based on the start time and exit conditions of pre-aeration, and provides the switching conditions and minimum maintenance timeout conditions for each working condition.

[0123] Each operating condition is assigned to a different target dissolved oxygen range, and the compensation parameter version number, incremental proportional-integral-differential algorithm parameter group, and upper and lower limits of the blower frequency converter are bound to it respectively.

[0124] The encapsulated control rule file is sent to the edge control layer and saved in version format, so that the edge control layer only needs to parse according to the current local version when executing, without having to re-request the cloud calculation results in each control cycle;

[0125] Furthermore, the control rule file associates the target range, switching conditions, and limiting parameters together, so that the edge control layer can determine the trigger node for entering or exiting a specific control phase, thereby reducing the probability of misjudging the working conditions during execution by the edge control layer.

[0126] If the rule file is not fully received during transmission, the edge control layer can continue to maintain the previous version of the rules and not switch to the incomplete new version, thereby ensuring that there is still an executable basis for the current control cycle;

[0127] After receiving the rule file, the edge control execution module parses the target dissolved oxygen range corresponding to the current operating condition and takes the center value of the range as the target dissolved oxygen value. At the same time, it reads the dynamic compensation parameters bound to the operating condition. The system corrects the real-time original dissolved oxygen based on the dynamic compensation coefficient to obtain the actual feedback amount.

[0128] For example, if the system determines that the probe surface contamination is causing the reading to be low and the returned dynamic compensation coefficient is a positive compensation ratio such as 0.055, the actual feedback quantity can be directly calculated using the equation:

[0129]

[0130] The actual feedback value is the only dissolved oxygen value used in the deviation calculation during the current control cycle. It is compared with the target dissolved oxygen value to form the deviation.

[0131] For this deviation, the edge control layer calculates the control increment according to the incremental proportional-integral-derivative (PI-DE) algorithm parameter set for the corresponding operating condition in the rule file. The specific execution rule is as follows:

[0132] Extract the current deviation, and call the locally stored deviations of the previous cycle and the cycle before that. Multiply the difference between the current deviation and the previous cycle deviation by the proportional coefficient in the parameter group to obtain the proportional term increment. Multiply the current deviation by the integral coefficient to obtain the integral term increment. Multiply the result of subtracting twice the previous cycle deviation from the current deviation and adding the deviation of the cycle before that by the differential coefficient to obtain the differential term increment. Summing the proportional term increment, integral term increment, and differential term increment gives the control increment.

[0133] The control increment is then added to the historical control quantity retained from the previous control cycle to obtain the unlimited control result; the result is then trimmed based on the upper and lower limits of the blower frequency converter set under the current operating conditions to obtain the final control quantity, which is then written into the blower frequency converter for execution.

[0134] Furthermore, if the real-time raw dissolved oxygen is missing in a certain control cycle, the edge control execution module can temporarily take the actual feedback amount after compensation in the previous effective cycle as the replacement value for the current cycle, while keeping the current operating condition unchanged, so as to avoid abnormal amplification of control increment due to sudden loss of feedback amount.

[0135] If no valid execution feedback is obtained after writing to the blower frequency converter, the execution record for that cycle will be marked as unconfirmed, and the final control quantity for this cycle will be retained for reference when accumulating the amplitude in the next cycle.

[0136] The edge control layer locally adjusts the flushing event judgment results, which are converted into target ranges and switching conditions, based on the compensated actual feedback.

[0137] Among them, the impact load judgment module is used to continuously monitor the flushing valve opening signal and liquid level rise rate in the time-series characteristic data; when the flushing valve opening signal is effective and the liquid level rise rate is greater than or equal to the preset threshold, it is determined that a water impact has occurred; based on the liquid level change during flushing and the system's preset load conversion rate, the target temporary dissolved oxygen increment is determined.

[0138] The start time of pre-aeration is determined based on the target temporary dissolved oxygen increment and the upper limit frequency of the current operating condition, and the shock resistance target range is determined accordingly; the start time of pre-aeration, the shock resistance target range and the exit conditions are combined into the shock resistance control target;

[0139] Among them, the control rule distribution module is used to obtain dynamic compensation parameters and shock resistance control targets; construct the working condition flow rules of normal working condition, pre-aeration working condition, shock resistance working condition and recovery working condition based on the pre-aeration start time and exit conditions, and configure working condition switching conditions and minimum maintenance timeout conditions for each working condition.

[0140] Each operating condition is mapped to the target dissolved oxygen range, and corresponding to the compensation parameter version number, incremental proportional-integral-differential algorithm parameter group, and upper and lower limits of the blower frequency converter.

[0141] The operating condition flow rules, target dissolved oxygen range, and control parameters are encapsulated into a control rule file and distributed to the edge control layer;

[0142] The edge control execution module is used to parse the control rule file, determine the target dissolved oxygen range corresponding to the current operating condition, select the center value of the range as the target dissolved oxygen value, and determine the dynamic compensation parameters; it compensates the real-time original dissolved oxygen based on the dynamic compensation coefficient to obtain the actual feedback amount.

[0143] Based on the deviation between the target dissolved oxygen value and the actual feedback amount, the control increment is calculated using an incremental proportional-integral-derivative algorithm.

[0144] The control increment is accumulated with the historical control quantity of the previous control cycle, and the amplitude is limited according to the upper and lower limits of the blower frequency converter corresponding to the current operating condition to obtain the final control quantity; the final control quantity is written into the blower frequency converter for execution.

[0145] In a preferred embodiment of the present invention, the network disconnection and degradation operation module is used to: continuously monitor the number of consecutive network disconnections and link latency; when the number of consecutive network disconnections or link latency is greater than or equal to a set threshold and continues to time out, it is determined that the network communication interruption state is entered.

[0146] Freeze the current control rule file and effective target dissolved oxygen value, and call the pre-fixed degradation maintenance instruction set. Based on the range of the liquid level data, directly output the preset low frequency value to the blower frequency converter for safe degradation control. After the network is restored, upload the equipment execution record during the communication interruption and wait for the restoration of normal control.

[0147] The operation assessment and parameter update module is used to: after the shock treatment cycle ends, summarize the working condition switching records, actual feedback data, effluent compliance data and total energy consumption data from equipment power metering to form cycle evaluation data;

[0148] The operation score is determined based on the total energy consumption, the level of ammonia nitrogen exceeding the standard in the effluent, and the frequency of operating condition switching; the penalty coefficient and load conversion rate are updated based on the operation score and written back to the cloud scheduling layer.

[0149] When the communication link latency or packet loss rate of the sewage treatment facility in the farm exceeds the preset threshold, even if the edge control layer has received the control rule file, the link latency may continue to increase or the connection may be lost continuously.

[0150] If we wait for the cloud to continue sending the working condition switching results, the blower control remains at the current control stage. Under the premise that the control rule file has been formed and executed by the edge control layer, the network disconnection and degradation operation module continuously monitors the number of consecutive network disconnections and the link latency value. If the number of consecutive network disconnections exceeds the first threshold, or the link latency exceeds the second threshold and the duration exceeds the confirmation window, the system determines that it has entered the network disconnection and communication interruption state.

[0151] The first threshold is set as the ratio of the maximum retransmission time allowed by the system to a single control cycle, rounded down, for example, 3 times. The second threshold is set as 50% of the duration of a single control cycle. The confirmation window is 3 consecutive control cycles to avoid misjudgment caused by occasional out-of-order data packets or instantaneous high latency.

[0152] Once the determination is successful, the edge control layer freezes the currently executing control rule file and its effective target dissolved oxygen value, and no longer executes new cloud-based operating condition switching instructions. At the same time, it calls the pre-frozen degradation maintenance instruction set in the local non-volatile memory.

[0153] This instruction set no longer relies on cloud-based updates. Instead, it directly outputs a preset low-frequency value to the blower inverter based on the range of liquid level data. This serves as a safety degradation control strategy to ensure that the processing unit still has basic aeration capabilities during link anomalies.

[0154] The specific mapping relationship is as follows: the full load depth of the pool is pre-divided into a low liquid level range, a normal liquid level range, and a high liquid level warning range, and each range is uniquely bound to the rated frequency of the equipment in the degradation maintenance instruction set. Proportional control parameters:

[0155] Will Set to the minimum operating frequency to prevent sedimentation. Set to a moderate sustaining frequency corresponding to normal energy consumption, The high-frequency unloading frequency is set to accelerate the processing. The edge control layer can directly match the corresponding preset low-frequency value by comparing the local real-time liquid level with the boundary of the above interval.

[0156] Furthermore, if only a single control cycle delay increases without meeting the confirmation window requirements, the system will continue to execute the original rules and will not immediately switch to the network communication interruption state to avoid frequent state changes caused by short-term network jitter.

[0157] If the liquid level data cannot be updated during the communication interruption, the edge control layer will continue to maintain the corresponding preset low frequency value according to the interval to which the most recent valid liquid level belongs within the first preset time window.

[0158] If the local sensor link fails simultaneously, causing the liquid level data to be missing for a period of time exceeding the physical maximum safe water storage time window calculated based on the tank cross-sectional area and the maximum influent flow rate, for example, 2 hours, the system determines that there is a dual failure state of the sensor and the network. At this time, the edge control layer will forcibly abandon the closed-loop maintenance strategy based on real-time status and directly switch to the lowest level open-loop fixed duty cycle mode, that is, according to the periodic output of a medium maintenance frequency with a fixed duty cycle alternating between 30 minutes of operation and 15 minutes of shutdown, in order to eliminate the risk of overflow or over-aeration caused by the complete failure of the link, until new valid liquid level data is obtained or the network is restored.

[0159] If the degradation maintenance instruction set version in the local non-volatile memory is abnormal, the currently executed valid target dissolved oxygen value will be frozen first, and the reception of unconfirmed new rule files will be prevented, thereby ensuring that at least the same control basis is maintained during the network outage.

[0160] Once the network is restored and the heartbeat verification is passed, the edge control layer first sends up the device execution records during the communication interruption, and then waits for the cloud to reissue the control rule file before resuming normal control, instead of directly using the working condition switching instructions before the network communication interruption.

[0161] Since the changes in liquid level, aeration results and effluent conditions during the communication interruption may have rendered the original judgment of the operating conditions inapplicable, it is necessary for the cloud to reorganize the current control basis based on the complete supplementary data.

[0162] After a shock treatment cycle is completed, the operation assessment and parameter update module summarizes the operating condition switching records, actual feedback data, effluent compliance data and total energy consumption data from the blower and booster pump power metering during the cycle to form cycle evaluation data.

[0163] The system uses total energy consumption, effluent ammonia nitrogen exceedance, and operating condition switching frequency together to determine the operating score;

[0164] After the scoring results are generated, the cloud updates the penalty coefficient and load conversion rate based on the score, and writes the updated results back to the scheduling data to be used in subsequent cycles, so that they can be called when calculating the dynamic compensation parameters and shock resistance control targets in the next round.

[0165] The specific assessment and parameter update logic is as follows: The historical energy consumption baseline is derived from the average total energy consumption of the system in the most recent 30 regular operating cycles in which all effluent water quality meets the standards. The steady-state limit is taken from the total number of normal unidirectional flows under various working conditions such as regular, pre-aeration, shock resistance, and recovery under the standard shock resistance treatment process, such as 4 times. An initial operating score of 100 points is set for this cycle.

[0166] If the total energy consumption exceeds the historical energy consumption baseline, the excess percentage will be deducted in segments, such as 5 points for every 10% excess; if any one instance of ammonia nitrogen exceeding the standard is detected in the effluent water quality, 20 points will be deducted as a penalty; when the actual operating condition switching frequency exceeds the preset steady-state limit, 2 points will be deducted for each additional switch.

[0167] The current cycle's performance score is calculated using the following formula, with a maximum score of 100:

[0168]

[0169] This calculation process deducts multiple penalty values ​​from the full score, among which, This represents the total energy consumption for this cycle. This serves as a historical energy consumption baseline. This represents the floor function; To take the larger value function; The number of times ammonia nitrogen levels in the effluent exceeded the standard; This refers to the actual switching frequency under operating conditions. The preset steady-state limit;

[0170] For example, if the total energy consumption of a certain cycle exceeds the historical energy consumption baseline by about 23%, which is more than 2 times 10%, 10 points will be deducted; if the ammonia nitrogen in the effluent exceeds the standard once, 20 points will be deducted; if the actual operating condition switching frequency is 6 times and the steady state limit is 4 times, which exceeds the limit twice, 4 points will be deducted; then, the operating score calculated by substituting into the formula is 66 points.

[0171] The system calculates the performance score based on a formula, and generates updated data on penalty coefficients and load conversion rates.

[0172] During the final cycle settlement, if the operating score drops to the preset lower limit, such as below 60 points, and the deduction score due to the effluent water quality exceeding the standard is strictly greater than the sum of the deduction scores for the total energy consumption exceeding the standard and the operating condition switching frequency exceeding the standard, then it is determined that the system has insufficient oxygen supply when dealing with water volume shock loads, and the load conversion rate is increased by the preset step size, such as by 10%, to proactively increase the next estimated oxygen supply demand.

[0173] Conversely, if the number of times the ammonia nitrogen exceeds the standard in the effluent during this cycle is 0, and the total energy consumption deduction score is greater than the preset maximum allowable energy consumption tolerance threshold, such as 10 points, it is determined that the system has aeration oxygen supply redundancy. The system will then lower the load conversion rate by the preset step size and raise the penalty coefficient used for the drift compensation module, for example, from the default 1.0 to 1.05, so as to suppress the abnormally large dynamic compensation amount caused by the probe's non-real drift in the next calculation and achieve closed-loop parameter optimization.

[0174] Furthermore, if the effluent compliance data has not been fully returned by the end of the current cycle, the operation assessment and parameter update module can temporarily postpone updating the penalty coefficient and load conversion rate, retain only the evaluation data for this cycle, and perform the update after the relevant data is supplemented;

[0175] If there are local missing data in the total energy consumption data, the score for this period can only generate pending results for the confirmed part, instead of directly overwriting the parameters of the previous period.

[0176] Through the above processing, the system maintains basic control during network anomalies, resumes rule execution after network recovery, and incorporates records generated during communication interruptions and normal operation into periodic evaluations to adjust subsequent compensation parameters and load conversion rates.

[0177] Among them, the network disconnection and degradation operation module is used to continuously monitor the number of consecutive network disconnections and the link latency value; when the number of consecutive network disconnections exceeds the first threshold, or the link latency exceeds the second threshold and the duration exceeds the confirmation window, the system is determined to enter the network disconnection and communication interruption state.

[0178] After entering the network communication interruption state, the currently executing control rule file and its effective target dissolved oxygen value are frozen, the new operating condition switching instructions issued by the cloud are stopped, and the pre-frozen degradation maintenance instruction set in the local non-volatile memory is called.

[0179] Based on the range of liquid level data, a preset low-frequency value is directly output to the blower frequency converter as a safety degradation control strategy; after the network is restored and the heartbeat verification is passed, the device execution record during the communication interruption is sent up, and normal control is restored after the control rule file is reissued.

[0180] The operation assessment and parameter update module is used to summarize the working condition switching records, actual feedback data, effluent compliance data and total energy consumption data from the power metering of blowers and booster pumps after the shock treatment cycle ends, forming periodic evaluation data.

[0181] The operation score is determined based on the total energy consumption, the level of ammonia nitrogen exceeding the standard in the effluent, and the frequency of operating condition switching; the penalty coefficient and load conversion rate are updated based on the operation score and written back to the cloud scheduling layer.

[0182] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A wastewater treatment and data management system for aquaculture environments based on the Internet of Things, characterized in that, The system comprises a cloud-based scheduling layer, an edge control layer, and a field device layer communicating with it. The field device layer includes dissolved oxygen probes, level sensors, blower frequency converters, booster pumps, and flushing valves. The system is equipped with: The data acquisition and processing module is used to acquire equipment acquisition data, including real-time raw dissolved oxygen data, equipment operating condition data, and network status data, and process them to obtain time-series characteristic data. The probe drift compensation module is used to determine dynamic compensation parameters based on the time-series characteristic data and the equipment operating condition data. The impact load judgment module is used to determine the impact resistance control target based on time-series characteristic data; The control rule distribution module is used to generate and distribute control rule files based on the dynamic compensation parameters and the shock resistance control target; The edge control execution module is used to determine the final control quantity based on the control rule file and real-time raw dissolved oxygen data, and to control the blower frequency converter to execute it; The network outage degradation operation module is used to execute a locally fixed set of degradation maintenance instructions when the network connection is lost or the link latency exceeds the limit; The operation assessment and parameter update module is used to update the system assessment parameters based on the system operation records and effluent compliance data; The data acquisition and processing module is used for: Acquire liquid level data collected by the liquid level sensor in the field equipment layer to determine the liquid level rise rate; acquire raw dissolved oxygen data collected by the dissolved oxygen probe; Acquire the current frequency of the blower, the start / stop status of the booster pump, the opening signal of the flushing valve, and network link latency data; The cleaning step difference is determined by the change in dissolved oxygen probe readings before and after probe cleaning, and the corresponding cleaning time markers are recorded. The acquired liquid level data, raw dissolved oxygen data, equipment operating data, liquid level rise rate, network link latency data, cleaning step difference, and cleaning time marker are subjected to noise reduction, missing data completion, and time synchronization processing to obtain time-series feature data. The probe drift compensation module is used for: Historical cleaning step difference data is extracted from the time-series feature data and compared with the current frequency of the blower to determine whether the dissolved oxygen probe is in a drift state. When it is determined that the state is in a drift state, the basic deviation rate is determined based on the time difference between the current time and the most recent cleaning time mark, as well as the correlation between the cleaning step difference and the original dissolved oxygen. The dynamic compensation coefficient is determined by combining the basic deviation rate with the preset initial penalty coefficient or the penalty coefficient of the historical control cycle, and the cleaning time mark, cleaning step difference, cleaning cycle and the penalty coefficient of the previous cycle are combined into dynamic compensation parameters.

2. The IoT-based wastewater treatment and data management system for aquaculture environments according to claim 1, characterized in that, The impact load determination module is used for: Continuously monitor the flush valve opening signal and liquid level rise rate in the time-series characteristic data; when the flush valve opening signal is active and the liquid level rise rate is greater than or equal to the preset threshold, it is determined that a water surge has occurred. The target temporary dissolved oxygen increment is determined based on the liquid level change during flushing and the system's preset load conversion rate. The pre-aeration start time and the shock resistance target range are determined based on the target temporary dissolved oxygen increment and the current operating condition limit frequency, and combined with the preset operating condition exit conditions to form the shock resistance control target.

3. The IoT-based wastewater treatment and data management system for aquaculture environments according to claim 2, characterized in that, The control rule distribution module is used for: Based on the start-up time and exit conditions of pre-aeration, the working condition flow rules for normal working condition, pre-aeration working condition, shock-resistant working condition and recovery working condition are constructed, and switching conditions and minimum maintenance timeout conditions are configured for each working condition. Each operating condition is mapped to the target dissolved oxygen range, and the compensation parameter version number, control algorithm parameter group, and upper and lower limits of the blower frequency converter are configured. The operating condition flow rules, target dissolved oxygen range, and control parameters are encapsulated into a control rule file and sent to the edge control layer.

4. The IoT-based wastewater treatment and data management system for aquaculture environments according to claim 3, characterized in that, The edge control execution module is used for: The control rule file is analyzed to determine the target dissolved oxygen value and dynamic compensation parameters corresponding to the current operating condition; the dynamic compensation parameters are used to compensate the real-time raw dissolved oxygen data to obtain the actual feedback quantity; The control increment is determined based on the deviation between the target dissolved oxygen value and the actual feedback value. The control increment is then combined with the historical control value from the previous control cycle and the upper and lower limits of the blower frequency converter for amplitude limiting. The final control value is then written into the blower frequency converter for execution.

5. The IoT-based wastewater treatment and data management system for aquaculture environments according to claim 1, characterized in that, The network outage and degradation operation module is used for: Continuously monitor the number of consecutive network disconnections and link latency; when the number of consecutive network disconnections or link latency is greater than or equal to the set threshold and continues to time out, it is determined that the network communication is interrupted. Freeze the current control rule file and effective target dissolved oxygen value, and call the pre-fixed degradation maintenance instruction set. Based on the range to which the liquid level data belongs, directly output the preset low frequency value to the blower frequency converter for safe degradation control. After the network is restored, upload the equipment execution record during the communication interruption and wait for the restoration of normal control.

6. The IoT-based wastewater treatment and data management system for aquaculture environments according to claim 1, characterized in that, The operational assessment and parameter update module is used for: After the shock treatment cycle is completed, the cycle evaluation data is formed by summarizing the working condition switching records, actual feedback data, effluent compliance data, and total energy consumption data from equipment power metering. The operation score is determined based on the total energy consumption, the level of ammonia nitrogen exceeding the standard in the effluent, and the frequency of operating condition switching; the penalty coefficient and load conversion rate are updated based on the operation score and written back to the cloud scheduling layer.

Images