A sewage treatment terminal operation state online analysis and early warning system

Abstract

The present application relates to the technical field of industrial control system monitoring, and discloses a sewage treatment terminal operation state online analysis and early warning system. The system acquires a control instruction sequence, an electromechanical execution state sequence and a biochemical reaction state sequence, takes the control instruction generation time as an event reference, performs relative time conversion and time axis matching on the three types of sequences, and divides electromechanical execution response periods and biochemical reaction response periods; an expected response relationship is established according to equipment start delay, biochemical reaction lag period and historical data, the consistency of electromechanical execution response and control instruction and the consistency of biochemical reaction response and electromechanical execution response are sequentially judged, and in combination with the response support state of associated parameters, electromechanical execution abnormalities, biochemical process transmission abnormalities or sensor abnormalities are identified, abnormal equipment, response transmission sections or abnormal sensors are determined, and early warning information is generated. The present application can reduce monitoring misjudgments caused by sampling cycle and response time delay differences, and improve the pertinence of abnormal identification and positioning early warning.

Description

Technical Field

[0001] This invention relates to the field of industrial control system monitoring technology, and in particular to an online analysis and early warning system for the operating status of a wastewater treatment terminal. Background Technology

[0002] With the development of intelligent manufacturing and the digitalization of industrial processes, wastewater treatment terminals are gradually adopting automatic control systems to continuously control and monitor the operation of electromechanical equipment such as pumps and mixers, as well as the biochemical treatment process. The control command signals output by the control system, the electrical execution feedback signals returned by the electromechanical equipment, and the acquisition signals output by the biochemical status sensors collectively reflect the operating status of the wastewater treatment terminal control system and its controlled components.

[0003] Existing technologies typically collect control commands, equipment start / stop status, flow rate, rotational speed, and monitoring parameters such as dissolved oxygen and ammonia nitrogen through programmable logic controllers (PLCs) or monitoring platforms. They then use single-item threshold comparisons, trend analysis, or historical curve comparisons to determine if the equipment and process are abnormal. Some systems also set normal ranges for the status of electromechanical equipment and biochemical indicators, outputting alarm information to the monitoring interface or operation and maintenance terminal when parameters exceed limits.

[0004] However, control commands, electromechanical execution responses, and biochemical reaction responses have different sampling periods and response delays. Existing monitoring methods typically lack relative time conversion based on the same control event, response period division, and cross-layer timing correlation, easily leading to the misclassification of equipment startup delays, biochemical reaction lags, or sensor acquisition deviations as the same type of anomaly. Furthermore, layered, independent threshold alarms struggle to determine the earliest deviation node by following the response sequence from control commands to electromechanical execution and then to biochemical reactions. They also fail to differentiate between electromechanical execution anomalies, biochemical process transmission anomalies, and sensor anomalies, resulting in a lack of specificity in identifying the abnormal object or response transmission segment of the control system and in issuing targeted warnings.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an online analysis and early warning system for the operating status of a wastewater treatment terminal, in order to solve the problems of existing monitoring methods lacking relative time conversion based on the same control event, response period division and cross-layer time sequence correlation, making it difficult to determine the earliest deviation node in the response sequence, and making it difficult to distinguish between electromechanical execution abnormalities, biochemical process transmission abnormalities and sensor abnormalities, resulting in inaccurate determination of abnormal objects or response transmission segments.

[0007] The technical solution of this invention is as follows: an online analysis and early warning system for the operating status of a wastewater treatment terminal, comprising: a status sequence acquisition module, used to acquire control command sequences formed by the control system, electromechanical execution status sequences formed by electromechanical equipment execution feedback, and biochemical reaction status sequences formed by biochemical status sensor acquisition terminals, wherein the three types of sequences respectively include time parameters and object identifiers; a cross-layer time sequence alignment module, used to convert the corresponding electromechanical execution status records and biochemical reaction status records into relative time intervals relative to the event benchmark, based on the control command generation time as the event benchmark, perform time axis matching, and divide the system into a first response period and a second response period; and a response association module, used to extract electromechanical execution response features and biochemical reaction response features from the first response period and the second response period respectively, and to correlate them according to the command issuance time and the electromechanical response start point. The system establishes a time-series response correlation relationship between points, electromechanical response stability nodes, and biochemical response initiation nodes; an expected relationship establishment module is used to divide operating conditions based on normal historical response data and establish equipment start-up delay range, electromechanical execution response range, biochemical reaction hysteresis range, and biochemical reaction response range; a progressive diagnosis module is used to determine the first consistency state between electromechanical execution response and control command and the second consistency state between biochemical reaction response and electromechanical execution response in sequence based on the time-series response correlation relationship and expected response relationship; an anomaly early warning module is used to identify electromechanical execution anomalies, biochemical process transmission anomalies, or sensor anomalies by combining the first consistency state, the second consistency state, and the response support state of associated monitoring parameters, determine the abnormal electromechanical equipment, response transmission segment, or abnormal sensor, and output early warning information.

[0008] Optionally, the state sequence acquisition module is further configured to: read the instruction target value, generation time, and device identifier from the control system; read the device identifier, start / stop status, actual flow rate or actual speed, and acquisition time from the electromechanical equipment execution feedback; and read the biochemical indicators, acquisition time, and sensor identifier from the biochemical state sensor acquisition terminal; associate the three types of records with the same control event according to the preset response correspondence between the device identifier and the biochemical monitoring parameters; and when multiple control events correspond to the same biochemical response observation period, mark the overlapping state of the control events and pause the determination of the second consistency state.

[0009] Optionally, the cross-layer timing alignment module is further configured to: convert the generation time and the acquisition time into the same time unit, respectively calculate the relative time interval between the acquisition time of the electromechanical execution state record and the biochemical reaction state record and the generation time of the corresponding control command; map the two types of state records to the same relative time axis according to the relative time interval, determine the electromechanical response start node and the electromechanical response stability node in the electromechanical execution state sequence to divide the first response period, and determine the biochemical response start node and the biochemical response stability node in the biochemical reaction state sequence to divide the second response period.

[0010] Optionally, the cross-layer timing alignment module is further configured to: determine the device start-up delay based on the time interval between the electromechanical response start node and the control command generation time, and determine the response time fluctuation based on the device start-up delay of continuous control events; when the response time fluctuation exceeds the fluctuation threshold, adjust the boundary of the first response period or the second response period according to each response node of the current control event, so that the adjusted response period covers the corresponding response node, and retain the original acquisition time of each state record.

[0011] Optionally, the response association module is further configured to: extract the actual operating parameters, direction of change, response amplitude, and stable state of the electromechanical system from the first response period; extract the direction of change, response amplitude, and stable state of the biochemical indicators from the second response period; determine the time interval from the time of control command generation to the electromechanical response start node as the equipment startup delay; determine the time interval from the electromechanical response stable node to the biochemical response start node as the biochemical reaction hysteresis period; and establish the time-series response association relationship according to the time sequence of each node.

[0012] Optionally, the response association module is further configured to: determine the timing offset based on the degree of deviation of the device start-up delay and biochemical reaction hysteresis period of the current control event relative to the historical statistical range of the corresponding operating conditions; when the timing offset exceeds the offset range, update the boundary of the corresponding response period according to the parameter with the larger deviation in the device start-up delay and biochemical reaction hysteresis period, and re-extract the electromechanical execution response features and biochemical reaction response features from the updated first response period and second response period to update the timing response association relationship.

[0013] Optionally, the expected relationship establishment module is further configured to: select three types of normal control events with complete state records from the normal historical response data; divide the operating conditions according to the target value of the control command and the initial biochemical state, and determine the equipment start-up delay range, the range of actual electromechanical operating parameters, the range of biochemical reaction hysteresis, and the range of biochemical reaction response for each operating condition; and establish parameter response support relationships based on the response direction and response time of the target biochemical monitoring parameters and associated monitoring parameters in the normal control events.

[0014] Optionally, the progressive diagnostic module is further configured to: compare the current equipment start-up delay, electromechanical execution response characteristics, and actual electromechanical operating parameters with the equipment start-up delay range, electromechanical execution response range, and control command target value of the corresponding operating condition to determine the first consistency state; when the first consistency state is consistent, compare the current biochemical reaction hysteresis period and biochemical reaction response characteristics with the biochemical reaction hysteresis range, biochemical reaction response range, and expected response direction of the corresponding operating condition to determine the second consistency state; when the data is incomplete, the operating condition is not matched, or the control events overlap, the corresponding consistency state is determined to be pending confirmation.

[0015] Optionally, the anomaly warning module is further configured to: determine whether the response deviation of the target biochemical monitoring parameter is supported by the response deviation of the associated monitoring parameter based on the parameter response support relationship, and determine the response support status; identify an electromechanical execution anomaly when the first consistency status is inconsistent; identify a biochemical process transmission anomaly when the first consistency status is consistent, the second consistency status is inconsistent, and the response support status is supported; and identify a sensor anomaly when the first consistency status is consistent, the second consistency status is inconsistent, and the response support status is not supported.

[0016] Optionally, the anomaly warning module is further configured to: identify abnormal electromechanical equipment based on the equipment identifier in the electromechanical execution status record; determine the response transmission segment between the electromechanical response node and the biochemical response node based on the time-series response correlation; identify abnormal sensors or acquisition channels based on the sensor identifier in the biochemical reaction status record; determine the warning level based on the anomaly type, anomaly duration, and response deviation degree; and output warning information containing the anomaly type, abnormal object or response transmission segment, and warning level to the monitoring interface or operation and maintenance terminal.

[0017] Beneficial effects:

[0018] This invention acquires control command signals, electromechanical execution feedback signals, and biochemical state acquisition signals. Using the control command generation time as a benchmark, it performs relative time conversion, time axis matching, and dual-response period division. This allows data with different sampling periods and response delays to form a relationship based on the same control event, reducing monitoring misjudgments caused by response misalignment. By determining equipment start-up delay and biochemical reaction hysteresis, and establishing expected response relationships under operating conditions based on historical data, the consistency between electromechanical execution response and control command, and between biochemical reaction response and electromechanical execution response, is sequentially judged. Response deviations are traced along the sequence of control command, electromechanical execution, and biochemical reaction to the earliest deviation node. By combining two types of consistency states with the response support states of associated monitoring parameters, it distinguishes between electromechanical execution anomalies, biochemical process transmission anomalies, and sensor anomalies. Based on equipment identifiers, response nodes, or sensor identifiers, it identifies abnormal equipment, response transmission sections, or abnormal sensors, improving the targeting of operational anomaly identification and location warning. Attached Figure Description

[0019] Figure 1 A schematic diagram of the composition and data flow of the online analysis and early warning system for the operating status of a wastewater treatment terminal provided in this embodiment of the invention; Figure 2 This is a schematic diagram illustrating the composition of three types of state sequences, the association of control events, and the data validity processing provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of cross-layer relative time axis matching and dual-response time period division provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the response feature extraction, temporal correlation, and response time period boundary update process provided in an embodiment of the present invention. Figure 5 A schematic diagram illustrating the division of operating conditions, expected response relationships, and parameter response support relationships provided for embodiments of the present invention; Figure 6 This is a schematic diagram of the progressive consistency diagnosis, anomaly location, and early warning output process provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0022] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion.

[0023] In this embodiment, "cross-layer" refers to the control command layer, the electromechanical execution feedback layer, and the biochemical state sensing layer. The system uses the control command signals output by the control system, the execution feedback signals returned by the electromechanical equipment, and the acquisition signals output by the biochemical state sensors as monitoring objects to perform online operation monitoring of the wastewater treatment terminal control system and its controlled components, without directly replacing the original control system in performing process control.

[0024] As mentioned earlier, existing wastewater treatment terminal monitoring methods typically perform threshold monitoring or trend judgment on control commands, electromechanical equipment execution status, and biochemical reaction status separately. They lack relative time conversion based on the same control event, response period division, and cross-layer time sequence correlation. Because control commands, electromechanical execution responses, and biochemical reaction responses have different sampling periods and response delays, existing methods easily conflate equipment start-up delays, biochemical reaction lags, or sensor acquisition deviations into the same type of anomaly. It is difficult to determine the earliest deviation node by following the response sequence from control commands to electromechanical execution and then to biochemical reaction, and it is also difficult to distinguish between electromechanical execution anomalies, biochemical process transmission anomalies, and sensor anomalies.

[0025] To address this problem, this invention uses the moment the control command is generated as the event benchmark, converts the electromechanical execution status record and the biochemical reaction status record into relative time intervals relative to the corresponding event benchmark, and divides the time into a first response period and a second response period. Based on each response node, the device start-up delay and biochemical reaction hysteresis period are determined, and the expected response relationship is established by combining normal historical response data. The first consistency state and the second consistency state are judged sequentially, and the anomaly type is identified by combining the response support state of the associated monitoring parameters. The abnormal electromechanical equipment, response transmission section or abnormal sensor is determined and an early warning information is output.

[0026] The following is combined Figures 1 to 6 This invention is described in detail.

[0027] Example 1:

[0028] like Figure 1 As shown, this embodiment of the invention provides an online analysis and early warning system for the operating status of a wastewater treatment terminal, including a status sequence acquisition module, a cross-layer time sequence alignment module, a response association module, an expected relationship establishment module, a progressive diagnosis module, and an anomaly early warning module.

[0029] The system can be configured in the operation monitoring platform, data processing server, or operation and maintenance platform of the wastewater treatment terminal, and establish data connections with the control system, electromechanical equipment execution feedback end, biochemical status sensing and acquisition end, and monitoring interface or operation and maintenance end of the wastewater treatment terminal, respectively.

[0030] The state sequence acquisition module is used to acquire control command sequences, electromechanical execution state sequences, and biochemical reaction state sequences.

[0031] Specifically, the state sequence acquisition module reads the target value, generation time, and corresponding device identifier of the control command from the control system, and records the control commands arranged according to their generation time to form a control command sequence; it reads the device identifier, start / stop status, actual flow rate or actual speed, and acquisition time from the electromechanical equipment execution feedback signal, and records the execution feedback corresponding to the same device identifier to form an electromechanical execution state sequence; it reads dissolved oxygen concentration, ammonia nitrogen concentration, or other biochemical indicators, acquisition time, and sensor identifier from the biochemical state sensor acquisition signal, and records corresponding to the same sensor identifier to form a biochemical reaction state sequence. The three types of sequences are then sent to the cross-layer timing alignment module.

[0032] The cross-layer timing alignment module is used to convert the electromechanical execution state record and biochemical reaction state record into relative time intervals relative to the corresponding event references, using the generation time of each control command as the event reference, and to perform time axis matching according to the relative time intervals to divide the first response period and the second response period respectively.

[0033] Specifically, the cross-layer timing alignment module reads the original timestamps carried by the three types of sequences and converts the control command generation time, electromechanical execution state acquisition time, and biochemical reaction state acquisition time into the same time unit. For each control command, it calculates the time interval between the acquisition time of the corresponding electromechanical execution state record and biochemical reaction state record and the generation time of the control command, and maps the two types of state records to a relative time axis starting from the generation time of the control command.

[0034] The cross-layer timing alignment module determines the electromechanical response start node and electromechanical response stability node in the mapped electromechanical execution state sequence to form the first response period; and determines the biochemical response start node and biochemical response stability node in the mapped biochemical reaction state sequence to form the second response period. The first and second response periods together form the cross-layer timing alignment sequence in chronological order and are sent to the response association module.

[0035] The response association module is used to extract electromechanical execution response features and biochemical reaction response features from the first response period and the second response period, respectively, and to establish a time-series response association relationship according to the time order of each response node.

[0036] Specifically, the response correlation module extracts the start / stop status, the direction of change of actual flow rate or actual rotation speed, the response amplitude, and the stable state from the first response period, and extracts the direction of change of target biochemical indicators, the response amplitude, and the stable state from the second response period. The device start-up delay is determined by the time interval between the control command generation time and the electromechanical response start node time, and the biochemical reaction hysteresis period is determined by the time interval between the electromechanical response stable node time and the biochemical response start node time.

[0037] The response association module associates the control command, electromechanical execution response, and biochemical reaction response according to the sequence of control command generation time, electromechanical response start node, electromechanical response stabilization node, biochemical response start node, and biochemical response stabilization node, forming a time-series response association relationship and sending it to the progressive diagnostic module.

[0038] The expected relationship establishment module is used to divide the operating conditions according to normal historical response data and establish the expected response relationship under each operating condition.

[0039] Specifically, the expected relationship establishment module acquires historical response data generated under normal operating conditions of the wastewater treatment terminal and divides the operating conditions according to the target values ​​of control commands and the corresponding initial biochemical states. For each operating condition, it determines the equipment start-up delay range, the actual electromechanical operating parameter range, the biochemical reaction hysteresis range, and the biochemical reaction response range, and records the time sequence and response correspondence between each range. The resulting expected response relationship is provided to the progressive diagnostic module.

[0040] The progressive diagnostic module is used to determine the first consistency state and the second consistency state sequentially based on the temporal response correlation and the expected response relationship.

[0041] Specifically, the progressive diagnostic module calls the corresponding expected response relationship according to the operating condition to which the current control command belongs, and compares the current equipment start-up delay, electromechanical execution response characteristics and actual electromechanical operating parameters with the equipment start-up delay range, electromechanical execution response range and control command target value to determine the first consistency state.

[0042] When the first consistency state is consistent, the progressive diagnostic module continues to compare the current biochemical reaction hysteresis period and biochemical reaction response characteristics with the biochemical reaction hysteresis range, biochemical reaction response range, and expected response direction under the corresponding operating conditions to determine the second consistency state. When any consistency state indicates a response deviation, the module determines the first deviating node along the time-series response correlation, forming a progressive diagnostic result, and sends the progressive diagnostic result to the anomaly warning module.

[0043] The anomaly warning module is used to identify the anomaly type by combining the first consistency state, the second consistency state, and the response support state, determine the anomaly object or response transmission segment, and output warning information.

[0044] Specifically, the anomaly warning module determines whether the response deviation of the target biochemical monitoring parameter is supported by the response deviation of the associated monitoring parameter based on the response direction and response time between the target biochemical monitoring parameter and the associated monitoring parameter.

[0045] When the first consistency state is inconsistent, it is identified as an electromechanical execution anomaly; when the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is supportive, it is identified as a biochemical process transmission anomaly; when the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is not supportive, it is identified as a sensor anomaly.

[0046] The anomaly warning module identifies abnormal electromechanical equipment based on the equipment identifier in the electromechanical execution status record, determines the response transmission segment between the electromechanical response node and the biochemical response node based on the time sequence response correlation, and identifies abnormal sensors or acquisition channels based on the sensor identifier in the biochemical reaction status record. It determines the warning level based on the anomaly type, anomaly duration, and response deviation, and sends the warning information to the monitoring interface or operation and maintenance terminal.

[0047] Based on the above system, this invention reduces misjudgments caused by different sampling periods and response delays by aligning the control command signals, electromechanical execution feedback signals, and biochemical state sensing signals under the same control event; it determines the earliest deviation node by implementing progressive consistency judgment based on equipment startup delay, biochemical reaction hysteresis period, and normal historical response range; and it distinguishes between electromechanical execution anomalies, biochemical process transmission anomalies, and sensor anomalies by combining the consistency state with the associated parameter response support state, and generates corresponding early warning information.

[0048] Example 2:

[0049] To further illustrate the above system, this embodiment describes the specific processing procedures of each module.

[0050] like Figure 2 As shown, the state sequence acquisition module receives operation records from the control system, the electromechanical equipment execution feedback end, and the biochemical state sensor acquisition end, respectively.

[0051] A control command record in a control command sequence includes at least the target value of the command, the time when the command was generated, and the corresponding device identifier. The target value of the command represents the operating state that the control system expects the corresponding electromechanical equipment to achieve, and can be a target flow rate value, a target speed value, or a start / stop state.

[0052] A state record in an electromechanical execution state sequence must include at least the equipment identifier, start / stop status, actual flow rate or actual rotational speed, and the time of data acquisition. For water pumps with flow feedback, the actual flow rate can be used to characterize the electromechanical execution response; for water pumps, mixers, or other rotating equipment with rotational speed feedback, the actual rotational speed can be used to characterize the electromechanical execution response; for equipment with only state feedback, the time point of start / stop status change can be used to characterize the electromechanical execution response.

[0053] A state record in a biochemical reaction state sequence includes at least the biochemical index value, the acquisition time, and the sensor identifier. The biochemical index may be dissolved oxygen concentration, ammonia nitrogen concentration, or other biochemical indexes corresponding to the current control event.

[0054] The status sequence acquisition module determines the biochemical reaction status record corresponding to the current control command based on the preset response correspondence between the device identifier and the biochemical monitoring parameters. The response correspondence includes at least the device identifier, sensor identifier, expected response direction, and normal response time range, and can be formed based on normal historical control events.

[0055] When multiple control events occur within the same biochemical response observation period, resulting in the same biochemical reaction state record potentially corresponding to multiple control events, the relevant control events are marked as overlapping control events. Control events in the overlapping control event state do not directly output a deterministic second consistency state; instead, they retain the relevant equipment identifiers, control commands, and biochemical reaction state records for verification by the operation and maintenance department.

[0056] To reduce the impact of instantaneous sensor jumps, duplicate communication messages, or missing data on subsequent timing matching, the state sequence acquisition module can perform outlier detection based on recorded parameter values, differences between adjacent samples, sampling intervals, and changes in device status. The allowable fluctuation range for judging abnormal records can be determined based on normal historical data.

[0057] Data marked as anomalous and unable to be verified by preceding and following trusted records will not be used as electromechanical or biochemical response nodes; data that can be recovered through adjacent trusted records can be supplemented with representative values ​​from adjacent trusted records.

[0058] For records lacking the time of command generation, data acquisition time, device identifier, or sensor identifier, the corresponding record will be marked as an incomplete data record. When any of the control command generation time, device identifier, or target value is missing, the corresponding control event can be set to a data insufficiency state, and a deterministic anomaly type will not be automatically output based on that control event.

[0059] like Figure 3 As shown, the cross-layer timing alignment module uses the moment when the control command is generated as the event reference to perform relative time axis conversion on the electromechanical execution state sequence and the biochemical reaction state sequence.

[0060] The relative time interval is used to represent the time difference between the acquisition time of the electromechanical execution status record or the biochemical reaction status record and the generation time of the corresponding control command. The cross-layer timing alignment module converts the three types of time parameters into the same time unit and subtracts the generation time of the control command from the acquisition time of the status record to obtain the corresponding relative time interval.

[0061] The cross-layer timing alignment module sorts the status records according to relative time intervals and retains the original timestamps and relative time intervals of each status record. The relative time conversion is only used to arrange records formed by different sampling periods on the time axis of the same control event, and does not correct for device startup delays or biochemical reaction hysteresis periods as acquisition clock errors.

[0062] The first response period is used to represent the time range within which the electromechanical equipment forms a measurable and executable response after the control command is generated. The starting point of the electromechanical response can be determined by the position where the start-stop state changes, the actual flow rate begins to change continuously, or the actual speed begins to change continuously; the stabilizing point of the electromechanical response can be determined by the actual operating parameters of the electromechanical equipment entering a stable state or entering the normal response range of the corresponding operating condition.

[0063] The second response period represents the time range within which the corresponding biochemical indicators begin to respond and exhibit identifiable changes after the electromechanical execution response is formed. The starting point of the biochemical response can be determined by the position where the target biochemical indicator undergoes a continuous directional change relative to the initial biochemical state; the stabilizing point of the biochemical response can be determined by the target biochemical indicator entering a stable state or entering the biochemical reaction response range of the corresponding operating condition.

[0064] Continuous directional change refers to a change in the target parameter that maintains a consistent direction throughout continuous sampling records, and whose magnitude exceeds the fluctuation range under normal steady-state conditions. For example, three consecutive sampling periods can be used as the criterion for determining continuity, and the normal fluctuation range can be defined as the range formed by the upper and lower standard deviations of the mean of the stable segment under the same operating conditions.

[0065] The cross-layer timing alignment module can also determine the response time fluctuation based on the device startup delay of consecutive control events. For example, the response time fluctuation can be represented by the standard deviation of multiple consecutive device startup delays, and the fluctuation threshold can be determined based on the device startup delay fluctuation range in normal historical control events.

[0066] When the response time fluctuation does not exceed the fluctuation threshold, the first and second response periods are determined according to the current relative time axis. When the response time fluctuation exceeds the fluctuation threshold, the start or end boundary of the first or second response period is adjusted according to the actual response node of the current control event, so that the adjusted response period covers the corresponding response node. The boundary adjustment does not change the original acquisition time and relative time interval of the status record.

[0067] In one implementation, the start time of the electromechanical response is denoted as... The stable node time of the electromechanical response is denoted as The time when the control command is generated is recorded as The device startup delay can be expressed by the following formula:

[0068]

[0069] in, Device startup delay, in seconds; The time of the start of the electromechanical response is in seconds. The time of control command generation is measured in seconds. All time parameters are converted to seconds before calculation.

[0070] The normal range for equipment start-up delay can be determined based on statistical results of the same equipment in normal historical control events (for example, the start-up delay range for pump equipment can be configured to be 1–4 seconds). The calculated... These are used to establish the temporal causal driving chain, determine the first consistent state, and calculate the temporal offset, respectively.

[0071] The moment when the target biochemical indicator begins to produce a sustained response is denoted as . The lag period of a biochemical reaction can be expressed by the following formula:

[0072]

[0073] in, The lag period of a biochemical reaction is measured in seconds. The time of the biochemical response node is in seconds; The time when the electromechanical equipment enters the stable response range is measured in seconds. For biochemical reaction state data recorded in minutes, the minutes are converted to seconds before calculation. The biochemical reaction hysteresis range can be determined based on normal historical control events under the same operating conditions (e.g., a hysteresis mean of 15 minutes and a standard deviation of 3 minutes; 12–18 minutes can be used as an example range for specific judgment). The calculated... It is used to divide the second response period, establish the time sequence response relationship, determine the second consistency state, and calculate the time sequence offset.

[0074] When the biochemical response start point cannot be determined within the preset observation window, the observation window can be extended to the maximum observation duration. For example, the initial observation window can be configured to 30 minutes, and the maximum observation duration can be configured to 2 hours. The specific values ​​are determined based on the normal historical reaction cycle and process rules of the corresponding biochemical indicator. If the biochemical response start point is still not determined after the maximum observation duration is reached, the control event is marked as having no valid biochemical response node, and the second consistency status is determined as pending confirmation.

[0075] like Figure 4 As shown, the response association module extracts the electromechanical response subsequence according to the first response time period and the biochemical response subsequence according to the second response time period.

[0076] For the electromechanical response subsequence, the changes in start / stop status, flow rate, rotational speed, direction of change, time required to reach the target value, and stable value are extracted to form electromechanical execution response characteristics. For the biochemical response subsequence, the initial value, starting point of change, direction of change, amplitude of change, and stable value of the target biochemical index are extracted to form biochemical reaction response characteristics.

[0077] The response association module uses the control event corresponding to the time the control command is generated as the command node, the electromechanical response initiation node and the electromechanical response stability node as the electromechanical response node, and the biochemical response initiation node and the biochemical response stability node as the biochemical response node, and establishes a time-series response association relationship according to the time order of each node. Each node must at least be associated with the object identifier, status value, node time, and the response time period to which it belongs.

[0078] Furthermore, the response correlation module determines the timing offset based on the degree of deviation of the device startup delay and biochemical reaction hysteresis period of the current control event from the historical statistical parameters of the corresponding operating conditions.

[0079] The timing offset can be determined by the following formula:

[0080]

[0081] in, This is the time offset, which is dimensionless. This is the current device startup delay, in seconds. This represents the historical average of the equipment startup delay under the corresponding operating conditions. The historical standard deviation of device startup delay, in seconds; The current lag period of the biochemical reaction is expressed in seconds. The historical average value of the lag period of the biochemical reaction under the corresponding operating conditions is given in seconds. This represents the historical standard deviation of the lag period in a biochemical reaction, in seconds. and These are the influence weights of equipment start-up delay and biochemical reaction hysteresis, respectively, and are dimensionless parameters. + =1. By taking the absolute value of the two standardized deviations, the mutual cancellation of equipment startup delay and biochemical reaction hysteresis when they deviate in opposite directions is avoided. The influence weights can be configured according to the role of different delay parameters in historical anomaly judgment (e.g., =0.60、 =0.40).

[0082] The offset range of the timing offset can be based on normal historical control events. The value distribution is determined. For example, the offset threshold can be configured to 2.50. When If the offset threshold is not exceeded, maintain the boundary of the current response period; when... When the offset threshold is exceeded, the corresponding response time boundary is updated according to the parameter with the larger value among the two standardized deviations.

[0083] When the standardized deviation of the device startup delay is large, adjust the boundary of the first response period; when the standardized deviation of the biochemical reaction hysteresis period is large, adjust the boundary of the second response period; when both deviations exceed the corresponding range, adjust the two response periods respectively. After boundary adjustment, re-extract the electromechanical execution response features and biochemical reaction response features, and update the timing response correlation.

[0084] like Figure 5 As shown, the expected relationship establishment module selects complete normal control events from the normal historical response data, including control command records, electromechanical execution status records, and biochemical reaction status records. Normal operating status can be determined based on equipment alarm records, operation and maintenance records, process status records, or manual verification results.

[0085] The expected relationship establishment module divides operating conditions according to the target value of the control command and the initial biochemical state. Operating conditions can be divided according to preset parameter ranges, or they can be grouped into the same operating condition based on the similarity between the target value of the control command, the initial electromechanical state, and the initial biochemical state.

[0086] For each operating condition, the equipment start-up delay, actual electromechanical operating parameters, biochemical reaction hysteresis period, and biochemical reaction response characteristics are extracted from normal historical control events. Based on the statistical boundaries of normal historical data, the ranges of equipment start-up delay, actual electromechanical operating parameters, biochemical reaction hysteresis, and biochemical reaction response are determined.

[0087] The statistical boundaries can be determined using the upper and lower quantile boundaries, mean and its fluctuation range of normal historical samples, or the normal range confirmed by operation and maintenance rules. This implementation uses the historical statistical range as the expected response relationship and does not rely on a specific machine learning model.

[0088] The expected relationship establishment module also establishes parameter response support relationships between target biochemical monitoring parameters and associated monitoring parameters based on normal historical control events. These parameter response support relationships record at least the target biochemical monitoring parameter, associated monitoring parameter, expected response direction, and normal response time range.

[0089] When historical samples are insufficient to form a valid operating condition, the current control event is marked as an unmatched operating condition. Instead of directly generating a deterministic progressive diagnostic conclusion, the current response data and the information closest to the operating condition are sent to the operation and maintenance terminal.

[0090] like Figure 6 As shown, the progressive diagnosis module and the anomaly warning module sequentially perform the first consistency status judgment, the second consistency status judgment, the response support status judgment, the anomaly type identification, the anomaly object or response transmission segment determination, and the warning output.

[0091] Specifically, the progressive diagnostic module determines the corresponding operating condition based on the current control command target value and the initial biochemical state, and calls the expected response relationship for that operating condition.

[0092] For the current response value to be compared and corresponding normal range Its deviation from the range can be expressed by the following formula:

[0093]

[0094] in, This is the normalized deviation of the current response value from the normal range, and is dimensionless. For equipment start-up delay, actual electromechanical operating parameters, biochemical reaction hysteresis period or biochemical reaction response parameters; and These are the lower and upper boundaries of the normal range for the same parameter, respectively. . , and Use the same units in the calculations. When historical data cannot form a satisfactory... When the effective range is reached, the range deviation calculation is not performed, and the corresponding parameter is set to the range insufficient state.

[0095] The progressive diagnostic module compares the current equipment startup delay, electromechanical execution response characteristics, and actual electromechanical operating parameters with the equipment startup delay range, electromechanical execution response range, and control command target value for the corresponding operating condition. When each necessary parameter is within its corresponding range and the actual equipment operating parameters meet the allowable deviation of the control command target value, the first consistency state is determined to be consistent; when any necessary parameter exceeds its corresponding range, the first consistency state is determined to be inconsistent.

[0096] When the first consistency state is consistent, the progressive diagnostic module continues to compare the current biochemical reaction hysteresis period and biochemical reaction response characteristics with the biochemical reaction hysteresis range, biochemical reaction response range, and expected response direction of the corresponding operating condition. When each necessary parameter is within the corresponding range and the direction of change conforms to the expected direction, the second consistency state is determined to be consistent; when any necessary parameter exceeds the corresponding range, the second consistency state is determined to be inconsistent.

[0097] When there are incomplete data, mismatched operating conditions, no valid biochemical response nodes, or overlapping control events, the corresponding consistency status is determined to be pending confirmation, and the reason for pending confirmation is recorded.

[0098] When the first consistency state is inconsistent, the electromechanical response node that first exceeds the normal range is identified as the deviation starting point, and the subsequent biochemical response deviations are recorded as downstream effects, and are not separately identified as abnormalities in the transmission of biochemical processes.

[0099] When the first consistency state is consistent and the second consistency state is inconsistent, trace from the electromechanical response stable node to the biochemical response initiation node to determine the response node where the biochemical reaction hysteresis exceeds the range or the biochemical reaction parameters exceed the range first.

[0100] The anomaly warning module determines the response support status based on the parameter response support relationship between the target biochemical monitoring parameters and the associated monitoring parameters.

[0101] When the target biochemical monitoring parameter deviates from its response, and at least one associated monitoring parameter deviates in the same direction within the corresponding time range, the response support status is determined to be "supported". When the target biochemical monitoring parameter deviates from its response, while the associated monitoring parameter remains within the normal range, or when the direction of change and response time of the associated monitoring parameter do not support the response deviation, the response support status is determined to be "not supported". When the associated monitoring parameter is missing or multiple associated monitoring parameters give conflicting results, the response support status is determined to be "pending confirmation".

[0102] When the first consistency state is inconsistent, it is identified as an electromechanical execution anomaly; when the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is supportive, it is identified as a biochemical process transmission anomaly; when the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is not supportive, it is identified as a sensor anomaly.

[0103] When any of the first consistency state, the second consistency state, or the response support state is pending confirmation, the corresponding control event will be marked as having insufficient diagnostic information and will not be forcibly classified into the above three types of anomalies.

[0104] After completing the aforementioned progressive consistency diagnosis and anomaly type identification, the anomaly warning module determines the anomaly object or response transmission segment based on the anomaly type. For electromechanical execution anomalies, the abnormal electromechanical equipment is identified based on the equipment identifier in the electromechanical execution status record; for biochemical process transmission anomalies, the response transmission segment between the electromechanical response node and the biochemical response node is determined based on the timing response correlation; for sensor anomalies, the abnormal sensor or acquisition channel is identified based on the sensor identifier in the biochemical reaction status record.

[0105] The anomaly warning module determines the warning level based on the duration of the anomaly and the degree of response deviation. The duration of the anomaly can be determined by the number of consecutive control events of the same anomaly type or the time interval from the first anomaly to the current time. The degree of response deviation can be determined by the maximum range deviation of the corresponding parameter or a representative value of multiple range deviations.

[0106] For example, a Level 1 warning can be configured when the maximum range deviation is no greater than 0.25 and occurs only in one control event; a Level 2 warning can be configured when the maximum range deviation is greater than 0.25 but not greater than 0.50, or when the same abnormality occurs consecutively in two control events; and a Level 3 warning can be configured when the maximum range deviation is greater than 0.50, or when the same abnormality occurs consecutively in three or more control events.

[0107] When the abnormal state recovers to the first consistency state, the second consistency state, and the response support state, all of which meet the corresponding normal conditions, the warning can be lifted after the lifting condition is met in a continuous recovery cycle. For example, the lifting condition can be configured to require two consecutive control events to return to normal, in order to reduce the repeated switching of warnings caused by a single normal sampling point.

[0108] The generated early warning information can include the anomaly type, early warning level, anomaly object or response transmission segment, corresponding control command, the associated node where the response deviation first occurred, the duration of the anomaly, and the degree of response deviation, and is sent to the monitoring interface or operation and maintenance terminal.

[0109] The following uses the flow rate adjustment of the return pump and the corresponding sludge concentration response as an example to illustrate the system operation process, where the values ​​are all example values.

[0110] At any moment The control system generates a reflux pump flow rate adjustment command, with a target flow rate of 40. And record the time when the instruction was generated. The return pump equipment is identified as P01. The status sequence acquisition module obtains the actual flow record and acquisition time corresponding to P01 from the return pump execution feedback end, and obtains the concentration record and acquisition time of the corresponding sludge concentration sensor S01 from the biochemical status sensor acquisition end.

[0111] Cross-layer timing alignment module with Using the event baseline, the collection times of the return pump flow rate records and sludge concentration records were uniformly converted to seconds, and the relative time of each collection time was calculated. The time interval. Time axis matching results show that the actual flow rate of the return pump is within... The value started to rise continuously at +5.8s. +20s to reach the stable range corresponding to the target flow rate; sludge concentration at It started to rise steadily after +18 minutes, and... It takes 45 minutes to reach a stable state. Therefore, the equipment startup delay is 5.8 seconds, and the biochemical reaction hysteresis period is 17 minutes and 40 seconds.

[0112] Assuming that under this operating condition, the average historical equipment startup delay is 2.5s with a standard deviation of 0.6s, and the average historical biochemical reaction lag period is 15min with a standard deviation of 3min, and... =0.60、 =0.40, then the timing offset is approximately 3.66, which is greater than the offset threshold of 2.50. Due to the large standardization deviation of the device startup delay, the response correlation module adjusts the first response period to cover the actual electromechanical response, and re-extracts the flow change amplitude, change direction, and stable flow.

[0113] The expected relationship establishment module is based on the target traffic of 40. The initial sludge concentration before the control command is generated is used to match this control event to the corresponding operating condition. It is assumed that under this condition, the normal range of equipment start-up delay is 1–4 seconds, and the stable flow rate is 38–42 ppm. The biochemical reaction lag range is 12–18 min, and the sludge concentration stability response range is 4000–4300 mg / L. The current stable flow rate is within the actual operating parameters of the electromechanical system, but the equipment startup delay of 5.8 seconds exceeds the normal range; therefore, the first consistency state is inconsistent.

[0114] The progressive diagnostic module identifies the starting node of the reflux pump's electromechanical response as the first associated node to deviate. The anomaly warning module identifies this control event as an electromechanical execution anomaly, determines the abnormal electromechanical equipment as reflux pump P01 based on the equipment identifier P01, and sends a warning message including the equipment identifier, target flow rate, actual stable flow rate, equipment start-up delay, and the first deviation node to the operation and maintenance terminal.

[0115] In another control event, if the reflux pump starts with a delay of 2.8 seconds and the stable flow rate is 40... All conditions were met for electromechanical execution. However, the sludge concentration only began to change continuously 28 minutes after the electromechanical response stabilized, exceeding the hysteresis range of the biochemical reaction. Furthermore, other related biochemical monitoring parameters also showed corresponding deviations in response direction within the corresponding time range. Therefore, the first consistency state was consistent, the second consistency state was inconsistent, and the response support state was supportive. The anomaly warning module identified this control event as an anomaly in the biochemical process transmission and determined the area between the return pump electromechanical response node and the sludge concentration biochemical response node as the response transmission segment, without further defining this segment as a specific biochemical process step or physical location.

[0116] In another control event, if the start-up delay and stable flow rate of the return pump are both within the normal range, but the sludge concentration output by sensor S01 continues to decrease, causing the biochemical reaction response characteristics to exceed the expected response range, while other related monitoring parameters remain within the normal range, then the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is unsupported. The anomaly warning module identifies this control event as a sensor anomaly and determines the abnormal sensor as S01 or its acquisition channel based on the sensor identifier S01.

[0117] Through the above three state branches, this invention starts from control command signals, electromechanical execution feedback signals and biochemical state sensing signals, and determines electromechanical execution anomalies, biochemical process transmission anomalies and sensor anomalies respectively through relative time conversion, response period division, response node association, operating condition matching, progressive consistency judgment and response support judgment, and forms corresponding anomaly objects, response transmission segments and early warning information.

[0118] The above formulas have a continuous relationship between data and application: the equipment startup delay formula and the biochemical reaction hysteresis formula calculate two core time parameters based on the response nodes obtained by time axis matching; the time offset formula uses the two time parameters as input to determine and update the response time period boundaries; the range deviation formula uses the updated extracted current response parameters and the normal historical range as input to determine the first consistency state, the second consistency state, and the degree of response deviation. The outputs of each formula enter subsequent processing steps and do not constitute isolated formulas.

[0119] The above description is merely a specific embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. All modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An online analysis and early warning system for the operating status of a wastewater treatment terminal, characterized in that, include: The state sequence acquisition module is used to acquire the control command sequence formed by the control system, the electromechanical execution state sequence formed by the electromechanical equipment execution feedback, and the biochemical reaction state sequence formed by the biochemical state sensor acquisition terminal. The three types of sequences respectively include time parameters and object identifiers. The cross-layer timing alignment module is used to convert the corresponding electromechanical execution state records and biochemical reaction state records into relative time intervals relative to the event benchmark, based on the time when the control command is generated, to perform time axis matching and to divide the first response period and the second response period. The response association module is used to extract electromechanical execution response features and biochemical reaction response features from the first response period and the second response period, respectively, and to establish a time-series response association relationship based on the instruction issuance time, electromechanical response start node, electromechanical response stability node and biochemical response start node. The expected relationship establishment module is used to classify operating conditions based on normal historical response data, and to establish the equipment start-up delay range, electromechanical execution response range, biochemical reaction hysteresis range, and biochemical reaction response range. The progressive diagnostic module is used to determine, in sequence, the first consistency state between the electromechanical execution response and the control command, and the second consistency state between the biochemical reaction response and the electromechanical execution response, based on the time-series response correlation and the expected response relationship. The anomaly warning module is used to combine the first consistency status, the second consistency status and the response support status of the associated monitoring parameters to identify electromechanical execution anomalies, biochemical process transmission anomalies or sensor anomalies, determine the abnormal electromechanical equipment, response transmission section or abnormal sensor, and output warning information.

2. The system according to claim 1, characterized in that, The state sequence acquisition module is further used for: The system reads the instruction target value, generation time and equipment identifier from the control system; the electromechanical equipment execution feedback reads the equipment identifier, start / stop status, actual flow rate or actual speed and acquisition time from the electromechanical equipment execution feedback; and the biochemical indicators, acquisition time and sensor identifier are read from the biochemical state sensor acquisition terminal. Based on the preset response correspondence between equipment identification and biochemical monitoring parameters, the three types of records are associated with the same control event; When multiple control events correspond to the same biochemical response observation period, the overlapping state of the control events is marked and the determination of the second consistency state is paused.

3. The system according to claim 1, characterized in that, The cross-layer timing alignment module is further used for: The generation time and the acquisition time are uniformly converted into the same time unit, and the relative time interval between the acquisition time of the computer electrical execution state record and the biochemical reaction state record and the generation time of the corresponding control command is calculated. The two types of state records are mapped to the same relative time axis according to the relative time interval. The electromechanical response start node and electromechanical response stability node are determined in the electromechanical execution state sequence to divide the first response period, and the biochemical response start node and biochemical response stability node are determined in the biochemical reaction state sequence to divide the second response period.

4. The system according to claim 3, characterized in that, The cross-layer timing alignment module is further used for: The device start-up delay is determined based on the time interval between the electromechanical response start node and the control command generation time, and the response time fluctuation is determined based on the device start-up delay of continuous control events. When the fluctuation of the response time exceeds the fluctuation threshold, the boundary of the first response period or the second response period is adjusted according to each response node of the current control event so that the adjusted response period covers the corresponding response node, and the original acquisition time of each status record is retained.

5. The system according to claim 1, characterized in that, The response association module block is further used for: The actual operating parameters, direction of change, response amplitude, and stable state of the electromechanical system are extracted from the first response period, and the direction of change, response amplitude, and stable state of the biochemical indicators are extracted from the second response period. The time interval from the moment the control command is generated to the electromechanical response start node is determined as the device startup delay, and the time interval from the electromechanical response stabilization node to the biochemical response start node is determined as the biochemical reaction hysteresis period. The time-series response association is established according to the time order of each node.

6. The system according to claim 5, characterized in that, The response association module is further used for: The timing offset is determined based on the degree of deviation of the current control event’s equipment start-up delay and biochemical reaction hysteresis period from the historical statistical range of the corresponding operating conditions; When the timing offset exceeds the offset range, the boundary of the corresponding response period is updated according to the parameter with the greater deviation between the device startup delay and the biochemical reaction hysteresis period. The electromechanical execution response features and biochemical reaction response features are then re-extracted from the updated first and second response periods to update the timing response correlation.

7. The system according to claim 1, characterized in that, The expected relationship establishment module is further used for: Three types of status records are selected from the normal historical response data to record complete normal control events; The operating conditions are divided according to the target value of the control command and the initial biochemical state. For each operating condition, the equipment start-up delay range, the actual electromechanical operating parameter range, the biochemical reaction hysteresis range, and the biochemical reaction response range are determined. Establish parameter response support relationships based on the response direction and response time of target biochemical monitoring parameters and associated monitoring parameters in normal control events.

8. The system according to claim 7, characterized in that, The progressive diagnostic module is further used for: The current equipment startup delay, electromechanical execution response characteristics, and actual electromechanical operating parameters are compared with the equipment startup delay range, electromechanical execution response range, and control command target value of the corresponding operating conditions to determine the first consistency state. When the first consistency state is consistent, the current biochemical reaction hysteresis period and biochemical reaction response characteristics are compared with the biochemical reaction hysteresis range, biochemical reaction response range and expected response direction of the corresponding operating conditions to determine the second consistency state; When data is incomplete, operating conditions do not match, or control events overlap, the corresponding consistency status will be determined as pending confirmation.

9. The system according to claim 8, characterized in that, The anomaly warning module is further used for: Based on the parameter response support relationship, determine whether the response deviation of the target biochemical monitoring parameter is supported by the response deviation of the associated monitoring parameter, and determine the response support status. When the first consistency state is inconsistent, it is identified as an electromechanical execution anomaly; When the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is supportive, it is identified as an anomaly in the transmission of biochemical processes. When the first consistency state is consistent, the second consistency state is inconsistent, and the response support state is not supported, the sensor is identified as abnormal.

10. The system according to claim 9, characterized in that, The anomaly warning module is further used for: Abnormal electromechanical equipment is identified based on the equipment identifier in the electromechanical execution status record; the response transmission segment between the electromechanical response node and the biochemical response node is determined based on the aforementioned timing response correlation; Identify abnormal sensors or acquisition channels based on the sensor identifiers in the biochemical reaction status record; The warning level is determined based on the anomaly type, duration of the anomaly, and degree of response deviation, and warning information including the anomaly type, anomaly object or response transmission segment and warning level is output to the monitoring interface or operation and maintenance terminal.

Images