Internal recirculation purification control system and method for volatile odors produced by industrial equipment

Abstract

The application discloses an industrial equipment generated volatile odor internal circulation purification control method, relates to the technical field of industrial waste gas treatment, and comprises the following steps: collecting volatile gas and extracting gas control indexes, determining a purification equipment starting state; analyzing a purification benefit ratio and judging stability thereof; judging whether a multi-equipment working condition analysis interval is consistent; if yes, a response curve is constructed to identify an optimal intervention interval; a time prediction model and a multi-equipment collaborative algorithm are constructed from historical data, a time period when the optimal intervention interval appears is predicted, and a collaborative control strategy is formulated. The industrial equipment generated volatile odor internal circulation purification control system comprises a state determination module, a stability analysis module, an interval identification module and a prediction control module. The application realizes efficient purification of industrial volatile odor, and improves system stability and energy-saving effect.

Description

Technical Field

[0001] This invention relates to the field of industrial waste gas treatment technology, specifically to an internal circulation purification control system and method for volatile odors generated by industrial equipment. Background Technology

[0002] In industrial production processes, volatile organic compounds (VOCs) are major pollutants in waste gas and odor, characterized by their high volatility and strong chemical reactivity, seriously affecting environmental quality and human health. Indoor TVOC concentrations must be strictly controlled to ≤0.6 mg / m³, but traditional purification methods face the following technical bottlenecks.

[0003] In industrial settings, multiple devices often experience synchronized fluctuations in operating conditions due to systemic factors such as concentrated emissions and abnormal ventilation (e.g., multiple printing presses operating simultaneously causing a sudden surge in TVOC concentration). However, existing control systems only support independent operation of individual devices and lack mechanisms for calculating the overlap and coordination of operating condition analysis intervals between devices. When pollution outbreaks occur, full-power operation of a single device can easily lead to overload, while remote devices may experience localized concentration exceedances due to response lags, resulting in an imbalance in overall processing capacity. Furthermore, the high peak power consumption of the cluster poses a risk of grid impact.

[0004] Therefore, the present invention provides a control system and method for internal circulation purification of volatile odors generated by industrial equipment. Summary of the Invention

[0005] The purpose of this invention is to provide a control system and method for the internal circulation purification of volatile odors generated by industrial equipment, so as to solve at least one of the above-mentioned problems in the prior art.

[0006] A method for controlling and purifying volatile odors generated by industrial equipment through internal circulation includes the following steps:

[0007] Collect volatile gases generated by industrial equipment and extract gas control indicators, judge and process the gas control indicators, and determine the start-up status of the purification equipment.

[0008] The average judgment index of the purification equipment is collected and curve analysis is performed to construct a purification efficiency group. Stability analysis is performed on the purification efficiency group to obtain the efficiency stability value. Based on the efficiency stability value, it is judged whether the purification efficiency of the purification equipment is stable.

[0009] If there is a discrepancy, extract the operating condition analysis interval, determine whether the operating condition analysis intervals of multiple devices are consistent, and if they are consistent, construct the response curve of the average judgment index and the purification effect, and identify the optimal intervention interval.

[0010] By extracting the time period corresponding to the priority intervention interval from historical data, constructing a time prediction model and a multi-device collaborative algorithm, predicting the time period of the preferred intervention interval, and formulating a collaborative control strategy.

[0011] As a further technical solution of the present invention: the method for determining the start-up state of the purification equipment is as follows:

[0012] The total volatile organic compound (TVOC) concentration was used as a control indicator. The TVOC concentration at the plasma module inlet was collected at multiple monitoring times during the monitoring period to construct a control indicator group.

[0013] Extract the characteristic indicators of the control indicator group and set the equipment start-stop judgment conditions, and determine the start-stop status of the equipment based on the start-stop judgment conditions.

[0014] As a further technical solution of the present invention: the method for obtaining the stable benefit value is as follows:

[0015] The average judgment index of the purification equipment is collected, and a curve showing the change of time versus the average judgment index is established in a two-dimensional coordinate system.

[0016] Obtain the start-up purification time interval of the purification equipment, and analyze the change curves of time and average judgment indicators according to the purification time interval to obtain the purification efficiency ratio.

[0017] A fitting analysis was performed on the purification efficiency ratio to obtain the stable efficiency value of the purification efficiency group.

[0018] As a further technical solution of the present invention: the method for obtaining the purification efficiency ratio is as follows:

[0019] The area of ​​the curves showing the change in calculation time and average judgment index is used as the purification index quantity.

[0020] Obtain the total energy consumption of the purification equipment within the purification time interval, and then calculate the ratio between the purification index and the total energy consumption to obtain the purification efficiency ratio.

[0021] As a further technical solution of the present invention: the method for performing the fitting analysis is as follows:

[0022] The purification efficiency ratios of multiple purification time intervals are combined to construct a purification efficiency group; the purification efficiency group is divided into multiple sub-intervals of different lengths, and the overscaling range of the sub-intervals is calculated.

[0023] Based on the rescaled range, linear regression is used to fit the points, resulting in the fitted line and its slope.

[0024] As a further technical solution of the present invention: the method for determining whether the operating condition analysis intervals of multiple devices are consistent is as follows:

[0025] Obtain the operating condition analysis intervals of multiple devices. If the operating condition analysis intervals of multiple devices overlap, calculate the overlap between the operating condition analysis intervals of any two devices.

[0026] Comparative analysis based on overlap is used to determine whether the operating condition analysis intervals of multiple devices are consistent.

[0027] As a further technical solution of the present invention: the method for identifying the preferred intervention interval is as follows:

[0028] Calculate the ion benefit ratio of the ion energy consumption index curve, and segment the curve based on the ion benefit ratio;

[0029] Obtain the range of device collaboration and perform boundary correction to identify the dynamic valid range;

[0030] Multidimensional constraint calibration is performed on the dynamic effective interval to identify the optimal intervention interval.

[0031] As a further technical solution of the present invention: the method for performing multidimensional constraint calibration is as follows:

[0032] In multidimensional constraint calibration, the boundary and minimum width limit of the efficient segment are calculated, and it is determined whether the dynamic effective interval meets the boundary and minimum width limit of the efficient segment. If it does, the overlapping scene factor correction is performed, and the collaborative correction interval is determined.

[0033] As a further technical solution of the present invention: the method for formulating the collaborative control strategy is as follows:

[0034] Acquire historical monitoring data of industrial gases and preprocess them to extract time features and gas features;

[0035] The time and gas characteristics are input and the time prediction model outputs periodic time windows and real-time probability warnings. When the prediction results coincide, the collaborative control strategy is triggered to predict the time period of the priority intervention interval.

[0036] Based on the time period of the predicted optimal intervention interval, a multi-device collaborative control strategy is formulated.

[0037] The internal circulation purification control system for volatile odors generated by industrial equipment includes the following modules:

[0038] Status determination module: used to collect volatile gases generated by industrial equipment and extract gas control indicators, process the gas control indicators, and determine the start-up status of the purification equipment.

[0039] Stability Analysis Module: Based on the determined start-up conditions, this module collects the average judgment indicators of the purification equipment and performs curve analysis to construct a purification benefit group. It then performs stability analysis on the purification benefit group to obtain a stable benefit value, and judges whether the purification benefit of the purification equipment is stable based on the stable benefit value.

[0040] Interval identification module: If there is a discrepancy, extract the operating condition analysis interval to determine whether the operating condition analysis intervals of multiple devices are consistent. If they are consistent, construct the response curve between the average judgment index and the purification efficiency, and identify the optimal intervention interval.

[0041] Predictive control module: Obtain the time period corresponding to the priority intervention interval from historical data, construct a time prediction model and multi-device collaborative algorithm, predict the time period of the preferred intervention interval, and formulate a collaborative control strategy.

[0042] The beneficial effects of this invention are:

[0043] 1. By setting up a TVOC detection module in the plasma gas circulation purification equipment, the VOC molecule concentration is collected in real time and a control index group is constructed. Start-up and shutdown judgment conditions including average judgment index and rate index are established to realize dynamic control of the start-up and shutdown status of the purification equipment, reduce ineffective operation in low-concentration stable scenarios, reduce equipment losses from frequent start-up and shutdown, and ensure timely intervention in the early stage of pollution concentration rise.

[0044] 2. The area under the concentration change curve within the purification time interval is used as the purification index. Combined with the total energy consumption within the interval, the purification efficiency ratio is calculated. The Hurst index is used to assess the stability of the efficiency. The ion energy consumption index curve is divided into different energy efficiency segments, prioritizing operation within the high-efficiency segment. By constraining the efficiency ratio and filtering with positive rate, inefficient edge data and concentration decrease intervals are eliminated, which helps improve the energy efficiency of the purification equipment.

[0045] 3. By calculating the average overlap of multiple equipment operating condition analysis intervals, equipment groups with consistent operating conditions are selected. For systemic factors such as centralized emissions from industrial equipment and abnormal ventilation, response curves between average judgment indicators and purification benefits are constructed. An initial interval is built based on the mean and standard deviation of the average judgment indicators for each equipment. After calibration using benefit ratio constraints, positive rate filtering, and multi-dimensional constraints, the optimal intervention interval is determined. In multi-equipment collaborative scenarios, staggered startup and dynamic power allocation reduce cluster power consumption while minimizing individual equipment overload, improving the overall balance and reliability of the system's processing capacity.

[0046] 4. A time prediction model is constructed through a dual-model fusion strategy, outputting periodic time windows and real-time probability warnings to predict the occurrence time of priority intervention intervals. Combined with a collaborative objective function, the voltage and airflow parameters of each device are dynamically adjusted to form control strategies such as tiered start-up and load balancing. This provides a predictable and adjustable intelligent solution for industrial waste gas treatment. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 is a structural diagram of the purification device provided by the present invention;

[0049] Figure 2 is a flowchart of the internal circulation purification and control method for volatile odors generated by industrial equipment provided by the present invention;

[0050] Figure 3 is a flowchart of the preferred intervention interval acquisition method provided by the present invention;

[0051] Figure 4 is a block diagram of the internal circulation purification control system for volatile odors generated by industrial equipment provided by the present invention. Detailed Implementation

[0052] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention. Example

[0053] As shown in Figure 1, the method for controlling the internal circulation purification of volatile odors generated by industrial equipment provided in this embodiment of the invention specifically includes the following steps:

[0054] Step 1: Collect volatile gases generated by industrial equipment and extract gas control indicators. Then, process and determine the gas control indicators to ascertain the start-up status of the purification equipment.

[0055] The method for collecting volatile gases generated by industrial equipment and extracting gas light control indicators is as follows:

[0056] A gas detection probe is installed in the plasma gas circulation purification equipment shown in Figure 1. The gas monitoring probe is equipped with a standard TVOC detection module to collect VOC molecule concentration in real time.

[0057] It should be explained that VOC (Volatile Organic Compounds) are a class of organic compounds that easily evaporate from a solid or liquid state to a gaseous state under normal temperature and pressure. They have strong volatility and chemical activity and are one of the main components of industrial waste gas and odor pollution.

[0058] TVOC (Total Volatile Organic Compounds) refers to the concentration of all volatile organic compounds in the air and is used for environmental quality assessment.

[0059] The indoor TVOC concentration limit is ≤0.6mg / m³;

[0060] The total volatile organic compounds (TVOC) concentration was used as a control indicator. The TVOC concentration at the plasma module inlet was collected at multiple monitoring times during the monitoring period to construct a control indicator group.

[0061] Extract the characteristic indicators of the control indicator group and set the equipment start-stop judgment conditions, and determine the start-stop status of the equipment based on the start-stop judgment conditions;

[0062] The method for extracting gas control indicators from the control indicator group is as follows:

[0063] The total volatile organic compounds (TVOC) concentration within the control indicator group was averaged to obtain the average judgment index.

[0064] The concentration change rate is obtained by calculating the concentration change rate of total volatile organic compounds at two adjacent monitoring times within the control indicator group.

[0065] The concentration evaporation rate at multiple monitoring times within the detection period is obtained, and regression fitting analysis is performed to determine the trend of concentration evaporation rate changes.

[0066] If the rate of concentration change is positive, it is used as a positive rate indicator; otherwise, it is used as a negative rate indicator. It should be noted that two adjacent monitoring times refer to monitoring times that are closely connected in the time dimension.

[0067] The average judgment index, positive rate index, and negative rate index are used as the gas control indexes in the control index group.

[0068] Establish equipment start-up and shutdown determination conditions. If the equipment meets any start-up and shutdown determination condition, change the start-up and shutdown status of the purification equipment.

[0069] Preferably, the first judgment condition is that if the average judgment index of the equipment is higher than the upper limit control value, the purification equipment is activated.

[0070] Judgment condition two: If the positive rate index is met and the positive rate is higher than the preset rate control value, then the purification equipment will be started;

[0071] Judgment Condition 3: If the average judgment index is lower than the lower limit control value and is in a negative rate index, the purification equipment shall be temporarily shut down.

[0072] For example, the upper limit control value is 0.6 mg / m³, and the lower limit control value is 0.3 mg / m³;

[0073] It needs to be explained that the gas passes through the low-temperature plasma module shown in Figure 1. Under the action of an external electric field, the low-temperature plasma module generates a large number of energetic electrons through dielectric discharge, which bombard VOC molecules, causing them to ionize, dissociate, and excite a series of physicochemical reactions. This degrades various large-molecule organic waste gases into simpler small-molecule substances, or transforms toxic and harmful substances into non-toxic, harmless, or less harmful substances, thereby degrading and removing VOCs. The treated air is then evenly discharged into the equipment from the top through a high-efficiency filter layer for continuous circulation. The internal gas circulation method can reuse the air in the workshop, avoiding the air conditioning energy consumption caused by the temperature difference between the inside and outside of the workshop when supplementing fresh air. The internal gas circulation can repeatedly treat the VOCs in the air, gradually reducing the concentration, achieving simultaneous volatilization and decomposition, reducing the problem of excessive pressure caused by the instantaneous decomposition due to accumulation. Example

[0074] As shown in Figure 2, the internal circulation purification and control method for volatile odors generated by industrial equipment also includes the following steps: Step 2: Based on the determined start-up conditions, collect the average judgment index of the purification equipment and perform curve analysis to construct a purification benefit group. Perform stability analysis on the purification benefit group to obtain the stable benefit value. Based on the stable benefit value, determine whether the purification benefit of the purification equipment is stable.

[0075] The average judgment index of the purification equipment is collected in real time, and a curve of the change of time and average judgment index is established in a two-dimensional coordinate system.

[0076] The time interval between the start-up and shutdown status of the purification equipment is used as the purification time interval, and the change curves of time and average judgment indicators are extracted according to the purification time interval.

[0077] The area of ​​the curves intersecting the changes in time and average judgment index is used as the purification index; the total energy consumption of the purification equipment within the purification time interval is obtained, and the purification index is compared with the total energy consumption to obtain the purification efficiency ratio.

[0078] The purification efficiency ratio of the purification equipment is obtained for multiple purification time intervals and a stability analysis is performed to determine whether the purification efficiency ratio is stable within multiple purification time intervals.

[0079] The method for determining whether the purification efficiency ratio is stable across multiple purification time intervals is as follows:

[0080] The purification efficiency ratios of multiple purification time intervals are combined to construct a purification efficiency group Xt;

[0081] in xt represents the purification benefit ratio for each of multiple purification time intervals in the purification benefit group, and n is the total number of purification benefit ratios;

[0082] Through the formula: Calculate the cumulative deviation X(t) from the starting purification time interval i to the current purification time interval t for each in the purification benefit group;

[0083] Where is the mean value of the purification benefit ratio;

[0084] Divide the purification benefit group into multiple sub-intervals of different lengths m, and calculate the range R(m) and standard deviation S(m) of the purification benefit ratios within the sub-intervals of length m;

[0085] Perform a ratio processing on the range and standard deviation of each sub-interval to obtain the rescaled range. Calculate the rescaled range corresponding to each sub-interval of different lengths m;

[0086] Taking log(m) as the abscissa and log(R(m) / S(m)) as the ordinate, plot the points of multiple different lengths m in the coordinate system, and use the linear regression method to fit the points to obtain the fitting line and the slope of the fitting line. Take the slope of the fitting line as the Hurst index;

[0087] Take the Hurst index as the benefit stability value of the purification benefit group, and compare the benefit stability value with the preset benefit stability threshold;

[0088] If the benefit stability value is close to the benefit stability threshold, it is considered that the purification benefit ratios within multiple purification time intervals are stable, otherwise they are unstable;

[0089] Those skilled in the art can understand that the benefit stability value (Hurst index) is an important indicator for evaluating the stability of the purification system;

[0090] When 0.45 ≤ H ≤ z, the benefit ratio fluctuation is close to a random walk, the equipment performance is stable and there is no obvious trend, which is a normal random fluctuation and does not require special intervention;

[0091] When 0.55 < H < 0.65 or 0.35 < H < 0.45, it is a mild deviation. H > 0.5 indicates that the benefit ratio has weak persistence (such as a slight upward trend brought by equipment optimization), and H < 0.5 has weak anti-persistence (such as a slow decrease in efficiency caused by slight filter element loss), and attention needs to be paid to the maintenance cycle or algorithm optimization;

[0092] When 0.65 ≤ H ≤ 0.85 or 0.15 ≤ H ≤ 0.35, it is a moderate deviation. H > 0.5 shows a significant continuous high efficiency (it is necessary to check the authenticity of the data), and H < 0.5 shows a significant anti-continuous low efficiency (indicating the precursor of filter element aging or failure), and it is necessary to trigger a maintenance warning and conduct targeted maintenance;

[0093] H>0.85 or H<0.15 indicates a high deviation; H>0.5 may indicate data anomalies (such as sensor misjudgment); H<0.5 usually indicates sudden equipment failure or extreme pollution impact, requiring immediate shutdown and investigation to avoid accident risks.

[0094] If the purification efficiency ratio is unstable within multiple purification time intervals, extract the sub-interval with the largest range of purification efficiency ratios in the purification efficiency group, and obtain the purification time with the lowest purification efficiency ratio in the sub-interval with the largest range as the working condition analysis interval.

[0095] It needs to be explained that the purpose of identifying the operating condition analysis interval is:

[0096] Function 1: Locating systemic pollution factors and triggering multi-equipment collaborative mechanisms. The operating condition analysis interval is a critical period extracted when the purification efficiency ratio is unstable. It is used to identify whether the operating conditions of multiple devices fluctuate synchronously due to common reasons. By calculating the overlap of the operating condition analysis intervals of multiple devices, if the average overlap is higher than a threshold, it is determined that there are systemic factors, triggering the multi-equipment collaborative control process. This reduces overload or response lag caused by independent processing of single devices and improves the overall system's ability to cope with concentrated pollution.

[0097] Secondly, it establishes a data foundation for collaborative control and optimizes intervention strategies. Multi-device data within the operating condition analysis range is the core input for identifying the optimal intervention range. By screening equipment groups with consistent operating conditions and constructing an initial range based on their concentration distribution characteristics, followed by cost-effectiveness constraints, positive rate filtering, and multi-dimensional calibration, the efficient operating range of the plasma module can be located. This shifts the control strategy from single-device empirical thresholds to multi-device data-driven approaches, improving the reliability and energy efficiency of the intervention range.

[0098] The technical solution of this embodiment is as follows: Based on the determined start-up conditions, the average judgment index of the purification equipment is collected and curve analysis is performed to construct a purification benefit group. The purification benefit group is then subjected to stability analysis to obtain a stable benefit value. Based on the stable benefit value, it is determined whether the purification benefit of the purification equipment is stable. Through benefit ratio constraints and positive rate filtering, inefficient edge data and concentration decrease intervals are eliminated, which is beneficial to improving the energy efficiency of the purification equipment. Example

[0099] like Figure 2 As shown, the method for controlling volatile odors generated by industrial equipment through internal circulation purification also includes the following steps:

[0100] Step 3: If there is a discrepancy, extract the operating condition analysis interval and determine whether the operating condition analysis intervals of multiple devices are consistent. If they are consistent, construct the response curve of the average judgment index and the purification efficiency, and identify the optimal intervention interval.

[0101] Preferably, multiple devices upload data to the server via a local area wireless network. The server automatically records the start and stop times, purification time intervals, and the change curves of average judgment indicators, and calculates the operating condition analysis interval.

[0102] The method for determining whether the operating condition analysis intervals of multiple devices are consistent is as follows:

[0103] Obtain the operating condition analysis intervals of multiple devices from the server. If there is an overlap between the operating condition analysis intervals of multiple devices, calculate the overlap between the operating condition analysis intervals of any two devices.

[0104] For example, if the operating condition analysis intervals of two devices are A=[a1,a2] and B=[b1,b2], where a1 and a2 are the start times of the operating condition analysis intervals and b1 and b2 are the end times of the operating condition analysis intervals;

[0105] If A and B have an overlapping portion, i.e. a1>b2 or b1>a2, then the length of the overlapping interval is 0, indicating that the operating condition analysis intervals of the two devices do not overlap.

[0106] If there is overlap, then the overlap length ;

[0107] Through the formula: Get overlap ;

[0108] Where min and max are the minimum and maximum value functions, min(a2, b2) represents selecting the smaller of a2 and b2, and max(a1, b1) represents selecting the smaller of a1 and b1.

[0109] Calculate the average overlap of all devices and compare it with the preset overlap threshold. If the average overlap is greater than or equal to the preset overlap threshold, the working condition analysis intervals of the devices are consistent; otherwise, they are inconsistent.

[0110] It should be explained that if the operating condition analysis range of the equipment is consistent, then there are systematic factors in the operating condition analysis range, such as concentrated emissions from industrial equipment, abnormal workshop ventilation, or common equipment problems.

[0111] If consistent, obtain the average energy consumption and average judgment index of the low-temperature plasma module in multiple monitoring cycles within the operating condition analysis range, and construct the ion energy consumption index curve.

[0112] It needs to be explained that the purpose of calculating overlap is:

[0113] Function 1: Identify the consistency of operating conditions of multiple devices and locate systemic pollution factors. Overlap is used to quantify the degree of time overlap in the analysis interval of multiple device operating conditions. It is the core indicator for judging whether the devices are driven by common factors. When the average overlap of all devices is higher than the preset threshold, it indicates that the start-up and shutdown time and concentration fluctuation trend of multiple devices are highly synchronized, and there is a systemic pollution cause.

[0114] Function 2: Trigger collaborative control strategies to optimize the operating efficiency of multiple devices. By dynamically adjusting voltage and air volume, load balancing is achieved, reducing efficiency loss or lifespan reduction caused by single device overload.

[0115] like Figure 3 As shown, the method for identifying the preferred intervention interval is as follows:

[0116] S1. Calculate the ion benefit ratio of the ion energy consumption index curve, and perform segmentation of the curve based on the ion benefit ratio.

[0117] Through the formula: Obtain the ion efficiency ratio at each time point t of the ion energy consumption index curve. ;

[0118] Among them, C in (j) represents the average judgment index of each point j on the ion energy consumption index curve, E mod (j) represents the average energy consumption at each point j, and Q(j) represents the air volume at each point j. This is an environmental correction function, where T represents temperature and H represents humidity, as defined by an expert in the field. noi For noise correction amount, E mod (j);

[0119] It needs to be explained that C in (j), C in Both (j) and Q(j) are dimensionless and then substituted into the formula to calculate the ion efficiency ratio of the ion energy consumption index curve;

[0120] The environmental correction function is fitted based on experimental data. The purification equipment is tested under different temperatures (T) and humidity levels (H), and the actual purification effect is recorded for the same pollutant concentration. By analyzing the mapping relationship between temperature and humidity changes and purification efficiency, a mathematical model is constructed using regression analysis, neural networks, and other methods to determine the environmental correction function. The specific expression is used to correct the influence of temperature and humidity on the purification process, so that the ion efficiency ratio calculation is more in line with the actual working conditions.

[0121] The determination of the noise correction amount needs to be achieved through experimental statistics and interference analysis. During equipment operation, the degree of interference of equipment mechanical vibration, circuit background noise, and sensor inherent error on the calculation of ion efficiency ratio is statistically analyzed. Through multiple sets of control experiments, the calculation deviations under normal operation and with interference are compared, and the average deviation is calculated to determine the noise correction amount. This is to eliminate the influence of irrelevant factors on the purification efficiency assessment and ensure the accuracy and reliability of the calculation results.

[0122] Based on the ion efficiency ratio of the ion energy consumption index curve, the ion energy consumption index curve is divided into low-efficiency, medium-efficiency and high-efficiency segments.

[0123] S2. Obtain the range of device collaboration and perform boundary correction to identify the dynamic valid range;

[0124] Obtain the overlap of all devices and calculate the mean to obtain the global overlap mean. Based on the global overlap mean, filter the devices and select those with a value higher than the global overlap mean. Based on the average judgment index mean and standard deviation of the devices with the global overlap mean, construct the initial interval. Remove inefficient edge data through the benefit ratio constraint, and then filter out the invalid intervals that are decreasing through the positive rate index to form the dynamic effective interval.

[0125] For example, calculate the global overlap mean and filter the devices: the overlap of the operating intervals of devices A and H are 75%, 60%, 85%, 50%, 70%, 65%, 55%, and 80% respectively;

[0126] The global overlap mean was 67.5%. Devices with an overlap of ≥67.5% were selected: A (75%), C (85%), and G (80%), a total of 3 devices.

[0127] An initial interval was constructed based on the average judgment index: the average judgment indexes of the three devices were 0.8 mg / m³, 1.0 mg / m³, and 0.9 mg / m³, respectively, with a mean μ of 0.9 mg / m³ and a standard deviation σ of 0.1 mg / m³.

[0128] Initial range: [μ-1.5,μ+1.5]=[0.75,1.05] (limited to the high-efficiency range of 0.6-1.2 mg / m³);

[0129] Inefficient data is eliminated based on the efficiency ratio constraint: The peak efficiency ratio of ions in the high-efficiency segment is 3.0, and a threshold of 2.7 is set to eliminate marginal data with an ion efficiency ratio <2.7 within the initial range.

[0130] For example, when the concentration of ions in device A is between 0.75 and 0.8 mg / m³, the ion efficiency ratio is 2.6, and the final retention range shrinks to [0.8, 1.05].

[0131] Forward rate filtering invalid intervals: Filters the initial intervals where the forward rate is higher than the preset rate control value;

[0132] For example, when the average judgment index of equipment C is between 1.0 and 1.05 mg / m³, the forward rate is 0.015, which is lower than the preset rate control value of 0.02 mg / m³. 3 *min, the final dynamic effective interval is [0.8, 1.0];

[0133] S3. Perform multi-dimensional constraint calibration on the dynamic effective interval to identify the optimal intervention interval;

[0134] In multidimensional constraint calibration, the boundary and minimum width limit of the efficient segment are calculated, and it is determined whether the dynamic effective interval meets the boundary and minimum width limit of the efficient segment. If it does, the overlapping scene factor correction is performed, and the collaborative correction interval is determined.

[0135] Finally, by finding the intersection of the dynamic effective interval and the collaborative correction interval, the optimal intervention interval that takes into account the benefit ratio achievement rate, equipment characteristics and system synergy is identified.

[0136] For example, multidimensional constraint calibration is performed on the dynamic effective range [0.8, 1.0 mg / m³] obtained in S2;

[0137] High-efficiency section boundary and minimum width limit: High-efficiency section boundary: 0.6-1.2 mg / m³, then the dynamic effective range is met;

[0138] Minimum width limit: The interval width is required to be ≥0.2mg / m³, and the current dynamic effective interval width is 0.2mg / m³, which meets the condition;

[0139] Correction for factors affecting highly overlapping scenes:

[0140] The overlap of the three devices is >80%, which indicates a strong systemic pollution. The pre-operation mode is activated, and the interval is extended to the left by 0.05 mg / m³. Therefore, the synergistic correction interval is [0.75, 0.95].

[0141] Calculate the intersection of the dynamic effective interval and the collaborative correction interval, and the preferred intervention interval is [0.75, 0.95].

[0142] Step 4: Obtain the time period corresponding to the priority intervention interval from historical data, construct a time prediction model and a multi-device collaborative algorithm, predict the time period of the preferred intervention interval, and formulate a collaborative control strategy.

[0143] The method for constructing the time prediction model is as follows:

[0144] Historical monitoring data of industrial gases is acquired and preprocessed to extract time and gas characteristics, and abnormal data is removed.

[0145] The time characteristics include: the start-up and shutdown time of the purification equipment and the production shifts;

[0146] Gas characteristics include: average judgment index of the priority intervention zone and ion efficiency ratio;

[0147] Then, a dual-model fusion strategy is adopted, using the Prophet model algorithm to capture the periodic patterns of time series, and combining it with the LSTM model algorithm to handle irregular fluctuations and long-term dependencies.

[0148] A time prediction model is constructed using the Rophet model algorithm and the LSTM model algorithm;

[0149] Input time and gas characteristics into the time prediction model, and output periodic time windows and real-time probability warnings through the time prediction model;

[0150] As will be understood by those skilled in the art, when constructing the Prophet model, time series data is decomposed into trend, seasonal, and holiday components. The trend component uses a linear or logistic growth model to describe the long-term trend of the data; the seasonal component uses Fourier series to fit periodic changes (such as daily, weekly, and yearly seasonality); and the holiday component is modeled by a custom list of holidays and their impact magnitude.

[0151] The Prophet model uses historical data to fit and optimize various parameters (such as trend slope, seasonal cycle amplitude, and holiday influence coefficient). When making predictions, future time points are substituted into the model to calculate the superposition value of each component and obtain the final prediction result. It can effectively capture the periodic and trend characteristics of time series.

[0152] When building an LSTM model, the input layer receives historical time-series data (such as feature vectors organized by time steps), and then processes the data through one or more LSTM layers. Each LSTM layer contains memory units, forget gates, input gates, and output gates to capture long-term dependencies and solve the gradient vanishing problem of traditional recurrent neural networks (RNNs). After processing by the LSTM layers, the data enters a fully connected layer, where the activation function outputs the predicted value. During model training, backpropagation algorithms (such as BPTT) are used to adjust the network weights, minimizing the error between the predicted and actual values ​​(such as mean squared error). During prediction, the sequence of historical time steps is input into the model, and the LSTM layers pass information through memory units, progressively calculating and outputting the predicted value for future time points. It excels at handling nonlinear and complex time-series patterns.

[0153] When the prediction results coincide, the collaborative control strategy is triggered to predict the time period of the priority intervention interval;

[0154] Based on the time period of the preferred intervention interval, a multi-device collaborative control strategy is formulated.

[0155] The multi-device collaborative control strategy is as follows:

[0156] Preferably, a collaborative objective function is established: ,in The real-time energy consumption of the purification equipment is given by load(i), where load(i) is the square of the equipment load deviation from the mean, and is the preset weight. These are preset weighting coefficients;

[0157] It needs to be explained that, This is defined for those skilled in the art. m represents the total number of purification devices, which limits the range of the summation operation, and i represents the index of the purification device, used to distinguish different individual purification devices.

[0158] The decision variables are determined based on the collaborative objective function, namely the voltage and air volume of each device, and the device's processing capacity constraints are incorporated (e.g., removal rate ≥ concentration rise rate).

[0159] Model predictive control is used to solve the objective function and find the optimal combination of parameters that balances energy consumption and load within a future time window.

[0160] Real-time data collection of energy consumption of each device Based on the load status, the load deviation load(i) is calculated and substituted into the objective function. The edge controller optimizes the output device operation instructions in real time and adjusts the voltage and air volume of each device.

[0161] The technical solution of this embodiment is as follows: If there is a discrepancy, the operating condition analysis interval is extracted to determine whether the operating condition analysis intervals of multiple devices are consistent. If they are consistent, the response curve of the average judgment index and the purification efficiency is constructed to identify the preferred intervention interval. The time period corresponding to the preferred intervention interval is obtained from historical data, and a time prediction model and a multi-device collaborative algorithm are constructed to predict the time period of the preferred intervention interval and formulate a collaborative control strategy. This provides a predictable and adjustable intelligent solution for industrial waste gas treatment.

[0162] Example 4

[0163] As shown in Figure 4, the internal circulation purification control system for volatile odors generated by industrial equipment includes the following modules:

[0164] Status determination module: used to collect volatile gases generated by industrial equipment and extract gas control indicators, process the gas control indicators, and determine the start-up status of the purification equipment.

[0165] Stability Analysis Module: Based on the determined start-up conditions, this module collects the average judgment indicators of the purification equipment and performs curve analysis to construct a purification benefit group. It then performs stability analysis on the purification benefit group to obtain a stable benefit value, and judges whether the purification benefit of the purification equipment is stable based on the stable benefit value.

[0166] Interval identification module: If there is a discrepancy, extract the operating condition analysis interval to determine whether the operating condition analysis intervals of multiple devices are consistent. If they are consistent, construct the response curve between the average judgment index and the purification efficiency, and identify the optimal intervention interval.

[0167] Predictive control module: Obtain the time period corresponding to the priority intervention interval from historical data, construct a time prediction model and multi-device collaborative algorithm, predict the time period of the preferred intervention interval, and formulate a collaborative control strategy;

[0168] 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 method for purifying and controlling the volatile odor generated in an industrial facility by internal circulation, characterized in that, The method comprises the following steps: extracting total volatile organic matter concentration and performing determination processing to determine the starting state of the purification equipment; collecting average determination indexes of the purification equipment and performing curve analysis to construct a purification benefit group, performing stability analysis on the purification benefit group to obtain a benefit stability value, and judging whether the purification benefit of the purification equipment is stable based on the benefit stability value; the benefit stability value is obtained in the following manner: collecting average determination indexes of the purification equipment, and establishing a change curve of time and average determination indexes in a two-dimensional coordinate system; obtaining a starting purification time interval of the purification equipment, and performing intercept analysis on the change curve of time and average determination indexes according to the purification time interval to obtain a purification benefit ratio; performing fitting analysis on the purification benefit ratio to obtain the benefit stability value; the purification benefit ratio is obtained in the following manner: calculating an intercept area of the change curve of time and average determination indexes as a purification index quantity; obtaining a total energy consumption value of the purification equipment in the purification time interval, performing ratio processing on the purification index quantity and the total energy consumption value to obtain the purification benefit ratio; the fitting analysis is performed in the following manner: combining the purification benefit ratios of multiple purification time intervals to construct a purification benefit group; dividing the purification benefit group into multiple subintervals of different lengths, and calculating a re-scaled range of the subintervals; performing fitting based on the re-scaled range using a linear regression method to obtain a fitting straight line and a slope of the fitting straight line; if not consistent, extracting a working condition analysis interval, judging whether the working condition analysis intervals of multiple devices are consistent, if consistent, constructing a response curve of average determination indexes and purification benefits, and identifying an optimal intervention interval; obtaining a time period corresponding to the optimal intervention interval from historical data, constructing a time prediction model and a multi-device collaboration algorithm, predicting a time period when the optimal intervention interval appears, and formulating a collaborative control strategy.

2. The method according to claim 1, wherein the starting state of the purification equipment is determined in the following manner: taking the total volatile organic matter concentration as a control index, collecting the total volatile organic matter concentration at the inlet of the plasma module at multiple monitoring moments in a monitoring period, and constructing a control index group; extracting feature indexes of the control index group and setting a device start-stop determination condition, and determining the start-stop state of the device based on the start-stop determination condition.

3. The method for internal circulation purification and control of volatile odors generated by industrial equipment according to claim 1, characterized in that, the manner of judging whether the working condition analysis intervals of multiple devices are consistent is as follows: obtaining the working condition analysis intervals of multiple devices, if the working condition analysis intervals of multiple devices have overlapping parts, calculating an overlap degree of the working condition analysis intervals of any two devices; performing comparison analysis based on the overlap degree to judge whether the working condition analysis intervals of multiple devices are consistent.

4. The method for internal circulation purification and control of volatile odors generated by industrial equipment according to claim 1, characterized in that, the manner of identifying the optimal intervention interval is as follows: calculating an ion benefit ratio of an ion energy consumption index curve, performing segmentation processing on the curve based on the ion benefit ratio; obtaining an interval of device collaboration and performing boundary correction to identify a dynamic effective interval; performing multi-dimensional constraint calibration on the dynamic effective interval to identify the optimal intervention interval.

5. The method according to claim 4, wherein the method is characterized by, the manner of performing multi-dimensional constraint calibration is as follows: in the multi-dimensional constraint calibration, the high-efficiency segment boundary and the minimum width limit are calculated, it is judged whether the dynamic effective interval meets the high-efficiency segment boundary and the minimum width limit, if it meets, an overlapping scene factor is corrected to determine a collaborative correction interval.

6. The method of claim 1, wherein the method further comprises: the manner of formulating a collaborative control strategy is as follows: Acquire historical monitoring data of industrial gases and preprocess them to extract time features and gas features; The time and gas characteristics are input and the time prediction model outputs periodic time windows and real-time probability warnings. When the prediction results coincide, the collaborative control strategy is triggered to predict the time period of the priority intervention interval. Based on the time period of the predicted optimal intervention interval, a multi-device collaborative control strategy is formulated.

7. A volatile odor generated by an industrial facility internal circulation purification control system for implementing the volatile odor generated by an industrial facility internal circulation purification control method according to any one of claims 1 to 6, characterized by, Includes the following modules: Status determination module: used to extract the total volatile organic compound concentration and perform judgment processing to determine the start-up status of the purification equipment; Stability Analysis Module: Based on a defined startup state, this module collects average judgment indicators of the purification equipment and performs curve analysis to construct a purification benefit group. It then performs stability analysis on the purification benefit group to obtain a stable benefit value, and uses this stable benefit value to determine whether the purification benefit of the purification equipment is stable. Interval identification module: If there is a discrepancy, extract the operating condition analysis interval to determine whether the operating condition analysis intervals of multiple devices are consistent. If they are consistent, construct the response curve between the average judgment index and the purification efficiency, and identify the optimal intervention interval. Predictive control module: Obtain the time period corresponding to the priority intervention interval from historical data, construct a time prediction model and multi-device collaborative algorithm, predict the time period of the preferred intervention interval, and formulate a collaborative control strategy.

Images