Biomass gas coupling multi-working condition collaborative control method and system

Abstract

The present application belongs to the technical field of industrial process control, and discloses a biomass gas coupling multi-working condition collaborative control method and system; comprising: collecting the inlet pressure, outlet pressure, combustible gas volume flow and heat exchanger outlet temperature of the combustible gas channel, and arranging the pipe section wall temperature measuring point; sorting the pipe section wall temperature measuring point temperature, combining the proportion threshold and the deviation threshold to determine the cold point position and the cold point side temperature; subtracting the control section temperature difference from the cold point side temperature difference to obtain the net temperature difference change; establishing the flow health baseline data set, and obtaining the net resistance growth from the channel pressure difference change; comparing the net temperature difference change and the net resistance growth with the risk threshold to obtain the risk state; executing differentiated treatment and collaborative constraint control based on the risk state; the present application realizes the early identification and effective treatment of tar condensation and deposition risk through the determination of the cold point side temperature, the three-state risk and the collaborative operation constraint control of the linkage of treatment action.

Description

Technical Field

[0001] This invention relates to the field of industrial process control technology, and more specifically, to a biomass gas coupled multi-condition collaborative control method and system. Background Technology

[0002] Biomass gas-coupled boilers are an engineering form that utilizes biomass energy and coal-fired boilers in a coordinated manner. Typically, biomass is converted into combustible gas through a biomass gasification device, and then the combustible gas is introduced into the furnace of a coal-fired boiler to participate in combustion. By adjusting the proportion of combustible gas under different loads, heat input can be supplemented and fuel structure optimized.

[0003] Existing engineering systems typically incorporate heat tracing and insulation into combustible gas channels, and manage the risk of tar condensation and deposition by setting lower limits for combustible gas outlet temperatures, channel differential pressure alarms, and regular purging and cleaning. In terms of operation control, the temperature and flow rate of combustible gas are usually maintained within allowable ranges by adjusting the inlet temperature of the heat exchange medium, the opening degree of the heat exchange bypass, and the opening degree of the combustible gas valve, so as to ensure the normal execution of blending ratio adjustment and boiler load adjustment.

[0004] During multi-condition coordinated operation, the temperature field of combustible gas along the pipeline is prone to non-uniform changes due to the influence of operating condition adjustments and external environmental fluctuations, and short-term low-temperature zones may form in local pipe sections. Tar condensation and adhesion deposition have threshold characteristics; when the local temperature exceeds the condensation condition, tar easily adheres to the pipe wall and gradually grows into a deposition layer. In the early stage of deposition, the impact on the total flow rate and overall pressure difference is often insignificant, and the pressure difference change shows a slow climbing characteristic. Relying solely on the lower limit of a single-point temperature or a single-point pressure difference alarm is insufficient to identify the initial stage of deposition in a timely manner, easily missing the intervention window. When the deposition layer grows to a certain extent, the effective cross-sectional area of ​​the combustible gas channel may rapidly decrease in a short period of time, causing a sudden increase in channel pressure difference, flow pulsation, and unstable gas supply, and may induce adverse consequences such as frequent venting, interlocked load reduction, and even unplanned shutdowns.

[0005] In addition, engineering practices often employ methods such as raising the inlet temperature of the heat exchange medium, adjusting the heat exchange bypass, increasing the heat tracing output, or triggering purging to address the issue. However, these actions themselves can cause short-term fluctuations in temperature and pressure differential. If the risk is determined to be eliminated or upgraded based solely on instantaneous thresholds, it can easily lead to misunderstandings, frequent switching of risk states, or wavering of the handling strategy. This can affect the continuous controllability of the blending ratio and boiler load, and may induce unnecessary downgrading, interlocking, or shutdown risks.

[0006] In view of this, the present invention proposes a biomass gas coupled multi-condition collaborative control method and system to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art, the present invention provides the following technical solution: a biomass gas coupled multi-condition coordinated control method, comprising:

[0008] Collect the inlet pressure, outlet pressure, volumetric flow rate of combustible gas, and outlet temperature of the heat exchanger in the combustible gas channel, and arrange Z pipe section wall temperature measuring points along the combustible gas delivery pipeline;

[0009] The pipe section wall temperature measuring points are sorted and the measuring point numbers of the lowest temperature measuring point and the second lowest temperature measuring point are recorded. The number of times the measuring point is recorded is counted and the low temperature deviation is calculated. The cold point location is determined by combining the proportion threshold and the deviation threshold. The pipe section wall temperature measuring point temperature corresponding to the cold point location is taken as the cold point side temperature.

[0010] Using the heat exchanger outlet temperature as the control temperature, the net temperature difference change is obtained by subtracting the control temperature difference from the cold point temperature difference between adjacent sampling periods.

[0011] A flow health baseline dataset is established based on a stable co-firing window. The current baseline pressure difference and the previous baseline pressure difference are obtained from the health baseline dataset according to the combustible gas volume flow rate of adjacent sampling periods. The difference between the two is calculated, and the net resistance increase is obtained by subtracting the difference between the two from the channel pressure difference change.

[0012] The window statistics are obtained by performing window statistics on the net temperature difference change and the net resistance increase. The window statistics are compared with the risk threshold to determine the risk status of tar condensation and deposition. The risk status includes condensation development state, delayed blockage state and resolution state.

[0013] Differentiated measures and collaborative constraint controls are implemented based on different risk states.

[0014] Furthermore, methods for obtaining the cold spot temperature include:

[0015] Take the most recent consecutive N sampling periods as the sliding time window, sort the temperature of each pipe section wall temperature measuring point from low to high in cycle by cycle within the sliding time window, and record the measuring point number of the lowest temperature measuring point and the second lowest temperature measuring point in each sampling period.

[0016] The measurement point numbers that have appeared as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window are summarized into a candidate measurement point set. The number of times each candidate measurement point is recorded as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window is counted, and cold point determination is performed.

[0017] The cold spot determination condition is: the proportion of the number of times a candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point within the sliding time window is not less than the total number of sampling periods of the sliding time window, and the low temperature deviation of the candidate measuring point is not less than the preset deviation threshold.

[0018] When there is a candidate measurement point in the candidate measurement point set that meets the cold point determination condition, the candidate measurement point is confirmed as the cold point location, and the current sampling period temperature corresponding to the cold point location is taken as the cold point side temperature; if multiple candidate measurement points meet the cold point determination condition at the same time, the one with the largest occurrence ratio is selected as the cold point location; when the occurrence ratios are the same, the candidate measurement point with the larger low temperature deviation is selected as the cold point location.

[0019] Furthermore, the method for obtaining the low-temperature deviation is as follows:

[0020] Calculate the median temperature of the candidate measurement points within each sampling period;

[0021] For each sampling period in which the candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point, calculate the difference between the median temperature of all candidate measuring points in that sampling period and the temperature of the candidate measuring point, and obtain the deviation difference for that sampling period.

[0022] The median value of the deviation difference within the sampling period is taken as the low temperature deviation of the candidate measurement point.

[0023] Furthermore, methods for determining the risk status of tar condensation and deposition include: performing window statistics on the net temperature difference change and net resistance increase; comparing the window statistics with risk thresholds; and using window statistics to obtain corresponding window statistics values.

[0024] Based on the sliding time window, the median value of the change in net temperature difference and the increase in net resistance are taken to obtain the corresponding window statistics.

[0025] Based on historical operating data, risk thresholds for net temperature difference change and net resistance increase are determined and stored in tiers according to the flow range used in the healthy baseline dataset.

[0026] In each sampling period, window statistics of the net temperature difference change and net resistance increase are read and compared with the risk threshold.

[0027] When the window statistical value of the net temperature difference change is not greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is not less than the risk threshold of the net resistance increase, the risk state is determined to enter the condensation development state.

[0028] When the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is still not less than the risk threshold of the net resistance increase, the risk state is determined to enter the delayed blockage state.

[0029] When the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is less than the risk threshold of the net resistance increase, and this is maintained for a preset number of periods, the risk state is determined to be in a de-escalation state.

[0030] Furthermore, methods for implementing differentiated treatment and collaborative constraint control based on different risk states include:

[0031] When the risk state is in the condensation development state, perform preventive thermal boundary programming and activate action blind window and differential verification;

[0032] When the risk state enters the delayed blockage state, a freeze constraint is applied to the blending ratio, keeping the blending ratio setting unchanged and prohibiting its increase; when it is necessary to reduce the risk of gas supply fluctuations, the blending ratio setting is lowered to the preset safety value; in the delayed blockage state, the cleaning and disposal actions are allowed to continue, but the blending ratio cannot be increased again before the freeze constraint is lifted.

[0033] When the risk state enters the de-escalation state and the risk state is determined to be de-escalation state for L consecutive sampling periods, select the historical data segment that has been continuously de-escalated and supplement the flow rate and differential pressure data in the historical data segment into the health baseline dataset.

[0034] Furthermore, when the risk state is in the condensation development state, the methods for performing preventative thermal boundary programming and initiating action blind windows and differential verification include:

[0035] When performing preventive thermal boundary scheduling actions, the current set values ​​of the heat exchange medium inlet temperature, heat exchange bypass opening, and heat tracing circuit output are used as the reference.

[0036] The inlet temperature setpoint of the heat exchange medium is increased by the preset temperature increment. If the increase exceeds the upper limit of the inlet temperature setting, the upper limit of the inlet temperature setting is used as the target inlet temperature value.

[0037] Increase the heat exchange bypass opening setting value by the preset opening increment. If the increase exceeds the upper limit of the opening, then use the upper limit of the opening as the target value of the bypass opening.

[0038] The output setting value of the heat tracing circuit is increased by the preset power increment. If the increase exceeds the upper limit of the heat tracing output setting, the upper limit of the heat tracing output setting is used as the target value of the heat tracing output.

[0039] The time period corresponding to M consecutive sampling cycles from the start of the preventive thermal boundary arrangement action is taken as the action blind window. During the duration of the action blind window, the change in net temperature difference and the increase in net resistance are continuously calculated and recorded, but the risk status adjustment or cancellation judgment is not performed based on the change results. After the action blind window ends, the differential review stage begins.

[0040] Furthermore, differential verification methods include:

[0041] The time period corresponding to the M consecutive sampling cycles after the end of the action blind window is used as the review cycle for judgment;

[0042] When the net resistance increase at the end of the review period is not greater than the net resistance increase at the beginning of the review period, and the median value of the net resistance increase in each sampling period within the review period is less than the risk threshold of the net resistance increase, and the net temperature difference change is higher than the risk threshold of the net temperature difference change, the process of determining the release state is initiated.

[0043] When the change in net temperature difference is higher than the risk threshold for the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold for the increase in net resistance, the risk state is determined to enter the delayed blockage state.

[0044] When the change in net temperature difference is not higher than the risk threshold of the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold of the increase in net resistance, the condensation development state is maintained and the cleaning action is triggered.

[0045] When the net resistance increase at the end of the review period is less than the risk threshold of the net resistance increase but does not meet the conditions for entering the resolution state determination process, the current risk state remains unchanged and differential review continues to be performed in subsequent sampling periods.

[0046] If differential verification continues for more than 2M consecutive sampling cycles after the automatic blind window ends, a differential verification timeout alarm will be output, and the current collaborative operation constraints will remain unchanged.

[0047] Furthermore, the cleaning and disposal actions include:

[0048] First, apply a limiting constraint to the boiler side to ensure that the change in the blending ratio setting value during the cleaning operation does not exceed the preset blending ratio limit, and that the boiler load deviation does not exceed the preset load deviation limit.

[0049] Under the premise of satisfying the aforementioned amplitude limit constraint, the existing purging circuit is triggered to perform channel cleaning action;

[0050] Once the cleaning action begins, the blind window timing is started simultaneously. After the blind window ends, differential verification is performed again.

[0051] When the net resistance increase is consistently below the risk threshold for net resistance increase during the review period, the risk status is allowed to enter the resolution determination process; when the net resistance increase is still not below the risk threshold for net resistance increase during the review period, the output channel is unavailable.

[0052] When the system is in a condensation development state and no output channel is available, a ramp constraint is applied to the blending ratio adjustment and boiler load adjustment to ensure that the adjustment of the blending ratio setpoint in any sampling period does not exceed the product of the preset blending ratio ramp limit and the sampling period duration, and that the adjustment of the boiler load setpoint in any sampling period does not exceed the product of the preset load ramp limit and the sampling period duration; at the same time, it is prohibited to perform blending ratio increase and load increase simultaneously in the same sampling period.

[0053] When the output channel is unavailable, a downgrade switching constraint is executed, causing the blending ratio setting value to be lowered to the preset safety value, and gas supply to the unavailable channel is stopped; when a backup channel or bypass channel is available, the system switches to the backup channel synchronously during the blending ratio reduction process to maintain gas supply continuity; when no backup channel is available, blending is stopped and the heat input is supplemented by coal input from the boiler side.

[0054] Furthermore, the deactivation process includes:

[0055] When the risk state enters the de-escalation state and is maintained for a preset number of periods, it is determined in turn whether two conditions are met: the net temperature difference change is higher than the risk threshold of the net temperature difference change, and the net resistance increase is continuously lower than the risk threshold of the net resistance increase.

[0056] When both conditions are met simultaneously, first release the freeze constraint, then release the downgrade switching constraint, and then release the ramp constraint to restore the blending ratio.

[0057] If either of the two conditions is not met, the existing cooperative operation constraints remain unchanged, and the release state determination process continues to be executed in subsequent sampling cycles until the conditions are met.

[0058] If the release condition is not met for more than 2L consecutive sampling cycles after the first time the release state entry condition is met, the release judgment timeout alarm will be output, and the freeze constraint and ramp constraint will be maintained. When the release judgment timeout alarm is triggered continuously for H times, the cleaning action will be performed. If the release judgment timeout alarm is still triggered after the cleaning action, the output channel will be unavailable.

[0059] A biomass gas coupled multi-condition coordinated control system, implementing the aforementioned biomass gas coupled multi-condition coordinated control method, includes:

[0060] Data acquisition module: Collects the inlet pressure, outlet pressure, volumetric flow rate of combustible gas, and outlet temperature of the heat exchanger in the combustible gas channel, and arranges Z pipe section wall temperature measuring points along the combustible gas delivery pipeline;

[0061] Cold point determination module: Sort the pipe section wall temperature measurement points and record the measurement point numbers of the lowest temperature measurement point and the second lowest temperature measurement point. Count the number of times the measurement point is recorded and calculate the low temperature deviation. Combine the proportion threshold and the deviation threshold to determine the cold point location. Take the pipe section wall temperature measurement point temperature corresponding to the cold point location as the cold point side temperature.

[0062] Temperature difference acquisition module: Using the heat exchanger outlet temperature as the reference section temperature, the net temperature difference change is obtained by subtracting the reference section temperature difference from the cold point temperature difference between adjacent sampling periods.

[0063] Differential pressure acquisition module: Based on the stable co-firing window, a flow health baseline dataset is established. The current baseline differential pressure and the previous baseline differential pressure are obtained from the health baseline dataset according to the combustible gas volume flow rate of adjacent sampling periods. The difference between the two is calculated, and the net resistance increase is obtained by subtracting the difference between the two from the channel differential pressure change.

[0064] Risk assessment module: Window statistics are performed on the net temperature difference change and net resistance increase to obtain the corresponding window statistics value. The window statistics value is compared with the risk threshold to determine the risk status of tar condensation and deposition. The risk status includes condensation development state, delayed blockage state and resolution state.

[0065] Execution module: Implements differentiated actions and collaborative constraint controls based on different risk states.

[0066] The technical effects and advantages of the biomass gas coupled multi-condition coordinated control method and system proposed in this invention are as follows:

[0067] First, this invention collects the inlet and outlet pressures of the combustible gas channel, the volumetric flow rate of the combustible gas, the outlet temperature of the heat exchanger, and the temperatures of multiple pipe section wall temperature measuring points. Within a sliding time window, the pipe section wall temperature measuring points are sorted, and the cold point location is determined based on the proportion threshold and the deviation threshold. The pipe section wall temperature measuring point temperature corresponding to the cold point location is taken as the cold point side temperature. Thus, under multi-condition fluctuations, it can still stably locate the local low-temperature sensitive location and reduce the risk of misjudgment caused by relying solely on the instantaneous lowest value of a single point.

[0068] Secondly, the present invention uses the heat exchanger outlet temperature as the control section temperature. The net temperature difference change is obtained by subtracting the temperature change of the cold point side from the temperature change of the control section in adjacent sampling cycles. This makes the discrimination quantity mainly reflect the additional cooling trend of the cold point side relative to the overall thermal boundary change, weakening the interference of the synchronous rise and fall of the overall temperature on the cold point discrimination, thus making it more suitable for the global temperature drift caused by the incorporation of proportional regulation and load regulation.

[0069] Furthermore, this invention performs preventative thermal boundary scheduling actions after the risk state enters the condensation development state, and sets action blind windows and differential verification to distinguish between short-term responses caused by handling actions and actual risk reduction; at the same time, it implements differentiated handling and collaborative operation constraint control based on the risk state, which suppresses the risk of condensation deposition while limiting the coupling disturbance between the mixing ratio and boiler load, thereby improving the stability of combustible gas supply and boiler-side combustion stability, and reducing the triggering frequency of cleaning and channel isolation handling.

[0070] In summary, this invention achieves early identification and reproducible handling of tar condensation and deposition risks by stabilizing the cold point temperature, decoupling the net temperature difference change and net resistance increase, identifying three-state risks, and coordinating operational constraints with the handling actions, thus balancing risk control effectiveness with multi-condition collaborative operation stability. Attached Figure Description

[0071] Figure 1 This is a flowchart of the biomass gas coupled multi-condition collaborative control method module in Embodiment 1 of the present invention;

[0072] Figure 2 This is a schematic diagram of the cold spot temperature acquisition process in Embodiment 1 of the present invention;

[0073] Figure 3 This is a block diagram of the biomass gas coupled multi-condition collaborative control system in Embodiment 2 of the present invention. Detailed Implementation

[0074] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] Example 1

[0076] See Figure 1 As shown, this embodiment provides a biomass gas coupled multi-condition coordinated control method, including:

[0077] Collect the inlet and outlet pressures of the combustible gas channel, the volumetric flow rate of the combustible gas, and the outlet temperature of the heat exchanger, and arrange Z pipe section wall temperature measuring points along the combustible gas delivery pipeline.

[0078] Pipeline wall temperature measuring points are preferably located at different distances after the heat exchanger outlet, and near bends, valve assemblies, and other locations where localized low-temperature zones are likely to form. Z is a positive integer greater than 0, and preferably not less than 5, to ensure that the temperature distribution along the pipe is observable. Pressure measuring points are preferably set at the outlet of the purification device and the boiler-side inlet. The volumetric flow rate of combustible gas is measured using the existing volumetric flow meter on the boiler.

[0079] Once in actual operation, data will be collected once according to the preset cycle.

[0080] The sampling period is the time interval for performing one data acquisition on temperature, pressure, and flow rate at a preset sampling frequency. The inlet pressure, outlet pressure, channel pressure difference, combustible gas volumetric flow rate, heat exchanger outlet temperature, and wall temperature measurement points for each pipe section are stored sequentially according to the sampling period. During each sampling period, the inlet pressure, outlet pressure, combustible gas volumetric flow rate, heat exchanger outlet temperature, and wall temperature measurement points are read. The channel pressure difference is obtained by subtracting the outlet pressure from the inlet pressure.

[0081] Determine the temperature on the cold spot side within the sliding time window.

[0082] See Figure 2 As shown, the method for obtaining the cold point side temperature is as follows:

[0083] Take the most recent consecutive N (N can be 30) sampling periods as the sliding time window, sort the temperature of each pipe section wall temperature measuring point from low to high in cycle within the sliding time window, and record the measuring point number of the lowest temperature measuring point and the second lowest temperature measuring point in each sampling period.

[0084] The measurement point numbers that have appeared as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window are summarized into a candidate measurement point set. The number of times each candidate measurement point is recorded as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window is counted, and cold point determination is performed.

[0085] The cold spot determination condition is: the proportion of the number of times a candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point within the sliding time window is not less than the total number of sampling periods of the sliding time window, and the low temperature deviation of the candidate measuring point is not less than the preset deviation threshold.

[0086] When there is a candidate measurement point in the candidate measurement point set that meets the cold point determination condition, the candidate measurement point is confirmed as the cold point location, and the current sampling period temperature corresponding to the cold point location is taken as the cold point side temperature; if multiple candidate measurement points meet the cold point determination condition at the same time, the one with the largest occurrence ratio is selected as the cold point location; when the occurrence ratios are the same, the candidate measurement point with the larger low temperature deviation is selected as the cold point location.

[0087] The method for obtaining the low-temperature deviation is as follows:

[0088] Calculate the median temperature of the candidate measurement points within each sampling period;

[0089] For each sampling period in which the candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point, calculate the difference between the median temperature of all candidate measuring points in that sampling period and the temperature of the candidate measuring point, and obtain the deviation difference for that sampling period.

[0090] The median value of the deviation difference within the sampling period is taken as the low temperature deviation of the candidate measurement point.

[0091] For example, the proportion threshold is defined as the number of times a candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point within the sliding time window being at least half of the total number of sampling periods in the sliding time window. The deviation threshold is defined as three times the nominal measurement error of the pipe section wall temperature measuring point.

[0092] For example, Z = 6 wall temperature measuring points, numbered T1 to T6; the sampling period is 1 second, and the sliding time window N = 30. The results of the "lowest temperature measuring point / second lowest temperature measuring point" records in the most recent 30 sampling periods are statistically analyzed: T3 appeared 19 times, T4 appeared 8 times, T2 appeared 3 times, and the remaining measuring points did not appear in the lowest / second lowest records. Therefore, the candidate measuring point set is {T2, T3, T4}. Taking a percentage threshold of 1 / 2, the percentage of occurrences of T3 is 19 / 30 ≈ 0.633 ≥ 1 / 2, which satisfies the percentage condition.

[0093] Further calculation of the low-temperature deviation: Taking one sampling period t as an example, the candidate measurement point temperatures are T2=198.4℃, T3=191.2℃, and T4=197.6℃, with a median temperature of 197.6℃. The deviation difference for this period is 197.6-191.2=6.4℃. For the 19 sampling periods where T3 is recorded as the lowest / second lowest, the deviation difference is calculated and the median value is taken, resulting in a low-temperature deviation of 6.1℃ for T3. If the nominal measurement error of the wall temperature measurement point is ±0.5℃, then the deviation threshold is taken as three times the error, which is 1.5℃. At this point, 6.1℃≥1.5℃, satisfying the deviation condition. Therefore, T3 is confirmed as the cold point location, and the temperature of T3 in the current sampling period is taken as the cold point side temperature.

[0094] As this example demonstrates, the location of the cold spot is constrained by both the probability of its occurrence and the magnitude of its deviation. This can suppress cold spot drift caused by a single, accidental low reading or noise, and anchor the temperature input to the high-risk section of the pipe that is actually undercooled along the pipeline.

[0095] After determining the cold spot temperature, a healthy baseline dataset is obtained by collecting flow rate and differential pressure data through a stable co-firing operation window; the net temperature difference change is calculated based on the control section temperature, and the net resistance increase is calculated based on the healthy baseline dataset.

[0096] Within the operating window where the equipment is in stable combustion and the channels are confirmed to be free of abnormal blockages, a healthy baseline dataset is formed by recording P (preferably 1000) sampling periods of combustible gas volumetric flow rate and channel pressure difference data. The channel pressure difference is obtained by subtracting the outlet pressure from the inlet pressure in the same sampling period. For example, the flow rate is divided into K continuous flow intervals based on a set lower and upper limit for the combustible gas volumetric flow rate, where K is a configurable parameter, preferably 6 to 12. The boundaries of each flow interval can be divided using equal quantile points or equal widths, and each flow interval must contain at least a preset minimum sample size S (e.g., 30) sampling points. For each flow interval, the baseline pressure difference is determined by performing a local linear fit on the combustible gas volumetric flow rate and channel pressure difference data within the interval, using the predicted value at the flow rate center of the interval as the baseline pressure difference. The baseline pressure differences corresponding to each flow interval are written into the healthy baseline dataset by interval index and stored in a lookup table. During operation, the corresponding baseline pressure difference is read using the flow interval into which the current sampling period's volumetric flow rate falls as the index.

[0097] For example, within a stable co-firing window, P = 1000 sampling points are collected. The lower limit of the volumetric flow rate Q is set to 1800 Nm³ / h, and the upper limit is set to 3000 Nm³ / h. K = 6 equal-width intervals are used, with each interval being 200 Nm³ / h, corresponding to the intervals [1800, 2000), [2000, 2200), ..., [2800, 3000]. Let the volumetric flow rate Qt = 2360 Nm³ / h in the current sampling period t, and the volumetric flow rate Qt-1 = 2280 Nm³ / h in the previous sampling period t-1. Then, they fall into the intervals [2200, 2400) and [2200, 2400) respectively.

[0098] Local linear fitting was performed on the samples in the interval [2200, 2400) to obtain the baseline pressure difference model ΔPbase=axQ+b, where the fitting coefficient a=0.004kPa / (Nm³ / h) and the fitting coefficient b=2.10kPa; the predicted value at the center flow rate Qmid=2300Nm³ / h in the interval is ΔPbase(Qmid)=0.004×2300+2.10=11.30kPa, which is written into the baseline pressure difference table for this interval.

[0099] During operation, if the current channel pressure difference ΔPt = 13.00 kPa and the previous cycle ΔPt-1 = 12.10 kPa is measured, then the change in channel pressure difference is ΔPt - ΔPt-1 = 0.90 kPa. Since Qt and Qt-1 fall within the same flow range, the change in baseline pressure difference is 11.30 - 11.30 = 0 kPa. Therefore, the net resistance increase is 0.90 - 0 = 0.90 kPa.

[0100] If the flow rate rises to 2440 Nm³ / h at the next time step t+1 and falls within the range [2400, 2600), the corresponding baseline pressure difference is 12.10 kPa, and ΔPt+1 = 13.40 kPa. Therefore, the channel pressure difference change is 0.40 kPa, the baseline pressure difference change is 12.10 - 11.30 = 0.80 kPa, and the net resistance increase is 0.40 - 0.80 = -0.40 kPa. This shows that the net resistance increase can still deduct the "normal pressure difference increment due to flow rate changes" when the flow rate changes across levels, avoiding misjudging the pressure difference increase caused by flow regulation as an increase in deposition resistance.

[0101] The method for obtaining the net temperature difference change is as follows:

[0102] The heat exchanger outlet temperature is selected as the control temperature, and the control temperature remains constant during operation.

[0103] Using the current sampling period and the previous sampling period as adjacent comparison periods, the temperature change on the cold spot side and the temperature change in the control segment are obtained by subtracting the temperature of the previous sampling period from the temperature of the current sampling period. The net temperature difference change is obtained by subtracting the temperature change in the control segment from the temperature change on the cold spot side.

[0104] The net temperature difference change is used to characterize the additional cooling trend of the cold point side relative to the overall thermal boundary change, so as to reduce the impact of the synchronous rise and fall of the overall temperature on the cold point identification.

[0105] The method for obtaining the net resistance increase is as follows:

[0106] Within the same sampling period, the inlet and outlet pressures are acquired and the channel pressure difference is calculated. The change in channel pressure difference is then calculated using adjacent comparison periods.

[0107] To mitigate baseline pressure fluctuations caused by flow rate changes, the corresponding baseline pressure differences are retrieved from the healthy baseline dataset based on the flow rate of the current sampling period and the flow rate of the previous sampling period. The change in baseline pressure difference is obtained by subtracting the baseline pressure difference of the previous sampling period from the current baseline pressure difference. The net resistance increase is obtained by subtracting the change in baseline pressure difference from the change in channel pressure difference.

[0108] The net resistance increase is used to characterize the trend of additional resistance growth relative to the baseline resistance change.

[0109] In this embodiment, multiple pipe section wall temperature measuring points are arranged along the combustible gas delivery pipeline to obtain the temperature distribution along the pipe wall. The measuring point temperatures of each sampling cycle are sorted within a sliding time window to determine the cold point location, thereby anchoring the temperature input for risk assessment to the pipe section most prone to local overcooling and condensation deposition. The measuring point temperature corresponding to the cold point location is taken as the cold point side temperature, and the temperature of the heat exchanger outlet is taken as the control section temperature. The difference between the temperature changes of adjacent sampling cycles is used to obtain the net temperature difference change. The net temperature difference change mainly reflects the additional cooling trend of the cold point side relative to the overall thermal boundary change, reducing the interference of synchronous temperature rise and fall of the overall operating conditions on the identification of local overcooling. A healthy baseline is established in the stable co-firing window, and the net resistance growth is obtained by subtracting the baseline pressure difference change from the channel pressure difference change. The net resistance growth mainly reflects the additional resistance growth related to deposition and does not change synchronously with the normal pressure difference fluctuations caused by flow regulation. Thus, together with the net temperature difference change, it supports the distinction and judgment of the condensation development state, delayed blockage state, and unblocking state.

[0110] After obtaining the net temperature difference change and net resistance increase, the risk status of tar condensation and deposition is determined and the risk status is output.

[0111] Specific methods for risk status assessment include:

[0112] Based on the sliding time window, the median value of the change in net temperature difference and the increase in net resistance are taken to obtain the corresponding window statistics.

[0113] Based on historical operating data, risk thresholds for net temperature difference change and net resistance increase are determined and stored in tiers according to the flow ranges used in the health baseline dataset.

[0114] In each sampling period, window statistics of the change in net temperature difference and the increase in net resistance are read and compared with the risk threshold.

[0115] When the window statistical value of the net temperature difference change is not greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is not less than the risk threshold of the net resistance increase, the risk state is determined to enter the condensation development state; when the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is still not less than the risk threshold of the net resistance increase, the risk state is determined to enter the delayed blockage state; when the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is less than the risk threshold of the net resistance increase, and this is maintained for a preset number of periods, the risk state is determined to enter the de-escalation state.

[0116] For example, the sliding time window is still N=30, and the median values ​​of the net temperature difference change and the net resistance increase are taken as the window statistics. Let the window statistics corresponding to the current sampling period t be: median value of net temperature difference change ΔTmed = -1.8℃, and median value of net resistance increase ΔRmed = 0.62kPa.

[0117] Within the current flow range [2200, 2400), the corresponding thresholds are: net temperature difference change risk threshold Tthr = -1.0℃ (below this value indicates that the cold spot is still experiencing additional cooling relative to the overall system), and net resistance increase risk threshold Rthr = 0.30kPa. Therefore, ΔTmed = -1.8℃ ≤ -1.0℃ and ΔRmed = 0.62kPa ≥ 0.30kPa, satisfying the entry conditions for the condensation development state.

[0118] If, after the preventive thermal boundary arrangement, the window statistics become ΔTmed = +0.4℃ and ΔRmed = 0.55kPa, then the condition "ΔTmed > Tthr and ΔRmed ≥ Rthr" is satisfied, and the system enters a delayed blockage state. This indicates that the cold point supercooling drive has been relieved, but the additional resistance remains, thus preventing premature release due to temperature rise.

[0119] Furthermore, if the subsequent window statistics become ΔTmed = +0.3℃ and ΔRmed = 0.12kPa after cleaning, and remain continuously for 10 sampling cycles, then the conditions for the risk state are met, and the risk state enters the risk state. Through this example, the risk state can be distinguished into three typical stages: "overcooling-dominated," "resistance-hysteresis," and "dual-indicator recovery," improving the reliability of judging the effectiveness of the response actions.

[0120] The tiered flow ranges are the flow ranges used when establishing the health baseline dataset. During runtime, the corresponding risk threshold is selected based on the flow range into which the volumetric flow rate of the combustible gas in the current sampling period falls.

[0121] When determining the risk thresholds for net temperature difference change and net resistance increase based on historical operating data, a historical data segment within the healthy baseline establishment window is selected for threshold calibration. Within the historical data segment, the minimum statistical value of the net temperature difference change is calculated, and a fixed temperature margin is subtracted from this minimum value to obtain the risk threshold for net temperature difference change. Similarly, the maximum statistical value of the net resistance increase is calculated, and a fixed pressure margin is added to this maximum value to obtain the risk threshold for net resistance increase. The fixed temperature margin and fixed pressure margin are pre-set constants used to cover measurement noise and short-term fluctuations.

[0122] It should be noted that the condensation development state indicates the existence of local low-temperature zones within the channel, where condensation conditions are forming or have already formed, and the increase in flow resistance caused by sediment formation begins to appear; the delayed blockage state indicates that the sediments already formed within the channel are still causing increased resistance or have not yet been effectively removed, with a lag risk of further evolution into blockage and gas supply fluctuations; the unblocked state indicates that the channel is in a recoverable low-risk operating range, showing neither continuous supercooling drive nor continuous growth in additional resistance related to sedimentation.

[0123] Through the above discrimination, the risk status simultaneously reflects the local supercooling trend and the trend of additional resistance growth, and can identify the process where the temperature recovers but the resistance continues to increase, thereby reducing the risk of prematurely resolving the issue based solely on temperature recovery. To avoid repeated switching near the threshold, the entry determination of the condensation development state and the delayed blockage state adopts a continuous determination method, that is, the state jump is triggered when the corresponding comparison relationship is true within at least two consecutive sampling periods; the hold count of the release state is reset to zero and recounted when any comparison relationship does not meet the release state condition.

[0124] Once the risk state enters the condensation development state, a preventative thermal boundary arrangement action is executed.

[0125] Methods for performing preventative thermal boundary choreography include:

[0126] When performing preventive thermal boundary scheduling actions, the heat exchange medium inlet temperature, heat exchange bypass opening, and heat tracing circuit output are all adjusted based on their respective current settings.

[0127] Increase the heat exchange medium inlet temperature setpoint by the preset temperature increment. If the increase exceeds the upper limit of the inlet temperature setting, the upper limit of the inlet temperature setting will be used as the target inlet temperature value. Increase the heat exchange bypass opening setting by the preset opening increment. If the increase exceeds the upper limit of the opening, the upper limit of the opening will be used as the target bypass opening value. Increase the heat tracing circuit output setting by the preset power increment. If the increase exceeds the upper limit of the heat tracing output setting, the upper limit of the heat tracing output setting will be used as the target heat tracing output value.

[0128] The aforementioned preset temperature increment, preset opening increment, and preset power increment are all constants preset before commissioning, and are determined by those skilled in the art based on their own experience with the commissioning and adjustment methods of heat exchange circuits and heat tracing circuits.

[0129] It should be noted that adjusting the inlet temperature of the heat exchange medium is used to increase the temperature level of the combustible gas at the outlet of the heat exchanger; adjusting the opening degree of the heat exchange bypass is used to reduce the heat exchange intensity of the combustible gas within the heat exchanger; and adjusting the output of the heat tracing circuit is used to increase the wall temperature level of the combustible gas delivery pipeline, thereby reducing the probability of localized supercooling of the combustible gas along the pipeline and triggering condensation and deposition. The adjustments to the aforementioned inlet temperature setpoint, bypass opening setpoint, and heat tracing output setpoint are performed incrementally according to the preset maximum rate of change. The preset maximum rate of change is also a constant pre-set before commissioning, and is limited with the sampling period as the execution granularity: within any sampling period, the actual adjustment of each setpoint does not exceed the product of the corresponding factory-set maximum rate of change and the sampling period duration, until the corresponding target value is reached.

[0130] The flow distribution relationship between the heat exchange bypass and the heat exchanger may differ in different projects. This means that increasing the bypass opening may reduce effective heat exchange and increase the heat exchanger outlet temperature in some structures, while in others it may increase effective heat exchange and decrease the heat exchanger outlet temperature. Therefore, the direction of adjusting the heat exchange bypass opening is to increase the heat exchanger outlet temperature. When increasing the heat exchange bypass opening does not increase the heat exchanger outlet temperature, the heat exchange bypass opening is adjusted in the opposite direction by the same increment.

[0131] After the preventative thermal boundary programming begins, start the action blind window timing.

[0132] The time period corresponding to M consecutive sampling cycles (M can be 30) starting from the start of the preventive thermal boundary arrangement action is defined as the action blind window. During the duration of the action blind window, the change in net temperature difference and the increase in net resistance are continuously calculated and recorded, but no risk state adjustment or cancellation judgment is made based on the changes. After the action blind window ends, the differential review stage begins.

[0133] For example, when entering the condensation development state, the heat exchange medium inlet temperature is set to 55°C, the bypass opening is set to 30%, and the heat tracing output is 40%. The preset temperature increment is +3°C, the preset opening increment is +5%, and the preset power increment is +10%, with their respective upper limits of 65°C, 60%, and 80%. After the action is executed, the target values ​​are an inlet temperature of 58°C, a bypass opening of 35%, and a heat tracing output of 50%, which gradually approach the target value over several sampling periods at the maximum rate of change.

[0134] After the action begins, a blind window of M = 30 (corresponding to 30 seconds) is set. Within the blind window, the heat exchanger outlet temperature is observed to briefly rise from 205℃ to 210℃ and then fall back to 206℃. The channel pressure difference exhibits transient fluctuations at the beginning of the action (e.g., rising from 13.0 kPa to 13.8 kPa and then falling back to 13.2 kPa). Since no release / reduction judgment is performed within the blind window, the aforementioned transient improvement or fluctuation will not trigger frequent state switching.

[0135] After the blind window ends, differential verification begins, with the verification period also set to M=30. If the net resistance increase at the beginning of the verification period is 0.48 kPa and at the end of the verification period is 0.22 kPa (the final value is not greater than the initial value), and the median net resistance increase during the verification period is 0.18 kPa < Rthr = 0.30 kPa, while the net temperature difference change is higher than Tthr (e.g., ΔTmed = +0.2℃ > -1.0℃), then the release state determination process begins. This shows that differential verification, using the combined constraints of "the difference between the beginning and end of the period and the median value within the window," can distinguish between short-term responses caused by actions and continuous improvements, thus improving the robustness of the release determination.

[0136] After the blind window for actions ends, perform differential verification.

[0137] In this embodiment, differential verification refers to using the initial and final values ​​of the indicators within a preset verification period as differential benchmarks after the action blind window ends, and combining the window statistics within the verification period to perform secondary verification on the net resistance increase and net temperature difference change. Through the joint constraints of differential trend and statistical threshold, the transient fluctuations caused by the action are distinguished from the continuous improvement such as the relief of cold spots along the process and the actual decline in resistance, thereby serving as a judgment gate for the downgrading, removal or transformation of the risk state.

[0138] Specifically, the time period corresponding to the M consecutive sampling cycles after the end of the action blind window is used as the review cycle for judgment;

[0139] When the net resistance increase at the end of the review period is not greater than the net resistance increase at the beginning of the review period, and the median value of the net resistance increase in each sampling period within the review period is less than the risk threshold of the net resistance increase, and the net temperature difference change is higher than the risk threshold of the net temperature difference change, the process of determining the release state is initiated.

[0140] When the change in net temperature difference is higher than the risk threshold for the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold for the increase in net resistance, the risk state is determined to enter the delayed blockage state.

[0141] When the change in net temperature difference is not higher than the risk threshold of the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold of the increase in net resistance, the condensation development state is maintained and the cleaning action is triggered.

[0142] When the net resistance increase at the end of the review period is less than the risk threshold of the net resistance increase but does not meet the conditions for entering the resolution state determination process, the current risk state remains unchanged and differential review continues to be performed in subsequent sampling periods.

[0143] If differential verification continues for more than 2M consecutive sampling cycles after the automatic blind window ends, a differential verification timeout alarm will be output, and the current collaborative operation constraints will remain unchanged.

[0144] Before the cleaning and disposal action is carried out, a limiting constraint is applied to the boiler side to ensure that the change of the mixing ratio setting value during the cleaning action does not exceed the preset mixing ratio limit and that the boiler load deviation does not exceed the preset load deviation limit.

[0145] Both the preset blending ratio limit and the preset load deviation limit are constants set before commissioning. For example, the preset blending ratio limit is 2% of the difference between the upper limit and the lower limit of the blending ratio setting, and is the smaller of this value and the blending ratio limit value configured in the boiler control. The preset load deviation limit is 1% of the boiler's rated load, and is the smaller of this value and the upper limit of the load deviation configured in the boiler control.

[0146] Under the premise of meeting the above-mentioned limiting constraints, the existing purging circuit is triggered to perform channel cleaning. After the cleaning action begins, the action blind window timer is started simultaneously. After the action blind window ends, differential verification is performed again: when the net resistance increase is stably lower than the risk threshold of net resistance increase within the verification period, the risk status is allowed to enter the de-escalation process; when the net resistance increase is still not lower than the risk threshold of net resistance increase within the verification period, the output channel is unavailable.

[0147] When the system is in a condensation development state and no output channel is available, a ramp constraint is applied to the blending ratio adjustment and boiler load adjustment to ensure that the adjustment of the blending ratio setpoint in any sampling period does not exceed the product of the preset blending ratio ramp limit and the sampling period duration, and that the adjustment of the boiler load setpoint in any sampling period does not exceed the product of the preset load ramp limit and the sampling period duration; at the same time, it is prohibited to perform blending ratio increase and load increase simultaneously in the same sampling period.

[0148] When the output channel is unavailable, a downgrade switching constraint is executed, causing the blending ratio setting value to be lowered to the preset safety value, and gas supply to the unavailable channel is stopped; when a backup channel or bypass channel is available, the system switches to the backup channel synchronously during the blending ratio reduction process to maintain gas supply continuity; when no backup channel is available, blending is stopped and the heat input is supplemented by coal input from the boiler side.

[0149] When the risk condition enters the delayed blockage state, a freeze constraint is applied to the blending ratio, keeping the blending ratio setpoint unchanged and prohibiting its increase. When it is necessary to reduce the risk of gas supply fluctuations, the blending ratio setpoint is lowered to a preset safety value, which is the lower limit of the blending ratio setpoint. Cleaning and disposal actions are allowed to continue in the delayed blockage state, but the blending ratio cannot be increased again until the freeze constraint is lifted.

[0150] The release state determination process includes:

[0151] When the risk state enters the de-escalation state and is maintained for a preset number of periods, it is determined in turn whether two conditions are met: the net temperature difference change is higher than the risk threshold of the net temperature difference change, and the net resistance increase is continuously lower than the risk threshold of the net resistance increase.

[0152] When both conditions are met simultaneously, first release the freeze constraint, then release the downgrade switching constraint, and then release the ramp constraint to restore the blending ratio.

[0153] If either of the two conditions is not met, the existing cooperative operation constraints remain unchanged, and the release state determination process continues to be executed in subsequent sampling cycles until the conditions are met.

[0154] If the release condition is not met for more than 2L consecutive sampling cycles after the first release state entry condition is met, the release judgment timeout alarm will be output, and the freeze constraint and ramp constraint will be maintained. When the release judgment timeout alarm is triggered continuously for H (H can be selected as 2) times, the cleaning action will be performed. If the release judgment timeout alarm is still triggered after the cleaning action, the output channel will be unavailable.

[0155] When the risk state enters the de-escalation state and the risk state is determined to be de-escalation state for a continuous L (L can be selected as 10) sampling periods, select the historical data segment that has been continuously de-escalated and supplement the flow rate and differential pressure data in the historical data segment into the health baseline dataset.

[0156] For example, the number of sampling periods for maintaining the de-escalation state is 10. After the risk state first meets the de-escalation entry conditions, in the subsequent 10 sampling periods, the statistical value of the net temperature difference change window remains between +0.1 and +0.6℃, which is higher than Tthr = -1.0℃; the statistical value of the net resistance increase window remains between 0.05 and 0.18 kPa, which is continuously lower than Rthr = 0.30 kPa. At this time, the freeze constraint, the downgrade switching constraint, and the ramp constraint are released sequentially, restoring the normal adjustment of the incorporation ratio.

[0157] If, during the release process, the net resistance increase window statistical value rises to 0.35 kPa ≥ Rthr in any sampling period, the release state count is reset to zero, and the release state determination process continues. This avoids releasing the constraint before the resistance has stabilized and fallen back, which could lead to an increase in the blending ratio and cause gas supply fluctuations. Through this example, the release state process uses a "continuous holding + two conditions" constraint recovery action to reduce the secondary risks caused by premature release.

[0158] Example 2

[0159] See Figure 3As shown, this embodiment provides a biomass gas coupled multi-condition collaborative control system, and the implementation of the biomass gas coupled multi-condition collaborative control method includes:

[0160] Data acquisition module: Collects the inlet pressure, outlet pressure, volumetric flow rate of combustible gas, and outlet temperature of the heat exchanger in the combustible gas channel, and arranges Z pipe section wall temperature measuring points along the combustible gas delivery pipeline;

[0161] Cold point determination module: Sort the pipe section wall temperature measurement points and record the measurement point numbers of the lowest temperature measurement point and the second lowest temperature measurement point. Count the number of times the measurement point is recorded and calculate the low temperature deviation. Combine the proportion threshold and the deviation threshold to determine the cold point location. Take the pipe section wall temperature measurement point temperature corresponding to the cold point location as the cold point side temperature.

[0162] Temperature difference acquisition module: Using the heat exchanger outlet temperature as the reference section temperature, the net temperature difference change is obtained by subtracting the reference section temperature difference from the cold point temperature difference between adjacent sampling periods.

[0163] Differential pressure acquisition module: Based on the stable co-firing window, a flow health baseline dataset is established. The current baseline differential pressure and the previous baseline differential pressure are obtained from the health baseline dataset according to the combustible gas volume flow rate of adjacent sampling periods. The net resistance increase is obtained by subtracting the difference between the two from the channel differential pressure change.

[0164] Risk assessment module: Window statistics are performed on the net temperature difference change and net resistance increase to obtain the corresponding window statistics value. The window statistics value is compared with the risk threshold to determine the risk status of tar condensation and deposition. The risk status includes condensation development state, delayed blockage state and resolution state.

[0165] Execution module: Implements differentiated actions and collaborative constraint controls based on different risk states.

[0166] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0167] In conclusion, the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A biomass gas coupled multi-condition coordinated control method, characterized in that, include: Collect the inlet pressure, outlet pressure, volumetric flow rate of combustible gas, and outlet temperature of the heat exchanger in the combustible gas channel, and arrange Z pipe section wall temperature measuring points along the combustible gas delivery pipeline; The pipe section wall temperature measuring points are sorted and the measuring point numbers of the lowest temperature measuring point and the second lowest temperature measuring point are recorded. The number of times the measuring point is recorded is counted and the low temperature deviation is calculated. The cold point location is determined by combining the proportion threshold and the deviation threshold. The pipe section wall temperature measuring point temperature corresponding to the cold point location is taken as the cold point side temperature. Using the heat exchanger outlet temperature as the control temperature, the net temperature difference change is obtained by subtracting the control temperature difference from the cold point temperature difference between adjacent sampling periods. A flow health baseline dataset is established based on a stable co-firing window. The current baseline pressure difference and the previous baseline pressure difference are obtained from the health baseline dataset according to the combustible gas volume flow rate of adjacent sampling periods. The difference between the two is calculated, and the net resistance increase is obtained by subtracting the difference between the two from the channel pressure difference change. The window statistics are obtained by performing window statistics on the net temperature difference change and the net resistance increase. The window statistics are compared with the risk threshold to determine the risk status of tar condensation and deposition. The risk status includes condensation development state, delayed blockage state and resolution state. Differentiated measures and collaborative constraint controls are implemented based on different risk states.

2. The biomass gas coupled multi-condition coordinated control method according to claim 1, characterized in that, The method for obtaining the cold spot temperature includes: Take the most recent consecutive N sampling periods as the sliding time window, sort the temperature of each pipe section wall temperature measuring point from low to high in the sliding time window, and record the measuring point number of the lowest temperature measuring point and the second lowest temperature measuring point in each sampling period. The measurement point numbers that have appeared as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window are summarized into a candidate measurement point set. The number of times each candidate measurement point is recorded as the lowest temperature measurement point or the second lowest temperature measurement point within the sliding time window is counted, and cold point determination is performed. The cold spot determination condition is: the proportion of the number of times a candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point within the sliding time window is not less than the total number of sampling periods of the sliding time window, and the low temperature deviation of the candidate measuring point is not less than the preset deviation threshold. When there is a candidate measurement point in the candidate measurement point set that meets the cold point determination condition, the candidate measurement point is confirmed as the cold point location, and the current sampling period temperature corresponding to the cold point location is taken as the cold point side temperature; if multiple candidate measurement points meet the cold point determination condition at the same time, the one with the largest occurrence ratio is selected as the cold point location; when the occurrence ratios are the same, the candidate measurement point with the larger low temperature deviation is selected as the cold point location.

3. The biomass gas coupled multi-condition coordinated control method according to claim 2, characterized in that, The method for obtaining the low-temperature deviation is as follows: Calculate the median temperature of the candidate measurement points within each sampling period; For each sampling period in which the candidate measuring point is recorded as the lowest temperature measuring point or the second lowest temperature measuring point, calculate the difference between the median temperature of all candidate measuring points in that sampling period and the temperature of the candidate measuring point, and obtain the deviation difference value of that sampling period. The median value of the deviation difference within the sampling period is taken as the low temperature deviation of the candidate measurement point.

4. The biomass gas coupled multi-condition coordinated control method according to claim 1, characterized in that, The method for determining the risk status of tar condensation and deposition by performing window statistics on the net temperature difference change and net resistance increase to obtain corresponding window statistical values, and comparing the window statistical values ​​with risk thresholds, includes: Based on the sliding time window, the median value of the change in net temperature difference and the increase in net resistance are taken to obtain the corresponding window statistics. Based on historical operating data, risk thresholds for net temperature difference change and net resistance increase are determined and stored in tiers according to the flow range used in the healthy baseline dataset. In each sampling period, window statistics of the net temperature difference change and net resistance increase are read and compared with the risk threshold. When the window statistical value of the net temperature difference change is not greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is not less than the risk threshold of the net resistance increase, the risk state is determined to enter the condensation development state. When the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is still not less than the risk threshold of the net resistance increase, the risk state is determined to enter the delayed blockage state. When the window statistical value of the net temperature difference change is greater than the risk threshold of the net temperature difference change and the window statistical value of the net resistance increase is less than the risk threshold of the net resistance increase, and this is maintained for a preset number of periods, the risk state is determined to be in a de-escalation state.

5. The biomass gas coupled multi-condition coordinated control method according to claim 1, characterized in that, The method for implementing differentiated treatment and collaborative constraint control based on different risk states includes: When the risk state is in the condensation development state, perform preventive thermal boundary programming and activate action blind window and differential verification; When the risk state enters the delayed blockage state, a freeze constraint is applied to the blending ratio, keeping the blending ratio setting unchanged and prohibiting its increase; when it is necessary to reduce the risk of gas supply fluctuations, the blending ratio setting is lowered to the preset safety value; in the delayed blockage state, the cleaning and disposal actions are allowed to continue, but the blending ratio cannot be increased again before the freeze constraint is lifted. When the risk state enters the de-escalation state and the risk state is determined to be de-escalation state for L consecutive sampling periods, select the historical data segment that has been continuously de-escalated and supplement the flow rate and differential pressure data in the historical data segment into the health baseline dataset.

6. The biomass gas coupled multi-condition coordinated control method according to claim 5, characterized in that, The method for performing preventative thermal boundary programming and initiating action blind windows and differential verification when the risk state is in the condensation development state includes: When performing preventive thermal boundary scheduling actions, the current set values ​​of the heat exchange medium inlet temperature, heat exchange bypass opening, and heat tracing circuit output are used as the reference. The inlet temperature setpoint of the heat exchange medium is increased by the preset temperature increment. If the increase exceeds the upper limit of the inlet temperature setting, the upper limit of the inlet temperature setting is used as the target inlet temperature value. Increase the heat exchange bypass opening setting value by the preset opening increment. If the increase exceeds the opening limit, the opening limit will be used as the target value for the bypass opening. The output setting value of the heat tracing circuit is increased by the preset power increment. If the increase exceeds the upper limit of the heat tracing output setting, the upper limit of the heat tracing output setting is used as the target value of the heat tracing output. The time period corresponding to M consecutive sampling cycles from the start of the preventive thermal boundary arrangement action is taken as the action blind window. During the duration of the action blind window, the change in net temperature difference and the increase in net resistance are continuously calculated and recorded, but the risk status adjustment or cancellation judgment is not performed based on the change results. After the action blind window ends, the differential review stage begins.

7. The biomass gas coupled multi-condition coordinated control method according to claim 6, characterized in that, The differential verification method includes: The time period corresponding to the M consecutive sampling cycles after the end of the action blind window is used as the review cycle for judgment; When the net resistance increase at the end of the review period is not greater than the net resistance increase at the beginning of the review period, and the median value of the net resistance increase in each sampling period within the review period is less than the risk threshold of the net resistance increase, and the net temperature difference change is higher than the risk threshold of the net temperature difference change, the process of determining the release state is initiated. When the change in net temperature difference is higher than the risk threshold for the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold for the increase in net resistance, the risk state is determined to enter the delayed blockage state. When the change in net temperature difference is not higher than the risk threshold of the change in net temperature difference and the increase in net resistance at the end of the review period is not less than the risk threshold of the increase in net resistance, the condensation development state is maintained and the cleaning action is triggered. When the net resistance increase at the end of the review period is less than the risk threshold of the net resistance increase but does not meet the conditions for entering the resolution state determination process, the current risk state remains unchanged and differential review continues to be performed in subsequent sampling periods. If differential verification continues for more than 2M consecutive sampling cycles after the automatic blind window ends, a differential verification timeout alarm will be output, and the current collaborative operation constraints will remain unchanged.

8. The biomass gas coupled multi-condition coordinated control method according to claim 7, characterized in that, The cleaning and disposal actions include: First, apply a limiting constraint to the boiler side to ensure that the change in the blending ratio setting value during the cleaning operation does not exceed the preset blending ratio limit, and that the boiler load deviation does not exceed the preset load deviation limit. Under the premise of satisfying the aforementioned amplitude limit constraint, the existing purging circuit is triggered to perform channel cleaning action; Once the cleaning action begins, the blind window timing is started simultaneously. After the blind window ends, differential verification is performed again. When the net resistance increase is consistently below the risk threshold for net resistance increase during the review period, the risk status is allowed to enter the resolution determination process; when the net resistance increase is still not below the risk threshold for net resistance increase during the review period, the output channel is unavailable. When the system is in a condensation development state and no output channel is available, a ramp constraint is applied to the blending ratio adjustment and boiler load adjustment to ensure that the adjustment of the blending ratio setpoint in any sampling period does not exceed the product of the preset blending ratio ramp limit and the sampling period duration, and that the adjustment of the boiler load setpoint in any sampling period does not exceed the product of the preset load ramp limit and the sampling period duration; at the same time, it is prohibited to perform blending ratio increase and load increase simultaneously in the same sampling period. When the output channel is unavailable, a downgrade switching constraint is executed, causing the blending ratio setting value to be lowered to the preset safety value, and gas supply to the unavailable channel is stopped; when a backup channel or bypass channel is available, the system switches to the backup channel synchronously during the blending ratio reduction process to maintain gas supply continuity; when no backup channel is available, blending is stopped and the heat input is supplemented by coal input from the boiler side.

9. The biomass gas coupled multi-condition coordinated control method according to claim 7, characterized in that, The release state determination process includes: When the risk state enters the de-escalation state and is maintained for a preset number of periods, it is determined in turn whether two conditions are met: the net temperature difference change is higher than the risk threshold of the net temperature difference change, and the net resistance increase is lower than the risk threshold of the net resistance increase. When both conditions are met simultaneously, first release the freeze constraint, then release the downgrade switching constraint, and then release the ramp constraint to restore the blending ratio. If either of the two conditions is not met, the existing cooperative operation constraints remain unchanged, and the release state determination process continues to be executed in subsequent sampling cycles until the conditions are met. If the release condition is not met for more than 2L consecutive sampling cycles after the first time the release state entry condition is met, the release judgment timeout alarm will be output, and the freeze constraint and ramp constraint will be maintained. When the release judgment timeout alarm is triggered continuously for H times, the cleaning action will be performed. If the release judgment timeout alarm is still triggered after the cleaning action, the output channel will be unavailable.

10. A biomass gas coupled multi-condition cooperative control system, used to execute the biomass gas coupled multi-condition cooperative control method according to any one of claims 1-9, characterized in that, include: Data acquisition module: Collects the inlet pressure, outlet pressure, volumetric flow rate of combustible gas, and outlet temperature of the heat exchanger in the combustible gas channel, and arranges Z pipe section wall temperature measuring points along the combustible gas delivery pipeline; Cold point determination module: Sort the pipe section wall temperature measurement points and record the measurement point numbers of the lowest temperature measurement point and the second lowest temperature measurement point. Count the number of times the measurement point is recorded and calculate the low temperature deviation. Combine the proportion threshold and the deviation threshold to determine the cold point location. Take the pipe section wall temperature measurement point temperature corresponding to the cold point location as the cold point side temperature. Temperature difference acquisition module: Using the heat exchanger outlet temperature as the reference section temperature, the net temperature difference change is obtained by subtracting the reference section temperature difference from the cold point temperature difference between adjacent sampling periods. Differential pressure acquisition module: Based on the stable co-firing window, a flow health baseline dataset is established. The current baseline differential pressure and the previous baseline differential pressure are obtained from the health baseline dataset according to the combustible gas volume flow rate of adjacent sampling periods. The difference between the two is calculated, and the net resistance increase is obtained by subtracting the difference between the two from the channel differential pressure change. Risk assessment module: Window statistics are performed on the net temperature difference change and net resistance increase to obtain the corresponding window statistics value. The window statistics value is compared with the risk threshold to determine the risk status of tar condensation and deposition. The risk status includes condensation development state, delayed blockage state and resolution state. Execution module: Implements differentiated actions and collaborative constraint controls based on different risk states.

Images