A fan energy-saving control method and device based on load prediction and a medium

Abstract

The application discloses a fan energy-saving control method and device based on load prediction, and a medium, relates to the technical field of industrial automatic control, and comprises the following steps: determining a fan group with a load coordination relationship in the same dust removal area, marking historical production events according to production beat changes from historical operation data, extracting fan operation parameters, pipe network constraint parameters and dust removal effect parameters before and after the events, correlating fluctuation time sequences of air volume, pressure difference, dust concentration and power consumption, and forming a process event load sample library. The application can determine predicted dust removal load and target required air volume in advance, and reduce load judgment lag and excessive air supply by similar matching of the current production event, the affected dust removal area, the dust removal load baseline and the dust removal control margin with historical samples in the process event load sample library, and predicting load increments in the next control period according to a load response curve and a similarity weight.

Description

Technical Field

[0001] This invention relates to the field of industrial automatic control technology, and in particular to a method, equipment and medium for energy-saving control of fans based on load prediction. Background Technology

[0002] Dust removal fans are widely used in continuous production scenarios such as metallurgy, building materials, and chemicals. Their operating status directly affects dust removal efficiency, pipeline stability, and production energy consumption. With the increasing requirements for industrial automation and energy conservation, fan control is gradually developing from fixed-frequency operation and manual adjustment to variable-frequency adjustment, centralized monitoring, and collaborative control. The load in a multi-fan dust removal area will fluctuate in sequence with changes in production cycle, valve status, pipeline pressure difference, and dust concentration.

[0003] Existing methods have some shortcomings. They lack the ability to anticipate fluctuations in dust removal load caused by changes in production cycle time, which can lead to problems such as increasing output only after the load rises and maintaining a high air volume even after the load drops, resulting in control lag and excessive air supply. In addition, when multiple fans are coordinated and regulated, the load is often allocated according to fixed priority, operating frequency, or manual experience, without comprehensively considering the incremental power consumption corresponding to the unit incremental air volume, changes in pipeline pressure difference, and changes in dust concentration control margin. This can easily lead to inefficient operation of fans, pipeline disturbances, and excessive dust removal constraints. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a wind turbine energy-saving control method based on load prediction to solve the problems of lagging dust removal load prediction and inaccurate matching of wind turbine energy efficiency.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a fan energy-saving control method based on load forecasting, comprising: identifying fan units with load coordination relationships within the same dust removal area; marking historical production events from historical operating data according to changes in production cycle time; extracting fan operating parameters, pipeline constraint parameters, and dust removal effect parameters before and after the event; associating the fluctuations in air volume, pressure difference, dust concentration, and power consumption in a time sequence to form a process event load sample library; identifying the current production event based on changes in production cycle time within the current control cycle; determining the affected dust removal area based on valve status; forming a dust removal load baseline based on the current fan operating parameters; forming a dust removal control margin based on the current pipeline constraint parameters and dust removal effect parameters; performing similarity matching between the current production event, the affected dust removal area, the dust removal load baseline, and the dust removal control margin and historical samples in the process event load sample library to determine historical similar samples; and predicting the load increment for the next control cycle based on the load response curve and similarity weight of the historical similar samples to obtain the predicted dust removal load and target required air volume. The target air volume requirement triggers the energy efficiency matching of the fan units. Combining the fan operating parameters, adjustable frequency range, valve adjustment status, and pipeline constraint parameters of each fan, the incremental power consumption and dust removal constraint change corresponding to the unit incremental air volume are calculated, generating marginal energy consumption evaluation values. These values ​​are then sorted from low to high to obtain the fan unit execution plan. After safety verification, the fan unit execution plan is converted into control instructions for PLC, DCS, frequency converter, and valve actuators. The predicted dust removal load is compared with the dust removal load baseline to obtain the predicted dust removal load change. When the predicted dust removal load change exceeds the control dead zone and meets the frequency change limit, pressure difference limit, and interlock protection conditions, the fan determined to increase output according to the fan unit execution plan increases its output, and the fan determined to decrease output decreases its output. The process event load sample library is updated with measured air volume, pipeline pressure difference, dust concentration, and fan power consumption.

[0007] As a preferred embodiment of the wind turbine energy-saving control method based on load forecasting described in this invention, the specific steps for forming the process event load sample library are as follows: Within the same dust removal area, identify fan units with load coordination relationships, align fan operating parameters, pipeline constraint parameters, dust removal effect parameters, and production cycle changes according to the collection time to obtain historical operating sequence data for the area; Among them, the fan units with load coordination relationship are determined based on the connection relationship between the fan and the dust removal pipeline, the valve adjustment range, and the dust removal points that work together; Historical production events are marked by changes in production cycle time in the historical runtime data of the region, and the fan operation parameters, pipeline constraint parameters and dust removal effect parameters before, during and after the historical production events are extracted to obtain historical event parameter fragments. By correlating the air volume, pressure difference, dust concentration, and power consumption in the historical event parameter fragments in chronological order, the load response relationship corresponding to the historical production events is obtained, and a sample library of process event loads is compiled.

[0008] As a preferred embodiment of the wind turbine energy-saving control method based on load prediction described in this invention, the specific steps for forming the dust removal load baseline are as follows: Within the current control cycle, the moment when the production cycle change state changes from a stable operating state to a changing operating state is marked as the start time of the current production event, and the process cycle change content corresponding to the production cycle change state is marked as the current production event; The dust collection area corresponding to the valve whose valve status changes is identified as the affected dust collection area; The data collection time that was in a stable operating state before the start time of the current production event was selected as the baseline time. The air volume and power consumption of the data collection cycle corresponding to the baseline time were used to form the dust removal load baseline. The remaining space between the differential pressure and the differential pressure limit at the start time of the current production event is determined as the differential pressure margin, and the remaining space between the dust concentration and the upper limit of the dust concentration control at the start time of the current production event is determined as the dust concentration margin. The differential pressure margin and the dust concentration margin together form the dust removal control margin.

[0009] As a preferred embodiment of the wind turbine energy-saving control method based on load forecasting described in this invention, the specific steps for determining historical similar samples are as follows: Historical samples that are consistent with current production events and historical production events, and whose affected dust removal areas are consistent with the dust removal areas corresponding to historical production events, are identified as candidate historical samples. The current air volume baseline and the current power consumption baseline in the dust removal load baseline are compared with the air volume benchmark value and the power consumption benchmark value in the candidate historical samples, respectively. The pressure difference margin and dust concentration margin in the dust removal control margin are compared with the pressure difference margin and dust concentration margin in the candidate historical samples, respectively, to obtain the matching distance. Candidate historical samples are sorted from low to high according to matching distance, and the candidate historical samples at the top of the ranking are selected to determine historical similar samples. Similarity is calculated based on the matching distance of historical similar samples, where a lower matching distance corresponds to a higher similarity. The similarity of each historical similar sample is then normalized to obtain a similarity weight.

[0010] As a preferred embodiment of the wind turbine energy-saving control method based on load prediction described in this invention, the specific steps for obtaining the predicted dust removal load and target required air volume are as follows: Extract the load change segment corresponding to the next control cycle from the load response curves of similar historical samples; The load change segments of each historical similar sample are weighted and fused according to the similarity weight to obtain the air volume increment, pressure difference increment, dust concentration increment and power consumption increment of the next control cycle. The load increment of the next control cycle is superimposed on the dust removal load baseline to obtain the predicted air volume, predicted pressure difference, predicted dust concentration and predicted power consumption of the next control cycle, and form the predicted dust removal load. When the predicted differential pressure does not exceed the differential pressure limit and the predicted dust concentration does not exceed the upper limit of dust concentration control, the predicted air volume is determined as the target required air volume. When either the predicted differential pressure exceeds the differential pressure limit or the predicted dust concentration exceeds the upper limit of dust concentration control, the load change segment where both differential pressure and dust concentration are within the limit range is determined from the load response curves of similar historical samples, and the corresponding air volume change is fused according to the similarity weight to determine the target required air volume.

[0011] As a preferred embodiment of the wind turbine energy-saving control method based on load forecasting described in this invention, the specific steps for generating the marginal energy consumption evaluation value are as follows: Determine the incremental air volume of the fan unit based on the target required air volume and the current actual total air volume of the fan unit. In a fan unit with load coordination relationship, the fan that is determined to be in normal operation, has adjustable frequency range, has valve adjustment status that is connected to and adjustable in the affected dust removal area, and meets the differential pressure limit of the pipeline constraint parameters is the fan to participate in energy efficiency matching. For wind turbines involved in energy efficiency matching, determine the incremental power consumption, pipeline pressure difference change, and dust concentration control margin change under unit incremental operating conditions; Based on the incremental power consumption, the amount of pipeline pressure difference change relative to the pressure difference margin, and the amount of dust concentration control margin change relative to the dust concentration margin, the marginal energy consumption evaluation value corresponding to each fan participating in energy efficiency matching is generated.

[0012] As a preferred embodiment of the wind turbine energy-saving control method based on load forecasting described in this invention, the specific steps for obtaining the wind turbine execution scheme are as follows: The wind turbines participating in energy efficiency matching are ranked from low to high according to their marginal energy consumption evaluation values, thus obtaining the marginal energy consumption ranking. When the incremental air volume of the wind turbine unit is greater than zero, the incremental air volume of the wind turbine unit is allocated from front to back according to the marginal energy consumption ranking, and the incremental air volume of the wind turbine unit that has not yet been allocated is preferentially allocated to the wind turbines in the front order of the marginal energy consumption ranking. Once the current fan reaches its allocable incremental limit, the remaining incremental air volume is allocated to the next fan until the incremental air volume of the fan group is fully allocated. The fan allocated to the incremental air volume is identified as the fan that increases output, and the corresponding frequency adjustment amount is determined based on the allocated incremental air volume, the current actual air volume, and the adjustable frequency range. When the incremental air volume of the fan unit is less than zero, the absolute value of the incremental air volume of the fan unit is determined as the air volume to be reduced. The fans participating in energy efficiency matching are verified according to the unit reduction operation status corresponding to the reduction output direction. The air volume to be reduced is allocated to the fans that meet the verification from the back to the front according to the marginal energy consumption ranking. The air volume to be reduced that has not yet been allocated is preferentially allocated to the fans that meet the verification in the later order of the marginal energy consumption ranking. The upper limit of the air volume that can be reduced is determined based on the difference between the current actual air volume and the air volume corresponding to the lower limit of the adjustable frequency range; Once the subsequent fan reaches the upper limit of the air volume that can be reduced, the remaining air volume to be reduced is allocated to the previous fan until the air volume to be reduced is completely allocated. The fan assigned to reduce the air volume is identified as the fan with reduced output, and the corresponding frequency adjustment amount is determined based on the assigned air volume to be reduced, the current actual air volume, and the adjustable frequency range. The execution plan for the fan unit is formed by summarizing the marginal energy consumption ranking, the incremental air volume allocated to each fan, the air volume to be reduced allocated to each fan, the fan output adjustment direction, the frequency adjustment amount, and the valve adjustment status.

[0013] As a preferred embodiment of the wind turbine energy-saving control method based on load forecasting described in this invention, the step of controlling wind turbines determined to increase output according to the wind turbine group execution plan to increase output and controlling wind turbines determined to decrease output to decrease output, and updating the process event load sample database with measured air volume, pipeline pressure difference, dust concentration, and wind turbine power consumption, is as follows: Based on the fan output adjustment direction, frequency adjustment amount, and valve adjustment status in the fan unit execution plan, the execution content to be verified is formed, and the predicted dust removal load is compared with the current dust removal load baseline to obtain the predicted dust removal load change amount. When the predicted change in dust removal load exceeds the control dead zone, a safety check is performed on the frequency adjustment amount, the expected pipeline pressure difference and interlock status corresponding to the content to be checked; When the frequency adjustment meets the frequency change limit, the expected pipeline differential pressure meets the differential pressure limit after execution, and the interlock status meets the interlock protection conditions, the execution content to be verified is converted into the fan number, fan output adjustment direction, frequency adjustment amount, valve adjustment status and execution time sent to the PLC and DCS, and frequency control instructions are sent to the frequency converter and valve adjustment instructions are sent to the valve actuator. According to the frequency control command, the fan determined to increase output increases output, and the fan determined to decrease output decreases output. The valve actuator is controlled to operate according to the valve adjustment command. During the execution of frequency control commands and valve adjustment commands, the measured air volume, pipeline pressure difference, dust concentration and fan power consumption are correlated according to the acquisition time to form the actual load response process, and the actual load response process is updated to the process event load sample library.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the wind turbine energy-saving control method based on load forecasting as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the wind turbine energy-saving control method based on load forecasting as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By similarly matching the current production event, affected dust removal area, dust removal load baseline, and dust removal control margin with historical samples in the process event load sample library, and predicting the load increment of the next control cycle according to the load response curve and similarity weight, the predicted dust removal load and target demand air volume can be determined in advance, reducing load judgment lag and excessive air supply; by calculating the incremental power consumption and dust removal constraint changes corresponding to the unit incremental air volume of each fan, the marginal energy consumption evaluation value is generated and sorted to form the fan group execution plan, which enables the fan group to coordinately adjust according to the comprehensive evaluation results of energy consumption and dust removal constraints, while meeting the air volume demand, pressure difference limit, and dust concentration control requirements, reducing ineffective energy consumption, and improving the accuracy and stability of energy-saving control of dust removal fans. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the 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.

[0018] Figure 1 This is a flowchart of a wind turbine energy-saving control method based on load forecasting.

[0019] Figure 2 This is a sample library of process event loads and a flowchart for load prediction.

[0020] Figure 3 This is a flowchart for evaluating and implementing the marginal energy consumption of wind turbine units.

[0021] Figure 4 Flowchart for security verification and closed-loop update.

[0022] Figure 5 This is a comparison chart of load forecast air volume tracking data.

[0023] Figure 6 This is a comparison chart of marginal energy consumption evaluation and incremental air volume allocation data. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 This is one embodiment of the present invention, which provides a wind turbine energy-saving control method based on load forecasting, including the following steps: S1. Identify the fan units with load coordination relationships within the same dust removal area, mark historical production events according to changes in production cycle from historical operating data, extract the fan operating parameters, pipeline constraint parameters, and dust removal effect parameters before and after the events, and correlate the fluctuations in air volume, pressure difference, dust concentration, and power consumption in sequence to form a process event load sample library.

[0028] S1.1. Identify the fan units with load coordination relationships within the same dust removal area, align the fan operating parameters (including air volume and power consumption), the pipeline constraint parameters (including pressure difference), the dust removal effect parameters (including dust concentration), and the production cycle change according to the collection time to obtain the historical operating sequence data of the area.

[0029] It should be noted that within the same dust removal area, fan units with load synergy are determined based on the connection relationship between the fan and the dust removal pipeline, the valve adjustment range, and the dust removal points that work together. Fan units with load synergy refer to a set of fans in the same dust removal area whose output changes can jointly affect the changes in air volume, pressure difference, and dust concentration at the corresponding dust removal points.

[0030] Extract the corresponding fan operating parameters, pipeline constraint parameters, dust removal effect parameters, and production cycle changes from historical operating data. The fan operating parameters include air volume and power consumption per acquisition cycle, the pipeline constraint parameters include pressure difference, the dust removal effect parameters include dust concentration, and the production cycle changes are determined according to the acquisition time in the production records.

[0031] Using the data collection interval of the fan operating parameters as a uniform time interval, the changes in fan operating parameters, pipeline constraint parameters, dust removal effect parameters, and production cycle time are time-aligned. When a corresponding parameter is missing at the same collection time, the data with the closest collection value from adjacent collection times is used to fill in the missing parameter. When there is duplicate data at the same collection time, the data with the later time sequence is retained, so that the same collection time corresponds to a set of fan operating parameters, pipeline constraint parameters, dust removal effect parameters, and production cycle time changes.

[0032] After time alignment, the historical runtime sequence data of the region is obtained; in the historical runtime sequence data of the region, each collection moment corresponds to the air volume, power consumption of the collection cycle, pressure difference, dust concentration and production cycle change status.

[0033] S1.2. Mark historical production events based on changes in production cycle time in the historical runtime sequence data of the region, and extract the fan operation parameters, pipeline constraint parameters and dust removal effect parameters before, during and after the historical production events to obtain historical event parameter fragments.

[0034] It should be noted that after the regional historical operating sequence data is arranged continuously according to the collection time, it is divided into stable operating state and changing operating state according to the change state of production rhythm. The collection time when the change state of production rhythm changes from stable operating state to changing operating state is marked as the start time of historical production event. The collection time when the change state of production rhythm changes from changing operating state to stable operating state is marked as the end time of historical production event. Historical production events are formed according to the start time and end time of historical production events.

[0035] Based on the time range of continuous stable operation before the start of the historical production event, the duration of the historical production event, and the time range of recovery to stable operation after the end of the historical production event, the regional historical operation sequence data is segmented to obtain the pre-event time sequence, the mid-event time sequence, and the post-event time sequence. The pre-event time sequence corresponds to the stable operation state before the occurrence of the historical production event, the mid-event time sequence corresponds to the operational changes during the process of the historical production event, and the post-event time sequence corresponds to the recovery state after the end of the historical production event.

[0036] In the pre-event, mid-event, and post-event time series, the time correspondence between the collection time of the fan operating parameters, pipeline constraint parameters, and dust removal effect parameters is preserved, so that the fan operating parameters, pipeline constraint parameters, and dust removal effect parameters correspond to the start time and end time of the historical production event, forming the parameter change process of the historical production event.

[0037] By collecting the start time, end time, pre-event time segment, mid-event time segment, post-event time segment, and corresponding fan operating parameters, pipeline constraint parameters, and dust removal effect parameters for the same historical production event, a fragment of historical event parameters is obtained.

[0038] S1.3. Correlate the fluctuations of air volume, pressure difference, dust concentration and power consumption in the historical event parameter fragments in chronological order to obtain the load response relationship corresponding to the historical production events, and summarize them to form a process event load sample library.

[0039] It should be noted that the air volume, pressure difference, dust concentration, and power consumption per data collection cycle in the pre-event time series characterize the stable operating state before the historical production event. Reference values ​​for air volume, pressure difference, dust concentration, and power consumption per data collection cycle are determined based on the air volume, pressure difference, dust concentration, and power consumption per data collection cycle at each data collection point in the pre-event time series. The reference values ​​are determined by the average value of the corresponding parameters within the pre-event time series, specifically: , ; , ; in, Indicates the air volume reference value; Indicates the differential pressure reference value; This represents the baseline value for dust concentration; This represents the baseline value of power consumption during the data collection cycle; Indicates the number of data collection points in the time series preceding the event; subscript Indicates the baseline value; subscript Indicates the sequence number of the time point collected in the time sequence preceding the event; Indicates the first time segment before the event Each data collection moment; Indicates the first Air volume at each sampling moment; Indicates the first Pressure difference at each sampling time; Indicates the first Dust concentration at each sampling time; Indicates the first Power consumption per acquisition cycle at each acquisition moment.

[0040] After determining the baseline values ​​for air volume, differential pressure, dust concentration, and power consumption during the data acquisition cycle, the changes in each data acquisition moment relative to the baseline values ​​are calculated according to the data acquisition time, within the time sequence of the event and after the event. This yields the air volume fluctuation process, differential pressure fluctuation process, dust concentration fluctuation process, and power consumption fluctuation process during the data acquisition cycle, specifically as follows: , ; , ; in, This represents any acquisition time within the time segment of the event and the subsequent time segment; express Changes in air volume over time; express The change in pressure difference at any given time; express The change in dust concentration over time; express The change in power consumption during the data acquisition cycle at any given time; express Air volume at any given moment; express Pressure difference at any given moment; express Dust concentration at any given time; express Power consumption per data acquisition cycle at any given time.

[0041] The changes in air volume, differential pressure, dust concentration, and power consumption during the collection cycle are arranged according to the collection time, so that the same collection time corresponds to a set of changes in air volume, differential pressure, dust concentration, and power consumption during the collection cycle, forming the load response relationship corresponding to historical production events. The load response relationship represents the linkage process of changes in air volume, differential pressure, dust concentration, and power consumption during the collection cycle over time after the occurrence of historical production events.

[0042] The historical event parameter fragments and load response relationships corresponding to different historical production events are collected according to the same dust removal area, the fan units with load coordination relationship and the changes in production cycle, to form a process event load sample library.

[0043] S2. Within the current control cycle, identify the current production event based on the change in production cycle time, determine the affected dust removal area based on the valve status, form a dust removal load baseline based on the current fan operating parameters, and form a dust removal control margin based on the current pipeline constraint parameters and dust removal effect parameters. Perform similarity matching between the current production event, the affected dust removal area, the dust removal load baseline, and the dust removal control margin and the historical samples in the process event load sample library to determine the historical similar samples. Predict the load increment of the next control cycle according to the load response curve and similarity weight of the historical similar samples to obtain the predicted dust removal load and target demand air volume.

[0044] S2.1 Within the current control cycle, identify the current production event based on the change in production cycle time, determine the affected dust removal area based on the valve status, and simultaneously extract the current fan operating parameters, current pipeline constraint parameters, and dust removal effect parameters. Form a dust removal load baseline based on the current fan operating parameters, and form a dust removal control margin based on the current pipeline constraint parameters and dust removal effect parameters.

[0045] It should be noted that, within the current control cycle, the moment when the production cycle change state transitions from a stable operating state to a changing operating state is recorded as the start time of the current production event; the change in the process cycle corresponding to the change in the production cycle state is recorded as the current production event.

[0046] The current valve status is determined according to the data collection time and the start time of the current production event. The dust collection area corresponding to the valve whose status changes is determined as the affected dust collection area. When the same current production event corresponds to multiple valve status changes, the same dust collection area where multiple valves work together is determined as the affected dust collection area.

[0047] Before the start of the current production event, the data collection time in a stable operating state is selected as the baseline time; the air volume and power consumption of the current fan operating parameters corresponding to the baseline time together form the dust removal load baseline; among them, the air volume represents the ventilation load level of the current dust removal area, and the power consumption of the current fan unit represents the energy consumption state corresponding to the current fan unit maintaining the ventilation load level.

[0048] In the current pipeline constraint parameters and dust removal effect parameters corresponding to the start time of the current production event, the remaining space between the differential pressure and the differential pressure limit in the current pipeline constraint parameters forms the differential pressure margin, and the remaining space between the dust concentration and the upper limit of dust concentration control in the dust removal effect parameters forms the dust concentration margin. The differential pressure margin and the dust concentration margin together form the dust removal control margin, specifically: ; ; in, Indicates differential pressure margin; subscript Indicates pressure difference; Indicates differential pressure limit; subscript Indicates the limit value; This represents the pressure difference corresponding to the start time of the current production event; Indicates the start time of the current production event; subscript Indicates the current production event; Indicates the dust concentration margin; subscript Indicates dust concentration; Indicates the upper limit of dust concentration control; This indicates the dust concentration at the start of the current production event.

[0049] S2.2. Perform similarity matching between the current production event, the affected dust removal area, the dust removal load baseline, the dust removal control margin and the historical samples in the process event load sample library. Filter the historical samples that match the production event and the dust removal area, and determine the historical similar samples according to the matching degree of the dust removal load baseline and the dust removal control margin. Form a similarity weight based on the matching degree of the historical similar samples.

[0050] It should be noted that the historical samples in the process event load sample library include historical production events, historical dust removal areas, historical event parameter segments, and load response curves; the historical dust removal load baseline is determined by the air volume benchmark value and the power consumption benchmark value of the acquisition cycle in the historical event parameter segment, and the historical dust removal control margin is determined by the pressure difference margin and dust concentration margin in the historical event parameter segment.

[0051] Historical samples where the current production event matches a historical production event and where the affected dust removal area matches a historical dust removal area are identified as candidate historical samples; for the first The candidate historical samples will be compared with the current air volume baseline and the current data collection cycle power consumption baseline in the dust removal load baseline. The air volume baseline value and the power consumption baseline value of the data collection cycle in the candidate historical samples are compared; the pressure difference margin and dust concentration margin in the dust removal control margin are compared with the first... The pressure margin and dust concentration margin in each candidate historical sample are compared to obtain the first... The matching distance of each candidate historical sample is as follows: ; in, Indicates the first Matching distance of each candidate historical sample; subscript Indicates the sequence number of the candidate historical sample; Indicates the current air volume baseline in the dust removal load baseline; subscript Indicates the currently matched object; Indicates the first Air volume baseline values ​​in candidate historical samples; This represents the current data collection cycle power consumption baseline within the dust removal load baseline; Indicates the first Baseline power consumption values ​​for the acquisition cycle in each candidate historical sample; Indicates the first Pressure margin in each candidate historical sample; Indicates the first Dust concentration margin in candidate historical samples.

[0052] Candidate historical samples are sorted from lowest to highest matching distance, and the number of historical similar samples to be included in load forecasting is determined based on the number of candidate historical samples. When the number of candidate historical samples does not reach the upper limit for the number of historical similar samples, all candidate historical samples are identified as historical similar samples. When the number of candidate historical samples reaches the upper limit for the number of historical similar samples, the candidate historical samples with the highest matching distance are selected as historical similar samples. The upper limit for the number of historical similar samples is determined based on the number of historical sample records corresponding to the same production event and the same dust removal area in the process event load sample library, and remains consistent during the load forecasting process for the same dust removal area. The lower the matching distance, the higher the degree of matching between the historical similar sample and the current production event, the affected dust removal area, the dust removal load baseline, and the dust removal control margin.

[0053] Similarity is calculated based on the matching distance of historical similar samples, where a lower matching distance corresponds to a higher similarity. The similarity scores of each historical similar sample are then normalized to obtain a similarity weight, as follows: ; ; in, Indicates the first Similarity between historical similar samples; Indicates the first Similarity weights for historical similar samples; Indicates the number of historically similar samples; subscript Indicates the index when summing historically similar samples; Indicates the first Similarity of historical similar samples.

[0054] S2.3 Extract the load change segment corresponding to the next control cycle from the load response curve of similar historical samples, merge the load change segments according to the similarity weight to obtain the load increment of the next control cycle, superimpose the load increment of the next control cycle onto the dust removal load baseline to obtain the predicted dust removal load, and determine the target demand air volume in combination with the dust removal control margin.

[0055] It should be noted that the load response curve of the historical similar sample takes the start time of the historical production event as the zero point of time, and records the changes in air volume, pressure difference, dust concentration and power consumption during the acquisition cycle after the occurrence of the historical production event over time; the time position from the start time of the current production event to the next control cycle corresponds to the relative time position in the load response curve of the historical similar sample, thereby determining the load change segment corresponding to the next control cycle.

[0056] The load change segments of historical similar samples include the changes in air volume, differential pressure, dust concentration, and power consumption during the next control cycle. The load change segments of each historical similar sample are weighted and fused according to similarity weights to obtain the increments in air volume, differential pressure, dust concentration, and power consumption during the next control cycle, specifically: , ; , ; in, This indicates the air volume increment for the next control cycle; This indicates the differential pressure increment for the next control cycle; This indicates the increase in dust concentration for the next control cycle; Indicates the power consumption increment during the next control cycle; subscript Indicates the next control cycle; This indicates the relative acquisition time of the next control cycle in the load response curve; Indicates the first Historically similar samples The corresponding change in air volume; Indicates the first Historically similar samples The corresponding change in pressure difference; Indicates the first Historically similar samples The corresponding change in dust concentration; Indicates the first Historically similar samples The corresponding change in power consumption during the data acquisition cycle.

[0057] The load increment for the next control cycle is superimposed onto the dust removal load baseline to obtain the predicted air volume, predicted pressure difference, predicted dust concentration, and predicted power consumption for the next data collection cycle, specifically: , ; , ; in, This indicates the predicted air volume for the next control cycle; This indicates the predicted pressure difference for the next control cycle; This indicates the predicted dust concentration for the next control cycle; This indicates the predicted data acquisition cycle power consumption for the next control cycle.

[0058] The predicted dust removal load is jointly characterized by the predicted air volume, predicted differential pressure, predicted dust concentration, and predicted power consumption of the next control cycle. The predicted differential pressure and predicted dust concentration are used to determine the utilization of the differential pressure margin and dust concentration margin, respectively. When the predicted differential pressure does not exceed the differential pressure limit and the predicted dust concentration does not exceed the upper limit of dust concentration control, the predicted air volume is determined as the target demand air volume. When either the predicted differential pressure exceeds the differential pressure limit or the predicted dust concentration exceeds the upper limit of dust concentration control, the load change segment where both differential pressure and dust concentration are within the limit range is determined from the load response curves of similar historical samples, and the corresponding air volume change is fused according to the similarity weight to determine the target demand air volume.

[0059] Figure 5 The paper demonstrates the changes in air volume tracking under different control methods when dust removal load increases and decreases due to changes in production cycle. It can be seen that the target air volume and the actual output air volume of the present invention can closely match the actual air volume in a timely manner. The actual output air volume of the real-time threshold control responds slowly during the load increase phase, but maintains a high output during the load decrease phase, showing obvious lag and excessive air supply. This indicates that the present invention can determine the target air volume in advance by matching historical similar samples and predicting the load increment of the next control cycle, thereby reducing load judgment lag and excessive air supply.

[0060] Figure 5 The “Actual Demand Air Volume” refers to the objective demand air volume of the affected dust removal area at each collection time after the current production event occurs; “Target Demand Air Volume of this Invention” refers to the target air volume predicted by this invention based on the current production event, the affected dust removal area, the dust removal load baseline, and the dust removal control margin, matching it with historical similar samples in the process event load sample library, and according to the load response curve and similarity weight; “Actual Output Air Volume of this Invention” refers to the actual fan output air volume formed after the target demand air volume is matched with the fan unit energy efficiency, safety verification, and control command execution; “Real-time Threshold Control Actual Output Air Volume” refers to the control scheme that does not perform load prediction, but adjusts the fan output only after the air volume deviation, pressure difference change, or dust concentration change reaches the trigger condition.

[0061] It should also be noted that existing technologies typically adjust fan output directly based on real-time air volume, power, differential pressure, or fixed thresholds, which suffers from problems such as load judgment lag, inaccurate target air volume, and excessive air supply. This solution uses the current production event, affected dust removal area, dust removal load baseline, and dust removal control margin to perform similarity matching with historical samples. It then predicts the load increment for the next control cycle based on the load response curves and similarity weights of historical similar samples, thus obtaining the predicted dust removal load and target air volume demand. This approach can reflect load fluctuations caused by changes in production cycle time in advance, while taking into account differential pressure limits and dust concentration control requirements, improving the accuracy of the target air volume demand, and providing a reliable basis for subsequent fan unit energy efficiency matching.

[0062] S3. The target air volume requirement triggers the energy efficiency matching of the fan unit. Combining the fan operating parameters, adjustable frequency range, valve adjustment status and pipeline constraint parameters of each fan, the incremental power consumption and dust removal constraint change corresponding to the unit incremental air volume are calculated, the marginal energy consumption evaluation value is generated, and the evaluation values ​​are sorted from low to high to obtain the fan unit execution plan.

[0063] S3.1 After the target required air volume enters the fan unit energy efficiency matching, the incremental air volume of the fan unit is determined based on the current actual air volume in the fan operating parameters, and the fans participating in the energy efficiency matching are determined based on the fan operating parameters, adjustable frequency range, valve adjustment status and pipeline constraint parameters of each fan.

[0064] It should be noted that after the target air volume requirement is entered into the fan unit energy efficiency matching, the current actual total air volume of the fan unit is calculated based on the current actual air volume of each fan in the fan unit with load coordination relationship in the same dust removal area; the target air volume requirement is compared with the current actual total air volume of the fan unit to determine the incremental air volume of the fan unit, specifically as follows: , ; in, Indicates the current actual total air volume of the fan unit; subscript Indicates the sum; Indicates the serial number of the fan in the fan unit; Indicates the number of fans in the fan unit; subscript Indicates a wind turbine unit; Indicates the first The current actual air volume of the typhoon fan; This indicates the incremental air volume of the fan unit; Indicates the target required air volume; subscript Indicates the goal.

[0065] When the incremental air volume of the fan unit is greater than zero, the incremental air volume of the fan unit represents the air volume that needs to be supplemented by the fan unit in the current control cycle; when the incremental air volume of the fan unit is not greater than zero, no more incremental air volume will be allocated, and the fan output will be reduced as the adjustment direction in the subsequent fan unit energy efficiency matching.

[0066] After the incremental air volume of the fan unit is determined, the status of each fan with load coordination relationship in the same dust removal area is confirmed. The fan operating parameters indicate that the fan is in normal operation, the adjustable frequency range indicates that the fan's current operating frequency has adjustable space, the valve adjustment status indicates that the corresponding valve of the fan is connected to the affected dust removal area and is in an adjustable state, and the pipeline constraint parameters indicate that the fan whose current pressure difference meets the pressure difference limit is determined to be the fan participating in energy efficiency matching.

[0067] The fans participating in energy efficiency matching retain the corresponding current actual air volume, current data collection cycle power consumption, current operating frequency, adjustable frequency range, valve adjustment status, and pipeline constraint parameters.

[0068] S3.2 For the fans participating in energy efficiency matching, calculate the incremental power consumption, pipeline pressure difference change and dust concentration control margin change corresponding to the unit incremental air volume, and generate the marginal energy consumption evaluation value of each fan based on the incremental power consumption and dust removal constraint change.

[0069] It should be noted that, regarding the first The fan participating in energy efficiency matching adjusts the current operating frequency to the adjacent executable frequency point along the direction of increasing output within the adjustable frequency range, forming a unit increment operating state; the unit increment operating state corresponds to the air volume after the unit increment air volume, the power consumption of the collection cycle, the pipeline pressure difference and the dust concentration control margin.

[0070] Adjacent executable frequency points are determined by the minimum frequency adjustment step size of the inverter and the adjustable frequency range of the corresponding fan. The air volume, power consumption during the acquisition cycle, pipeline pressure difference, and dust concentration control margin under unit incremental operation are preferentially extracted from historical operation records in the process event load sample library for the same affected dust removal area, the same valve adjustment state, and whose operating frequency has been adjusted from the current operating frequency to the adjacent executable frequency point. When the corresponding historical operation record is missing in the process event load sample library, the air volume, power consumption during the acquisition cycle, and pipeline pressure difference under unit incremental operation are determined according to the correspondence between the operating frequency, air volume, power consumption during the acquisition cycle, and pipeline pressure difference of the same fan within the adjustable frequency range in the historical operation data. The dust concentration control margin is determined based on the remaining space between the dust concentration under unit incremental operation and the upper limit of dust concentration control.

[0071] Under unit increment operation, the valve regulation status remains connected to the affected dust removal area, the pipeline constraint parameters remain within the differential pressure limit range, and both differential pressure margin and dust concentration margin have remaining space, enabling the unit increment operation state to characterize the first The impact of the increased air volume of the wind turbines participating in energy efficiency matching on energy consumption and dust removal constraints.

[0072] No. The unit incremental power consumption, pipeline pressure difference change, and dust concentration control margin change corresponding to the wind turbines participating in energy efficiency matching are determined using the following methods: ; , ; in, Indicates the first The unit incremental power consumption corresponding to the wind turbines participating in energy efficiency matching; superscript Indicates the unit increment operation status; Indicates the first The power consumption of the wind turbines participating in energy efficiency matching under unit incremental operation conditions during the data collection cycle; Indicates the first The current data collection cycle power consumption of the wind turbines participating in energy efficiency matching; Indicates the first The air volume of the fan participating in energy efficiency matching under unit incremental operation; Indicates the first The change in pipeline pressure difference corresponding to the wind turbines participating in energy efficiency matching; Indicates the first The pipeline pressure difference of the wind turbines participating in energy efficiency matching under unit incremental operation conditions; Indicates the first The change in dust concentration control margin corresponding to the wind turbines participating in energy efficiency matching; Indicates the first The dust concentration margin of the wind turbines participating in energy efficiency matching under unit incremental operation.

[0073] After determining the unit incremental power consumption, pipeline pressure difference change, and dust concentration control margin change for each fan participating in energy efficiency matching, first determine the average unit incremental power consumption, and then generate the next... The marginal energy consumption evaluation value corresponding to the wind turbines participating in energy efficiency matching: ; ; in, This represents the average power consumption per unit increment. Indicates the number of wind turbines participating in energy efficiency matching; subscript This indicates participation in energy efficiency matching; Indicates the first The marginal energy consumption evaluation value corresponding to the wind turbines participating in energy efficiency matching; Indicates the first The absolute value of the pipeline pressure difference change corresponding to the wind turbines participating in energy efficiency matching.

[0074] The lower the marginal energy consumption evaluation value, the better. The lower the incremental power consumption of the fan that participates in energy efficiency matching when it undertakes the incremental air volume, the smaller the impact on the pipeline pressure difference and dust concentration control margin.

[0075] S3.3. Sort the fans participating in energy efficiency matching according to the marginal energy consumption evaluation value from low to high, and determine the incremental air volume allocation method and the air volume to be reduced allocation method according to the positive and negative values ​​of the incremental air volume of the fan group, so as to form a fan group execution plan that includes the fan output adjustment direction, frequency adjustment amount and valve adjustment status.

[0076] It should be noted that the wind turbines participating in energy efficiency matching are ranked from low to high according to their marginal energy consumption evaluation values, resulting in a marginal energy consumption ranking. Wind turbines ranked higher in the marginal energy consumption ranking have lower comprehensive evaluations of incremental power consumption and dust removal constraint changes when undertaking incremental air volume per unit, and are given priority in participating in the allocation of incremental air volume for wind turbine groups.

[0077] When the incremental air volume of the wind turbine units is greater than zero, the incremental air volume of the wind turbine units is allocated from front to back according to the marginal energy consumption ranking; for the first in the marginal energy consumption ranking... For each fan unit, the upper limit of the allocable incremental air volume is determined based on the difference between the upper limit of the adjustable frequency range and the current actual air volume. Unallocated incremental air volume is prioritized for allocation to the fan unit preceding the one in the marginal energy consumption ranking. Once the preceding fan reaches its allocable incremental air volume limit, the remaining incremental air volume is allocated to the next fan unit, until the incremental air volume of the fan unit is fully allocated. Specifically: ; in, This indicates the order of marginal energy consumption. Incremental air volume allocated to the fan; This indicates the order of marginal energy consumption. The serial number of the fan; subscript Indicates the position in the marginal energy consumption ranking; subscript This indicates the position in the marginal energy consumption ranking that has been assigned. This indicates the order of marginal energy consumption. Incremental air volume allocated to the fan; This indicates the order of marginal energy consumption. The upper limit of the allocatable increment corresponding to the adjustable frequency range of the fan; This indicates taking the maximum value; This indicates taking the minimum value.

[0078] When the incremental air volume of the fan unit is less than zero, the absolute value of the incremental air volume of the fan unit is determined as the air volume to be reduced. For the fans participating in energy efficiency matching, within the adjustable frequency range, the current operating frequency is adjusted to the adjacent executable frequency point along the reduction output direction to form a unit reduction operation state. Based on the air volume, power consumption during the sampling cycle, pipeline pressure difference, and dust concentration control margin under the unit reduction operation state, the power consumption change, pipeline pressure difference change, and dust concentration control margin change corresponding to the unit air volume to be reduced are determined. Fans that still meet the pressure difference limit and dust concentration control upper limit under the unit reduction operation state are designated as the air volume to be reduced allocation objects. For the marginal energy consumption ranking, the first... For each fan, the upper limit of the reducible air volume is determined based on the difference between the current actual air volume and the air volume corresponding to the lower limit of the adjustable frequency range. The unallocated air volume to be reduced is then prioritized for allocation to fans ranked later in the marginal energy consumption order that meet the unit reduction operation status verification. Once a subsequent fan reaches its upper limit of reducible air volume, the remaining air volume to be reduced is allocated to the fan ranked higher in the order that meets the unit reduction operation status verification, until all air volumes to be reduced are allocated. Specifically: ; in, This indicates the order of marginal energy consumption. The air volume to be reduced by the fan; This indicates that the airflow needs to be reduced. This indicates the order of marginal energy consumption. The air volume that has been allocated to the fan for reduction; This indicates the order of marginal energy consumption. The upper limit of air volume can be reduced for the fan within the adjustable frequency range.

[0079] After the incremental air volume of the fan unit is allocated, the fan allocated the incremental air volume is designated as the fan with increased output, and the corresponding frequency adjustment amount is determined based on the allocated incremental air volume, the current actual air volume, and the adjustable frequency range. After the air volume to be reduced is allocated, the fan allocated the air volume to be reduced is designated as the fan with reduced output, and the corresponding frequency adjustment amount is determined based on the allocated air volume to be reduced, the current actual air volume, and the adjustable frequency range. When the incremental air volume of the fan unit is zero, neither incremental air volume nor air volume to be reduced is allocated.

[0080] The valve adjustment status is kept connected to the affected dust removal area and consistent with the output adjustment direction of the corresponding fan; the marginal energy consumption ranking, the incremental air volume allocated to each fan, the air volume to be reduced allocated to each fan, the fan output adjustment direction, the frequency adjustment amount, and the valve adjustment status are summarized to form the fan group execution plan.

[0081] Figure 6The relationship between marginal energy consumption evaluation values ​​and allocated incremental air volume under different control methods is shown. It can be seen that the method of the present invention enables fans with lower marginal energy consumption evaluation values ​​to undertake more incremental air volume, and fans with higher marginal energy consumption evaluation values ​​to undertake less incremental air volume. The fixed priority allocation method and the average allocation method do not reflect the differences in energy consumption and dust removal constraints between fans. This shows that the present invention can coordinate the adjustment of fan units according to the comprehensive evaluation results of energy consumption and dust removal constraints, while meeting the target demand air volume, pressure difference limit and dust concentration control requirements, reducing ineffective energy consumption and improving control stability.

[0082] Figure 6 The "method of this invention" refers to a control method that, after determining the target required air volume, generates a marginal energy consumption evaluation value by comprehensively considering the unit incremental power consumption of each fan, the change in pipeline pressure difference, and the change in dust concentration control margin. The incremental air volume of the fan group is then allocated from low to high according to the marginal energy consumption evaluation value. The "fixed priority allocation method" refers to a control scheme that allocates incremental air volume according to a preset fan number or fixed order, without dynamically adjusting the allocation order based on changes in current energy consumption and dust removal constraints. The "average allocation method" refers to a baseline scheme that evenly allocates the incremental air volume of the fan group to the participating fans, without distinguishing the differences in incremental power consumption, pressure difference impact, and dust concentration margin occupation of different fans. F1, F2, F3, and F4 represent the fans participating in energy efficiency matching, respectively.

[0083] Figure 5 and Figure 6 The corresponding data all come from historical operating data and control execution records within the same dust removal area, and the collection cycle is consistent with the collection cycle of fan operating parameters, pipeline constraint parameters, and dust removal effect parameters; the test scenarios include the process of production cycle changing from a stable operating state to a changing operating state, and from a changing operating state back to a stable operating state; the fans participating in energy efficiency matching are... Figure 6 F1, F2, F3, and F4 are shown; Figure 5 The comparative scheme is to control the actual output air volume in real time with a threshold. The control rule is not to predict the load increment for the next control cycle, but to adjust the fan output only after the air volume deviation, pressure difference change or dust concentration change reach the trigger condition. Figure 6 The fixed priority allocation method allocates incremental air volume according to the preset fan number or fixed order, while the average allocation method distributes the incremental air volume of the fan group equally to the fans participating in the adjustment. The air volume tracking error is calculated based on the absolute difference between the actual output air volume and the actual required air volume within the same collection time range. The energy saving rate is calculated as the proportion of the difference between the total collection cycle power consumption of the control scheme and the total collection cycle power consumption of the present invention within the same collection time range to the total collection cycle power consumption of the control scheme.

[0084] When the target required air volume is lower than the current actual total air volume of the fan unit, this invention allocates the air volume to be reduced from back to front according to the marginal energy consumption ranking, so that the fans with higher marginal energy consumption evaluation values ​​and greater dust removal constraints are given priority to reduce output, thereby reducing ineffective air supply and high marginal energy consumption operation during the load reduction phase.

[0085] It should also be noted that existing technologies typically allocate loads based on fan power, frequency, or manual experience, easily overlooking the differences in power consumption per unit incremental air volume and the impact of dust removal constraints, leading to high energy consumption, pressure fluctuations, and inefficient fan operation. This solution determines the incremental air volume of the fan unit by determining the target required air volume, calculates the incremental power consumption of each fan, the change in pipeline pressure difference, and the change in dust concentration control margin, generates marginal energy consumption evaluation values, and allocates air volumes in a ranked manner. This allows for the priority allocation of incremental loads to fans with low marginal energy consumption and minimal impact from dust removal constraints, ensuring that the fan unit execution plan simultaneously meets the target required air volume, pressure difference limits, and dust concentration control requirements, thereby improving the accuracy and stability of the coordinated energy-saving control of the fan unit.

[0086] S4. After safety verification, the fan unit execution plan is converted into control instructions for PLC, DCS, frequency converter and valve actuator. The predicted dust removal load is compared with the dust removal load baseline to obtain the predicted dust removal load change. When the predicted dust removal load change exceeds the control dead zone and meets the frequency change limit, pressure difference limit and interlock protection conditions, the fan determined to increase output is controlled to increase output according to the fan unit execution plan, and the fan determined to decrease output is controlled to decrease output. The process event load sample library is updated with the measured air volume, pipeline pressure difference, dust concentration and fan power consumption.

[0087] S4.1. Based on the fan output adjustment direction, frequency adjustment amount, and valve adjustment status in the fan unit execution plan, the execution content to be verified is formed. The predicted dust removal load is compared with the current dust removal load baseline to obtain the predicted dust removal load change.

[0088] It should be noted that the fan output adjustment direction, frequency adjustment amount, and valve adjustment status are organized according to the fan number to form the execution content to be verified; the execution content to be verified represents the adjustment actions that each fan and corresponding valve is prepared to perform in the current control cycle.

[0089] The fan output adjustment direction includes increasing and decreasing the output; the frequency adjustment amount is determined by the incremental air volume allocated in the fan group execution plan, the air volume to be reduced, the current actual air volume, and the adjustable frequency range of the corresponding fan; the valve adjustment state is kept connected to the affected dust removal area and consistent with the output adjustment direction of the corresponding fan.

[0090] The predicted dust removal load is jointly characterized by the predicted air volume, predicted pressure difference, predicted dust concentration, and predicted power consumption of the next control cycle; the current dust removal load baseline is jointly characterized by the current air volume baseline, the current power consumption baseline of the current acquisition cycle, the pressure difference corresponding to the start time of the current production event, and the dust concentration corresponding to the start time of the current production event; the difference between the predicted dust removal load and the current dust removal load baseline is compared to obtain the predicted dust removal load change, specifically: ; in, This indicates the predicted change in dust removal load for the next control cycle.

[0091] The differences in air volume, pressure difference, dust concentration, and power consumption during the sampling cycle in the predicted changes in dust removal load are all calculated using absolute values. This is used to characterize the overall degree of change in the next control cycle relative to the current dust removal load baseline, thus avoiding the mutual cancellation of differences in different directions of change.

[0092] S4.2 When the predicted change in dust removal load exceeds the control dead zone, the frequency adjustment amount, pipeline pressure difference and interlock status corresponding to the execution content to be verified are verified for safety. After the verification results meet the frequency change limit, pressure difference limit and interlock protection conditions, the execution content to be verified is converted into control instructions for PLC, DCS, frequency converter and valve actuator.

[0093] It should be noted that in the process event load sample library, stable operation samples with no change in production cycle time, no adjustment in fan output, and dust concentration meeting control requirements are selected; based on the air volume, pressure difference, dust concentration, and power consumption of the collection cycle in the stable operation samples, the change in dust removal load between adjacent collection times is determined, and the natural fluctuation range of load under stable operation conditions is obtained.

[0094] Based on the minimum executable adjustment of the PLC, DCS, frequency converter, and valve actuator, the dust load change corresponding to a minimum control action is determined; among them, the minimum frequency adjustment step of the frequency converter corresponds to the air volume change, and the minimum opening adjustment step of the valve actuator corresponds to the pressure difference change; the upper limit of the natural fluctuation range of the load under stable operation is compared with the dust load change corresponding to a minimum control action, and the maximum value of the two is determined as the control dead zone.

[0095] The control dead zone is used to distinguish between effective load changes caused by production events and natural fluctuations under stable operating conditions. When the predicted change in dust removal load does not exceed the control dead zone, no new control commands are issued in the current control cycle, and the fan operation status and valve adjustment status remain unchanged. When the predicted change in dust removal load exceeds the control dead zone, the content to be verified enters the safety verification.

[0096] The expected pipeline pressure difference after execution is determined based on the fan output adjustment direction, frequency adjustment amount, and valve adjustment status in the execution content to be verified. Specifically, the frequency adjustment amount of each fan is converted into the corresponding number of unit operating status changes according to adjacent executable frequency points. When the fan output adjustment direction is to increase output, the pipeline pressure difference change corresponding to the unit incremental operating status is called. When the fan output adjustment direction is to decrease output, the pipeline pressure difference change corresponding to the unit decreasing operating status is called. When multiple fans are adjusted simultaneously, the pipeline pressure difference changes corresponding to each fan connected to the affected dust removal area are superimposed according to the adjustment direction, and the superposition result is added to the pressure difference corresponding to the start time of the current production event to obtain the expected pipeline pressure difference after execution. When multiple valves are adjusted simultaneously, only the pipeline pressure difference changes corresponding to the valve adjustment status connected to the affected dust removal area are included in the superposition.

[0097] Safety verification includes frequency change limit verification, differential pressure limit verification, and interlock protection condition verification. Frequency change limit verification is performed one by one according to the fan number, comparing the frequency adjustment amount with the corresponding fan's allowed frequency change limit. If the frequency adjustment amount does not exceed the frequency change limit, the frequency change limit verification is considered passed. Differential pressure limit verification determines the expected pipeline differential pressure after execution based on the fan output adjustment direction, frequency adjustment amount, and valve adjustment status corresponding to the content to be verified. If the expected pipeline differential pressure after execution does not exceed the differential pressure limit, the differential pressure limit verification is considered passed. Interlock protection condition verification confirms the fan operating status, inverter operating status, valve actuator status, and dust removal area interlock status. If the fan, inverter, valve actuator, and dust removal area are all in an allowable adjustment state, the interlock protection condition verification is considered passed.

[0098] Once the frequency variation limit verification, differential pressure limit verification, and interlock protection condition verification all pass, the content to be verified is converted into control instructions. Among them, the PLC is a programmable logic controller, which is responsible for the on-site sequential control and interlock control of the fan, frequency converter, and valve actuator. The DCS is a distributed control system, which is responsible for the centralized monitoring and coordinated control of the fan operating status, pipeline differential pressure, dust concentration, and control instruction execution status within the dust removal area. The control instructions include the fan number, fan output adjustment direction, frequency adjustment amount, valve adjustment status, and execution time sent to the PLC and DCS, and correspondingly form frequency control instructions sent to the frequency converter and valve adjustment instructions sent to the valve actuator.

[0099] If any of the verifications—frequency change limit verification, differential pressure limit verification, and interlock protection condition verification—fail, the control command corresponding to the content to be verified will not be issued, and the current fan operating status and valve adjustment status will be maintained.

[0100] S4.3. Control the fans determined to increase output according to the control command, and control the fans determined to decrease output to decrease output. During the execution of the control command, collect the measured air volume, pipeline pressure difference, dust concentration and fan power consumption, and update the actual load response process after the execution of the control command to the process event load sample library.

[0101] It should be noted that the PLC and DCS control the fan and valve actuator according to the fan number in the control command, and send the frequency adjustment amount to the corresponding strain gauge and the valve adjustment status to the corresponding valve actuator.

[0102] For fans determined to increase output, the frequency converter increases the operating frequency according to the frequency adjustment amount in the control command, thereby increasing the output of the corresponding fan; for fans determined to decrease output, the frequency converter decreases the operating frequency according to the frequency adjustment amount in the control command, thereby decreasing the output of the corresponding fan; the valve actuator operates according to the valve adjustment state in the control command, keeping the load channel of the affected dust removal area connected, and matching the valve adjustment state with the output adjustment direction of the corresponding fan.

[0103] During the execution of control commands, the measured air volume, pipeline pressure difference, dust concentration, and fan power consumption are continuously recorded according to the acquisition time after the execution time; among them, fan power consumption is recorded according to the power consumption of the acquisition cycle. The measured air volume represents the change in actual ventilation load after the execution of control commands, the pipeline pressure difference represents the change in pipeline constraints after the execution of control commands, the dust concentration represents the change in dust removal effect after the execution of control commands, and the power consumption of the acquisition cycle represents the change in energy consumption after the execution of control commands.

[0104] The measured air volume, pipeline pressure difference, dust concentration, and power consumption during the data collection cycle are correlated in a time sequence according to the collection time to form the actual load response process after the control command is executed. The actual load response process is then collected and updated to the process event load sample library after corresponding with the current production event, the affected dust removal area, the fan unit execution plan, and the control command. This update supplements the process event load sample library with the actual response data of the current production event after control execution and serves as a historical sample for subsequent similar matching of the current production event and prediction of the load increment in the next control cycle.

[0105] This embodiment also provides a computer device applicable to the wind turbine energy-saving control method based on load forecasting, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the wind turbine energy-saving control method based on load forecasting as proposed in the above embodiment.

[0106] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0107] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the wind turbine energy-saving control method based on load forecasting as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0108] In summary, this invention achieves this by: matching the current production event, affected dust removal area, dust removal load baseline, and dust removal control margin with historical samples in the process event load sample library; and predicting the load increment for the next control cycle based on the load response curve and similarity weight. This allows for the early determination of the predicted dust removal load and target air volume demand, reducing load judgment lag and excessive air supply. By calculating the incremental power consumption and dust removal constraint changes corresponding to the unit incremental air volume of each fan, marginal energy consumption evaluation values ​​are generated and sorted to form a fan group execution plan. This enables the fan group to coordinately adjust according to the comprehensive evaluation results of energy consumption and dust removal constraints, reducing ineffective energy consumption and improving the accuracy and stability of energy-saving control of dust removal fans while meeting air volume demand, pressure difference limits, and dust concentration control requirements.

[0109] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A wind turbine energy-saving control method based on load forecasting, characterized in that, include: Within the same dust removal area, identify fan units with load coordination relationships, mark historical production events according to changes in production cycle from historical operating data, extract fan operating parameters, pipeline constraint parameters, and dust removal effect parameters before and after the events, and correlate the fluctuations in air volume, pressure difference, dust concentration, and power consumption in sequence to form a process event load sample library; Within the current control cycle, the current production event is identified based on the change in production cycle time, and the affected dust removal area is determined by the valve status. A dust removal load baseline is formed based on the current fan operating parameters, and a dust removal control margin is formed based on the current pipeline constraint parameters and dust removal effect parameters. The current production event, affected dust removal area, dust removal load baseline, and dust removal control margin are matched with historical samples in the process event load sample library to determine historical similar samples. The load increment of the next control cycle is predicted according to the load response curve and similarity weight of the historical similar samples, thus obtaining the predicted dust removal load and target demand air volume. The target demand air volume triggers the energy efficiency matching of the fan unit. Combining the fan operating parameters, adjustable frequency range, valve adjustment status and pipeline constraint parameters of each fan, the incremental power consumption and dust removal constraint change corresponding to the unit incremental air volume are calculated, the marginal energy consumption evaluation value is generated, and the evaluation values ​​are sorted from low to high to obtain the fan unit execution plan. After safety verification, the fan unit execution plan is converted into control instructions for PLC, DCS, frequency converter and valve actuator. The predicted dust removal load is compared with the dust removal load baseline to obtain the predicted dust removal load change. When the predicted dust removal load change exceeds the control dead zone and meets the frequency change limit, pressure difference limit and interlock protection conditions, the fan determined to increase output is controlled to increase output according to the fan unit execution plan, and the fan determined to decrease output is controlled to decrease output. The process event load sample library is updated with the measured air volume, pipeline pressure difference, dust concentration and fan power consumption.

2. The wind turbine energy-saving control method based on load forecasting as described in claim 1, characterized in that, The specific steps for forming the process event load sample library are as follows: Within the same dust removal area, identify fan units with load coordination relationships, align fan operating parameters, pipeline constraint parameters, dust removal effect parameters, and production cycle changes according to the collection time to obtain historical operating sequence data for the area; Among them, the fan units with load coordination relationship are determined based on the connection relationship between the fan and the dust removal pipeline, the valve adjustment range, and the dust removal points that work together; Historical production events are marked by changes in production cycle time in the historical runtime data of the region, and the fan operation parameters, pipeline constraint parameters and dust removal effect parameters before, during and after the historical production events are extracted to obtain historical event parameter fragments. By correlating the air volume, pressure difference, dust concentration, and power consumption in the historical event parameter fragments in chronological order, the load response relationship corresponding to the historical production events is obtained, and a sample library of process event loads is compiled.

3. The wind turbine energy-saving control method based on load forecasting as described in claim 1, characterized in that, The specific steps for establishing the dust removal load baseline are as follows: Within the current control cycle, the moment when the production cycle change state changes from a stable operating state to a changing operating state is marked as the start time of the current production event, and the process cycle change content corresponding to the production cycle change state is marked as the current production event. The dust collection area corresponding to the valve whose valve status changes is identified as the affected dust collection area. The data collection time that was in a stable operating state before the start time of the current production event was selected as the baseline time. The air volume and power consumption of the data collection cycle corresponding to the baseline time were used to form the dust removal load baseline. The remaining space between the differential pressure and the differential pressure limit at the start time of the current production event is determined as the differential pressure margin, and the remaining space between the dust concentration and the upper limit of the dust concentration control at the start time of the current production event is determined as the dust concentration margin. The differential pressure margin and the dust concentration margin together form the dust removal control margin.

4. The wind turbine energy-saving control method based on load forecasting as described in claim 3, characterized in that, The specific steps for determining historical similar samples are as follows: Historical samples that are consistent with current production events and historical production events, and whose affected dust removal areas are consistent with the dust removal areas corresponding to historical production events, are identified as candidate historical samples. The current air volume baseline and the current power consumption baseline in the dust removal load baseline are compared with the air volume benchmark value and the power consumption benchmark value in the candidate historical samples, respectively. The pressure difference margin and dust concentration margin in the dust removal control margin are compared with the pressure difference margin and dust concentration margin in the candidate historical samples, respectively, to obtain the matching distance. Candidate historical samples are sorted from low to high according to matching distance, and the candidate historical samples at the top of the ranking are selected to determine historical similar samples. Similarity is calculated based on the matching distance of historical similar samples, where a lower matching distance corresponds to a higher similarity. The similarity of each historical similar sample is then normalized to obtain a similarity weight.

5. The wind turbine energy-saving control method based on load forecasting as described in claim 4, characterized in that, The specific steps for obtaining the predicted dust removal load and target required air volume are as follows: Extract the load change segment corresponding to the next control cycle from the load response curves of similar historical samples; The load change segments of each historical similar sample are weighted and fused according to the similarity weight to obtain the air volume increment, pressure difference increment, dust concentration increment and power consumption increment of the next control cycle. The load increment of the next control cycle is superimposed on the dust removal load baseline to obtain the predicted air volume, predicted pressure difference, predicted dust concentration and predicted power consumption of the next control cycle, and form the predicted dust removal load. When the predicted differential pressure does not exceed the differential pressure limit and the predicted dust concentration does not exceed the upper limit of dust concentration control, the predicted air volume is determined as the target required air volume. When either the predicted differential pressure exceeds the differential pressure limit or the predicted dust concentration exceeds the upper limit of dust concentration control, the load change segment where both differential pressure and dust concentration are within the limit range is determined from the load response curves of similar historical samples, and the corresponding air volume change is fused according to the similarity weight to determine the target required air volume.

6. The wind turbine energy-saving control method based on load forecasting as described in claim 1, characterized in that, The specific steps for generating the marginal energy consumption evaluation value are as follows: Determine the incremental air volume of the fan unit based on the target required air volume and the current actual total air volume of the fan unit. In a fan unit with load coordination relationship, the fan that is determined to be in normal operation, has adjustable frequency range, has valve adjustment status that is connected to and adjustable in the affected dust removal area, and meets the differential pressure limit of the pipeline constraint parameters is the fan to participate in energy efficiency matching. For wind turbines involved in energy efficiency matching, determine the incremental power consumption, pipeline pressure difference change, and dust concentration control margin change under unit incremental operating conditions; Based on the incremental power consumption, the amount of pipeline pressure difference change relative to the pressure difference margin, and the amount of dust concentration control margin change relative to the dust concentration margin, the marginal energy consumption evaluation value corresponding to each fan participating in energy efficiency matching is generated.

7. The wind turbine energy-saving control method based on load forecasting as described in claim 6, characterized in that, The specific steps for obtaining the wind turbine unit execution plan are as follows: The wind turbines participating in energy efficiency matching are ranked from low to high according to their marginal energy consumption evaluation values, thus obtaining the marginal energy consumption ranking. When the incremental air volume of the wind turbine unit is greater than zero, the incremental air volume of the wind turbine unit is allocated from front to back according to the marginal energy consumption ranking, and the incremental air volume of the wind turbine unit that has not yet been allocated is preferentially allocated to the wind turbines in the front order of the marginal energy consumption ranking. Once the current fan reaches its allocable incremental limit, the remaining incremental air volume is allocated to the next fan until the incremental air volume of the fan group is fully allocated. The fan allocated to the incremental air volume is identified as the fan that increases output, and the corresponding frequency adjustment amount is determined based on the allocated incremental air volume, the current actual air volume, and the adjustable frequency range. When the incremental air volume of the fan unit is less than zero, the absolute value of the incremental air volume of the fan unit is determined as the air volume to be reduced. The fans participating in energy efficiency matching are verified according to the unit reduction operation status corresponding to the reduction output direction. The air volume to be reduced is allocated to the fans that meet the verification from the back to the front according to the marginal energy consumption ranking. The air volume to be reduced that has not yet been allocated is preferentially allocated to the fans that meet the verification in the later order of the marginal energy consumption ranking. The upper limit of the air volume that can be reduced is determined based on the difference between the current actual air volume and the air volume corresponding to the lower limit of the adjustable frequency range; Once the subsequent fan reaches the upper limit of the air volume that can be reduced, the remaining air volume to be reduced is allocated to the previous fan until the air volume to be reduced is completely allocated. The fan assigned to reduce the air volume is identified as the fan with reduced output, and the corresponding frequency adjustment amount is determined based on the assigned air volume to be reduced, the current actual air volume, and the adjustable frequency range. The execution plan for the fan unit is formed by summarizing the marginal energy consumption ranking, the incremental air volume allocated to each fan, the air volume to be reduced allocated to each fan, the fan output adjustment direction, the frequency adjustment amount, and the valve adjustment status.

8. The wind turbine energy-saving control method based on load forecasting as described in claim 7, characterized in that, The process involves controlling fans that are determined to increase output according to the fan unit execution plan to increase their output, and controlling fans that are determined to decrease output to decrease their output. The process event load sample database is updated with measured air volume, pipeline pressure difference, dust concentration, and fan power consumption. The specific steps are as follows: Based on the fan output adjustment direction, frequency adjustment amount, and valve adjustment status in the fan unit execution plan, the execution content to be verified is formed, and the predicted dust removal load is compared with the current dust removal load baseline to obtain the predicted dust removal load change amount. When the predicted change in dust removal load exceeds the control dead zone, a safety check is performed on the frequency adjustment amount, the expected pipeline pressure difference and interlock status corresponding to the content to be checked; When the frequency adjustment amount meets the frequency change limit, the differential pressure of the pipeline network is expected to meet the differential pressure limit after execution, and the interlock status meets the interlock protection conditions, the execution content to be verified is converted into the fan number, fan output adjustment direction, frequency adjustment amount, valve adjustment status and execution time sent to the PLC and DCS, and frequency control instructions are sent to the frequency converter and valve adjustment instructions are sent to the valve actuator. According to the frequency control command, the fan determined to increase output increases output, and the fan determined to decrease output decreases output. The valve actuator is controlled to operate according to the valve adjustment command. During the execution of frequency control commands and valve adjustment commands, the measured air volume, pipeline pressure difference, dust concentration and fan power consumption are correlated according to the acquisition time to form the actual load response process, and the actual load response process is updated to the process event load sample library.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the wind turbine energy-saving control method based on load forecasting as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the wind turbine energy-saving control method based on load forecasting as described in any one of claims 1 to 8.

Images