Metallographic polishing liquid supply control method based on polishing liquid concentration feedback

Abstract

The application provides a metallographic polishing liquid supply control method based on polishing liquid concentration feedback, comprising: obtaining current operating states of each switching valve, circulating pump and liquid supply branch, determining a current sampling path from a mixed liquid source position to a concentration detection point and a current liquid supply path to a polishing liquid supply point, and determining common source relationship and common mixing relationship of the two paths; comparing the current path with a corresponding path before switching to determine a target pipe section newly entering the current sampling path or the current liquid supply path due to flow path switching, and determining a previous residual liquid state according to an effective liquid holding capacity of the target pipe section and a polishing liquid state before switching. The application simultaneously checks the common source relationship, common mixing relationship of the sampling path and the liquid supply path, and the material replacement state after flow path switching, so that the concentration detection value only participates in proportional control when it can correspond to the current actual liquid supply liquid, thereby reducing improper liquid supply correction caused by the previous residual liquid.

Description

Technical Field

[0001] This invention relates to the field of fluid proportioning control technology, and in particular to a metallographic polishing fluid supply control method based on polishing fluid concentration feedback. Background Technology

[0002] In metallographic polishing, a polishing slurry, composed of a concentrate and a diluent in a specific ratio, is continuously supplied to the polishing station. The concentration of the polishing slurry is monitored by concentration sensors installed in the circulation pipeline or sampling branch, and the supply ratio of the two liquids is adjusted based on the monitoring results. During long-term operation of the supply system, switching valves, circulation pumps, and different supply branches may change their operating states to accommodate filtration, circulation, cleaning, station switching, or adjustments to the supply path, thus altering the actual flow path between the concentration detection point and the polishing supply point. Such switching is part of the normal operation of the supply system and does not indicate a malfunction of the concentration sensor or the supply actuator.

[0003] When the flow path is switched, the liquid sampled at the concentration detection point may temporarily no longer originate from the same flow path as the liquid actually supplied to the polishing supply point. Even if the valves and pumps have been switched, and the sampling path and supply path once again share a common source in their connection relationship, the newly activated pipe sections, valve chambers, or other liquid-holding spaces may still retain polishing fluid from before the switch. This residual liquid will be gradually replaced as the current polishing fluid continues to flow. Therefore, during the transition phase after the flow path switch, the concentration sensor itself can output a stable detection value, but the liquid corresponding to this detection value may not be from the same source and material state as the current actual supply liquid.

[0004] If the current concentration measurement value is directly used for adjusting the ratio of concentrate to diluent during the aforementioned stage, the concentration changes of residual liquid before the switch or liquid not corresponding to the current supply may be interpreted as a deviation in the current supply concentration, leading to unnecessary or inappropriate ratio corrections. The reason is not that the concentration measurement itself is inaccurate, but rather that after the flow path change, the correspondence between the measured liquid and the actual supply liquid needs to be re-established and confirmed. Therefore, how to determine whether the current concentration measurement result can truly represent the current actual supply liquid after a switch in the supply flow path, and how to prevent unrepresentative measurement results from prematurely participating in the supply ratio control, is a technical problem that needs to be solved in metallographic polishing supply control. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a metallographic polishing solution supply control method based on polishing solution concentration feedback, thereby solving the technical problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The metallographic polishing solution supply control method based on polishing solution concentration feedback includes the following steps:

[0008] S1. Obtain the current operating status of each switching valve, circulating pump and liquid supply branch, determine the current sampling path from the source of the mixed liquid to the concentration detection point and the current liquid supply path to the polishing liquid supply point, and determine the common source relationship and common mixing relationship of the two paths;

[0009] S2. Compare the current sampling path and the current liquid supply path with the corresponding path before the switch to determine the target pipe segment that has been newly entered into the two paths due to the switch, and determine the state of the residual liquid in the previous sequence based on the effective liquid holding capacity of the target pipe segment and the state of the polishing liquid before the switch.

[0010] S3. Obtain the actual liquid flow rate through the target pipe section after switching, determine the cumulative liquid volume, and update the previous residual liquid state according to the cumulative liquid volume and the effective liquid holding volume to obtain the material replacement state of the current sampling path and the current liquid supply path.

[0011] S4. Based on the common source relationship, common mixing relationship and material replacement state, determine whether the sampling liquid and the supply liquid corresponding to the current concentration detection value have the same flow path source and corresponding material state, and determine the concentration detection value obtained after the common source relationship and common mixing relationship are established and the current sampling path and the current supply path meet the corresponding material replacement requirements as the effective concentration detection value.

[0012] S5. Based on the deviation between the effective concentration detection value and the target polishing fluid concentration and the current total supply volume, determine the target supply volume of the concentrate and diluent and control the supply branch; for concentration detection values ​​that do not meet the requirements of common source relationship, common mixing relationship or corresponding material replacement, do not adjust the amount based on the formation of a new supply ratio.

[0013] The beneficial effects of this invention are as follows:

[0014] This invention simultaneously verifies the common source relationship and mixing relationship between the sampling path and the liquid supply path, as well as the material replacement state after flow path switching. This ensures that the concentration detection value only participates in proportional control when it corresponds to the current actual liquid supply, reducing improper liquid supply corrections caused by residual liquid from the previous flow path. By identifying the target pipe segment newly entering the sampling or liquid supply path after flow path switching and updating its replacement state based on the effective liquid holding capacity and the actual cumulative liquid volume, this invention distinguishes the transition process where the flow path connection is completed but the residual liquid has not been fully replaced, avoiding the need to judge whether the liquid has synchronized solely based on the valve or pump status.

[0015] This invention correlates effective liquid holding capacity with clearly defined residual liquid replacement criteria, and employs verifiable determination methods for regular pipe sections and pipe sections with local volume or backmixing effects. This provides an executable basis for replacement judgment under different flow path structures, reducing judgment deviations caused by inconsistencies between fixed waiting times and actual replacement processes. By associating the flow path source conditions, material state conditions, and the collection time and path version of concentration detection values, it prevents the retrospective use of historical concentration data obtained during the switching transition phase after material replacement is completed, ensuring that the concentration information used in the liquid supply ratio calculation corresponds to the current flow path state over a specified period.

[0016] This invention determines the change in the concentrate ratio based on the deviation between the effective concentration detection value and the target polishing slurry concentration. It then calculates the target supply amounts of concentrate and diluent by combining the total supply volume and the executable range of the two branches. This ensures that concentration feedback is translated into specific branch control quantities, while avoiding the generation of control commands exceeding the actuator's capabilities. When the current concentration detection value is not representative of the supply or when new changes occur in the flow path, the invention avoids accumulating new concentration corrections using that detection value. Instead, it maintains the actual supply state that can be confirmed by the execution feedback. This reduces repeated ratio adjustments caused by continuous ineffective feedback and preserves a stable control basis for obtaining a new effective concentration detection value. Attached Figure Description

[0017] Figure 1 This is a flowchart of the metallographic polishing solution supply control method based on polishing solution concentration feedback according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figure 1 As shown, this embodiment provides a metallographic polishing solution supply control method based on polishing solution concentration feedback, including the following steps:

[0020] S1. Obtain the current operating status of each switching valve, circulating pump and liquid supply branch, determine the current sampling path from the source of the mixed liquid to the concentration detection point and the current liquid supply path to the polishing liquid supply point, and determine the common source relationship and common mixing relationship of the two paths;

[0021] S2. Compare the current sampling path and the current liquid supply path with the corresponding path before the switch to determine the target pipe segment that has been newly entered into the two paths due to the switch, and determine the state of the residual liquid in the previous sequence based on the effective liquid holding capacity of the target pipe segment and the state of the polishing liquid before the switch.

[0022] S3. Obtain the actual liquid flow rate through the target pipe section after switching, determine the cumulative liquid volume, and update the previous residual liquid state according to the cumulative liquid volume and the effective liquid holding volume to obtain the material replacement state of the current sampling path and the current liquid supply path.

[0023] S4. Based on the common source relationship, common mixing relationship and material replacement state, determine whether the sampling liquid and the supply liquid corresponding to the current concentration detection value have the same flow path source and corresponding material state, and determine the concentration detection value obtained after the common source relationship and common mixing relationship are established and the current sampling path and the current supply path meet the corresponding material replacement requirements as the effective concentration detection value.

[0024] S5. Based on the deviation between the effective concentration detection value and the target polishing fluid concentration and the current total supply volume, determine the target supply volume of the concentrate and diluent and control the supply branch; for concentration detection values ​​that do not meet the requirements of common source relationship, common mixing relationship or corresponding material replacement, do not adjust the amount based on the formation of a new supply ratio.

[0025] S1 specifically includes the following sub-steps:

[0026] S110. The industrial control software reads the operating status of equipment related to the liquid flow direction in the metallographic polishing liquid supply system and calls the established flow path connection data. Equipment operating status includes the actual position of each switching valve, the start / stop status of the circulating pump and the liquid supply pump, and whether each liquid supply branch is in a permitted liquid supply state. The flow path connection data is established according to the actual pipeline of the equipment and is used to describe the physical connection relationship between the liquid storage area, mixing area, each pipe section, switching valve, pump, concentration detection point, and polishing liquid supply point. Each pipe section must at least record the pipe section identifier, upstream connection node, downstream connection node, and permitted flow direction.

[0027] For switching valves that can change the connection relationship, record the connection nodes corresponding to each working position; for circulating pumps and liquid supply pumps, record the pipe section they are located in and the flow direction that is allowed to be established during normal operation. The above connection data should be entered according to the piping diagram and the on-site connection situation during equipment installation and commissioning, and should be re-verified after pipeline modification, valve replacement, or change of connection relationship.

[0028] Equipment status is recorded according to a unified time base of the industrial control software. When status data returned by different controllers has timestamps, the timestamps are used to confirm their sequence; when there are no independent timestamps, the moment when the control software actually receives and confirms the completion of the equipment execution is taken as the status moment. If a valve switching command has been issued but feedback has not yet been received, or a pump start / stop command has not yet received confirmation of its operating status, the command value is not directly taken as the actual status of the equipment.

[0029] At this point, the current flow path status is recorded as undetermined. The most recently confirmed path record is only saved as a historical comparison benchmark before the flow path switch and will not be used as the current valid path. Concentration detection values ​​obtained during this undetermined state will not be included in the new liquid supply ratio adjustment.

[0030] The concentration detection point is based on the actual location of the polishing fluid being tested, obtained by the concentration sensor. The polishing fluid supply point is based on the supply interface before the liquid enters the current polishing station. When the equipment has multiple polishing stations, the corresponding polishing fluid supply point is selected according to the current station status. If it includes a CMP polishing station, the supply interface of that station is still treated as a specific polishing fluid supply point according to the same rules, without changing the subsequent path identification and concentration representation qualification judgment method. When the equipment is started for the first time, the first current flow path status record is established only after all valves, pumps, and supply branches involved in the current flow path judgment have obtained a verifiable status.

[0031] S120. Using the source location of the mixed solution as the starting point of the path, and the concentration detection point and the current polishing liquid supply point as the ending points of the path, the current sampling path and the current liquid supply path are determined based on the confirmed equipment operating status. The source location of the mixed solution is the outlet of the storage area or the mixing area, which can serve as the common upstream source for both the sampling liquid and the supply liquid after the concentrate and diluent have completed their preliminary mixing. This location is determined based on the actual pipeline during equipment installation and commissioning, and is bound and saved with the flow path connection data.

[0032] When determining the path, first exclude connections blocked by closed switching valves and branches containing pumps without a defined flow direction from the flow path connection data. The remaining connections represent the flow paths that liquid is allowed to pass through under the current equipment status. Start from the source of the mixture and search segment by segment along the allowed flow direction towards the concentration detection point. At each valve, pump, or branch node, determine the subsequent passable pipe segment according to the currently confirmed equipment status until the concentration detection point is reached.

[0033] The current sampling path is obtained by saving the pipe segment identifiers and connection nodes traversed according to the actual flow sequence. The current polishing liquid supply point is located using the same method to obtain the current liquid supply path. Both paths record the start point, end point, pipe segments arranged in flow sequence, relevant equipment status, path determination time, and path version.

[0034] When multiple branches are simultaneously connected to the same endpoint, the actual branch participating in the flow is determined by combining pump status, valve status, and existing branch flow information. For branches with independent flow detection, whether the flow detection result reaches the range that the calibrated sensor can reliably distinguish from zero flow is used as an auxiliary confirmation criterion.

[0035] When there is no independent branch flow detection, all parallel branches that may actually participate in the flow are saved according to the confirmed connection status of the pump and valves, and no single branch is arbitrarily selected as the sole flow path. If a continuous flow path cannot be established from the source of the mixture to the concentration detection point or the polishing supply point, the corresponding path is marked as invalid. During the period of invalid path, no new supply ratio adjustment is generated based on the current concentration detection value.

[0036] S130. Compare the upstream sources and the actual flow paths that the current sampling path and the current supply path share to determine their common source relationship and common mixing relationship. If both paths start from the same mixed liquid source location and there is no independent upstream input that changes the liquid source of one path before reaching their respective endpoints, they are considered to have a common source relationship.

[0037] If the starting points are different, or if one of the paths reconnects to liquid from another storage area after separation, the common source relationship does not hold. When multiple branches are running in parallel, the source of each branch actually involved in the liquid supply should be verified; the fact that one branch meets the conditions cannot be used to represent all actual flow branches.

[0038] After a common source relationship is established, starting from the source location of the mixed liquid, compare the pipe sections and nodes that the two paths pass through in the direction of liquid flow to determine the last common node before separation. If the sampled liquid and the supplied liquid have passed through the same mixing area or the same continuous common pipe section before separation, and the mixing area or common pipe section has been verified for mixing uniformity under the equipment's permissible operating conditions, and there is no independent liquid input that enters only one path after separation and can change the state of the polishing liquid composition, then the two are considered to have a common mixing relationship. The mixing uniformity verification is conducted during equipment commissioning. Samples are taken at each actual outlet location corresponding to the separation point of the common flow path, and the same concentration metering caliber is used for measurement. Under the condition that the flow rate, liquid ratio, and operating status of the mixing device remain unchanged, it is confirmed that the concentration difference at each location is within the permissible measurement error range of the concentration detection method used, and remains for a period not less than the response time of the concentration sensor. The flow rate range, liquid ratio range, and operating status of the mixing device used during verification are saved along with the flow path connection data; if the current operating conditions are not within the verified range, the common mixing relationship is not considered to be established.

[0039] If the two paths have separated before co-mixing, or if only one side is reconnected to the concentrate, diluent, or other liquid that alters the composition of the polishing slurry after separation, then the co-mixing relationship is not valid. This judgment only indicates whether the two liquids have the same source and mixing basis in the spatial flow path; it does not determine whether the residual liquid before the flow path switching has been replaced by the current liquid.

[0040] The current sampling path, current supply path, common source relationships, common mixing relationships, last common node, path version, and path determination time are all stored together. The previously confirmed path record is retained separately as a pre-switch history record. A new current path version is generated only after all new valve or pump states are confirmed; when the flow path is in a switching transition state, the previous history record is not used to replace the current state. The above records are used for subsequent comparison of the actual pipe segment composition before and after the switch, and to determine which pipe segments have newly entered the current sampling path or current supply path due to this flow path switch.

[0041] S2 specifically includes the following sub-steps:

[0042] S210. Read the current path version and the historical path record confirmed last time before the flow path switch. Compare the sampling path and liquid supply path before and after the switch according to the same pipe segment identifier and actual flow direction. Pipe segments that are in an effective flow state before and after the switch, and whose liquid source and flow direction have not changed, are recorded as continuously used pipe segments. Pipe segments that no longer participate in the current sampling or liquid supply after the switch are recorded as exited pipe segments. Pipe segments that only enter the current sampling path or current liquid supply path after the switch, but did not perform the same liquid transport function in the corresponding effective path before the switch, are identified as target pipe segments.

[0043] If the same physical pipe segment was part of the liquid supply path before the switch and is part of the sampling path after the switch, or if its flow direction changes, not only should the pipe segment identification be compared, but the liquid source and flow direction before and after the switch should also be verified. Only when the source and flow direction remain consistent can the original material state be used; otherwise, it should still be treated as the target pipe segment. This can prevent the same physical pipe segment from incorrectly inheriting residual liquid states that are not applicable to the current flow path due to a change in function.

[0044] The path comparison uses the confirmed completion time of the switchover as a unified time reference. No new target pipe segment set is generated until the flow path switchover has been confirmed. In addition to the target pipe segments directly determined in this switchover, pipe segments still involved in the current flow that have not yet completed their previous replacement or whose status is unknown, as well as downstream pipe segments that may receive liquid discharged from the aforementioned segments, are all included in the target pipe segment set used for this replacement verification. Downstream pipe segments are searched segment by segment according to the current flow direction until the concentration detection point or polishing supply point is reached, including pipe segments that are continuously used but may receive residual liquid from the previous flow. Only when the current path is completely consistent with the historical path, all previous replacement requirements have been met, and there are no affected pipe segments as described above, is the target pipe segment set recorded as empty, and the residual liquid status is not reinitialized.

[0045] If historical records are missing before the system's initial startup or switchover, or if the historical records are earlier than the most recent unregistered manual pipeline adjustment, the previous liquid state will not be inferred from those records. After on-site verification and confirmation that the current flow path connection data is consistent with the actual pipeline, all pipe segments actually involved in the flow in the current sampling path and current liquid supply path will be included in the replacement verification; if the initial liquid state can be confirmed, the corresponding initialization record will be used; if it cannot be confirmed, it will be recorded as unknown. After completing the corresponding replacement and filling confirmation, the historical state insufficiency mark will be removed, and the currently confirmed path and material state will be saved as the subsequent comparison benchmark. Before this, the current concentration detection value will not obtain new liquid supply representative status.

[0046] Each target pipe segment records its segment identifier, its current sampling path or current liquid supply path, its participation status before switching, its flow direction after switching, its current liquid source, and the time of switching completion, so as to determine the effective liquid holding capacity and the state of the previous residual liquid.

[0047] S220. Determine the effective liquid holdup for the target pipe section. Effective liquid holdup is an equivalent liquid volume parameter corresponding to the residual liquid replacement criterion used in this scheme. It indicates how much current liquid needs to pass through the target pipe section after flow path switching in order to determine, according to the established detection criterion, that the impact of the residual liquid on the outlet state before the switching has been reduced to an acceptable range. Effective liquid holdup is not directly equivalent to geometric volume under all circumstances. For regular straight pipes with simple structure, clear flow direction, and whose commissioning results show that the geometric volume corresponds to the same replacement criterion, its geometric liquid holdup volume can be used as the effective liquid holdup.

[0048] The geometric liquid-holding volume of a regular circular pipe section is calculated using the following formula:

[0049]

[0050] in, This represents the geometric liquid holding volume of the j-th target pipe segment, in meters. 3 ; This indicates the actual inner diameter of the pipe section, in meters, and is derived from equipment drawings or on-site measurements. This indicates the effective length of the pipe segment that actually participates in liquid holding within the current flow path, expressed in meters (m), and is derived from flow path connection data and on-site verification results. If the pipe segment contains valve chambers, filter housings, detection chambers, or other internal spaces that cannot be reflected by the regular pipe segment dimensions, these spaces are not included in the aforementioned geometric volume.

[0051] For pipe sections with significant local volume or structural changes, or those prone to backmixing, the effective liquid holdup is calibrated using a volume displacement method. During calibration, the flow path to be calibrated is first filled with a liquid of known initial concentration, and then replaced with another liquid of stable and known concentration following the normal liquid supply direction of the equipment. Simultaneously, the cumulative input liquid volume and outlet concentration are recorded.

[0052] When the outlet concentration of the replacement fluid is within the sensor's allowable measurement error range corresponding to the known concentration, and remains within this range for at least the response time specified in the sensor's technical documentation, the corresponding cumulative input fluid volume is taken as the effective liquid hold-up for that pipe section. The unit is m 3 The rule-based straight tube only applies when similar permutation verification shows... Only when the above criteria are met can it be ordered If the inner diameter of the pipeline, valve body, filter assembly, or local cavity structure is changed, the corresponding parts should be recalibrated. Normal valve opening and closing does not change the established fixed liquid holding parameters.

[0053] When multiple permissible flow rates exist for the same target pipe section during normal equipment operation, or when changes in polishing fluid temperature and viscosity alter the displacement process within the pipe section, volume displacement calibration should be performed under the corresponding permissible operating conditions. Calibration should also cover the actual permissible flow direction for the pipe section, as well as the permissible concentration ranges of the preceding and current liquids used by the equipment. In cases of reverse use, the displacement process in both directions should be verified separately; the calibration result of one direction should not be directly used to replace the result of the other. The concentration difference of the calibration liquid should cover the maximum concentration difference in permissible switching scenarios, and ensure that the initial liquid and the displacement liquid can be reliably distinguished by the concentration detection method employed. The maximum cumulative input liquid volume required to reach the same residual liquid displacement criterion under the covered operating conditions should be determined as the effective liquid holdup of the target pipe section.

[0054] When multiple target pipe sections are used in series, the entire path should be verified for replacement according to the actual flow path and the rule of accumulating from upstream to downstream as described later. This verifies that after each pipe section completes the judgment based on its effective liquid holding capacity, the concentration at the end of the path meets the same replacement criterion. If not, the calibrated replacement liquid volume of the corresponding pipe section should be increased and re-verified. The verified effective liquid holding capacity and applicable flow path should be saved. If the specific value of the preceding concentration is unknown, but the equipment medium usage record confirms that it is within the calibrated range, the effective liquid holding capacity covering that range can be used. If even the medium type or concentration range cannot be confirmed, the unknown state should not be cleared solely by the accumulated liquid volume. If the current operating condition exceeds the calibrated range, the corresponding material replacement state should be recorded as unknown. Replacement should only be confirmed after the operating condition returns to the calibrated range and the required replacement is completed again, or after recalibration is completed.

[0055] S230. Determine the state of the preceding residual liquid based on the last actual liquid source connected to the target pipe segment before the flow path switching. The state of the preceding residual liquid describes the liquid that may still occupy the target pipe segment before the switching is completed, and includes at least the target pipe segment identifier, effective liquid holding capacity, preceding polishing fluid concentration, preceding liquid source, and the time when the state was formed.

[0056] The concentration of the preceding polishing fluid is preferentially based on the concentration test result obtained when the pipe section was in an effective flow state and qualified as a representative of the fluid supply. If the system is starting for the first time and there is no historical effective concentration record, after the initial fluid preparation is completed and stable circulation is achieved, the liquid at the source of the mixture is directly sampled and measured using the same concentration metering caliber as the online concentration detection. The measured concentration, liquid source, and sampling time are recorded as the initial concentration state. This state is only used for pipe sections that have been confirmed to be filled with this initial fluid; pipe sections that have not yet been confirmed to be filled or may retain other liquids are still recorded as having an unknown state of the preceding residual liquid. The initial concentration state is only used to provide initial liquid information and does not replace the judgment of replacement completion and filling completion of the current path.

[0057] If a target pipe segment is closed but filled with liquid before switching, the last confirmed liquid source and concentration before closure will be used, and the current concentration in the storage area will not be directly replaced. If multiple consecutive target pipe segments are occupied by the same liquid source before switching, they can be associated with the same previous polishing fluid concentration, but each pipe segment will still retain its own effective liquid holding capacity so that it can be updated according to its actual liquid flow rate later. If a target pipe segment has been emptied before switching and the equipment status confirms that there is no residual liquid that needs to be replaced, the previous residual liquid volume will be recorded as 0, and the pipe segment will be marked as not needing replacement.

[0058] "No need for replacement" only indicates the absence of residual liquid from upstream pipelines, not that the pipeline segment has been fully filled. After a drained pipeline segment is reactivated, existing full-liquid detection or outlet drainage checks should be performed to confirm continuous liquid flow to the corresponding pipeline segment and its downstream path. The confirmation result and time should be recorded in the current path record. Simply having pump / valve feedback in place is not sufficient to confirm completion of filling. Paths that have not yet completed filling confirmation do not meet the conditions for subsequent material status. If the empty pipeline filling process may receive residual liquid from upstream pipelines, the "no need for replacement" status for that pipeline segment should be cancelled, and a replacement assessment should be performed again based on its effective liquid holdup and the completion status of upstream replacement.

[0059] When the source or concentration of liquid in a target pipe section before switching cannot be determined, the current concentration or other estimated value is not automatically assigned; instead, the previous residual liquid state is marked as unknown. This unknown state is only allowed to be removed after the actual liquid flow reaches the effective liquid holdup of the pipe section and meets the same replacement criterion. The target pipe section, effective liquid holdup, and previous residual liquid state are used together as the initial data for calculating the cumulative liquid flow after switching and updating the material replacement state.

[0060] S3 specifically includes the following sub-steps:

[0061] S310: Read the target pipe segment set, the effective liquid holding capacity of each target pipe segment, and the previous residual liquid state, and obtain the actual liquid flow rate through the corresponding target pipe segment after the flow path switching is completed. When the target pipe segment is equipped with a branch flow sensor, the actual flow rate detection value of that branch after validity verification is directly used.

[0062] When multiple target pipe sections are connected in series and there is no flow diversion, replenishment, or other liquid input between them, the effective flow detection result that can represent the same actual flow rate in the series flow path can be used. When the target pipe section is located in a parallel branch, only the data that can reflect the actual flow rate of that branch is used; when the flow distribution relationship of each parallel branch cannot be determined, the total liquid supply flow rate is not used to directly replace the actual flow rate of the branch.

[0063] Flow data is aligned with the time of flow path switching completion according to the timestamp. Only data whose flow direction is consistent with the permitted direction of the current path, and whose valve, pump, and related branch status have been confirmed as valid, is included in the accumulation. When the concentration control software receives data that is explicitly marked as communication failure, out of range, or invalid by the flow sensor, it will not write that data into the valid flow sequence. Instantaneous fluctuations within the normal range are retained and the actual accumulated liquid volume is not altered to obtain a smooth curve.

[0064] If the corresponding valve is switched back, the circulating pump stops, the flow direction changes, or the current path version becomes invalid, the flow accumulation under that path version is stopped, and the confirmed accumulation results are saved. If the target pipe segment does not have independent flow detection, but the flow rate can be uniquely determined by the effective flow data in the same unsplit-flow series path, then that data is allowed; otherwise, if a reliable flow source cannot be obtained, the material replacement status is recorded as unknown. This yields an effective flow sequence corresponding to the target pipe segment identifier, switching completion time, path version, and flow direction.

[0065] S320. Based on the current flow direction and the upstream and downstream relationship between the target pipe segments, determine the cumulative starting point for each target pipe segment used for displacement judgment. For target pipe segments that do not have upstream pipe segments still needing displacement and can directly receive the current mixture, accumulation begins after the flow path switching is completed and continuous effective flow data is obtained. For target pipe segments where upstream pipe segments are still in a state of incomplete displacement or unknown status, the liquid volume passing through them is not included in their displacement accumulation; only after all upstream pipe segments that actually supply liquid to this pipe segment have completed displacement, and the intermediate pipe segments that may receive residual liquid have also completed verification according to the same rules, is the accumulated liquid volume of this pipe segment set to 0 and accumulation begins. Parallel branches are judged separately, and the pipe segments after merging must wait for all upstream pipe segments that actually participate in the merging to meet the conditions. In this way, it is confirmed that the current liquid has reached the inlet of each target pipe segment, avoiding the inclusion of the previous residual liquid discharged from the upstream into the effective displacement liquid volume of the downstream. The industrial control software performs discrete accumulation according to the effective flow records obtained after each pipe segment starts accumulating and their actual time intervals:

[0066]

[0067] in, This represents the cumulative liquid volume that has passed through the j-th target pipe segment during the current continuous effective cumulative period after meeting its cumulative start-up conditions, in meters. 3 ; This represents the actual flow rate corresponding to the k-th valid record of the j-th target pipe segment, in meters. 3 / s; This represents the time interval between the previous valid flow record and the kth valid flow record, in seconds. For the initial accumulation, it represents the time interval between the actual start of accumulation for the target pipe segment and the first valid flow record, excluding the period waiting for upstream replacement. n represents the number of valid flow records participating in the current continuous valid accumulation. When the control cycle changes normally, the actual time interval is used, without assuming that each sampling cycle is absolutely equal. For duplicate records corresponding to the same timestamp, only one confirmed as valid by the control system is retained to avoid duplicate accumulation.

[0068] When communication failure, over-range, invalid marker, or missing data occurs between adjacent flow records, the current continuous valid accumulation is interrupted. The flow rate during the abnormal period is not calculated based on the two consecutive valid flow records, and the material replacement status of the corresponding target pipe segment is recorded as unknown. Downstream target pipe segments relying on this pipe segment for liquid supply simultaneously cease replacement accumulation, and their incomplete accumulation results are not used for release. Previously completed downstream statuses are also not used as current valid completion statuses while the upstream material status cannot be confirmed. After the flow data resumes continuous validity, the upstream replacement status is rechecked. If the upstream accumulation start conditions have been met, the recovery time is used as the new accumulation start point; if the upstream conditions have not yet been met, the process continues to wait until the conditions are met before starting a new accumulation. Each pipe segment that restarts accumulation will be used for replacement judgment. If set to 0, the cumulative liquid volume before the anomaly is only saved as historical data and is not merged with the cumulative liquid volume after recovery. The replacement completion state is only restored after the corresponding pipe section meets the cumulative start-up conditions again and the continuously confirmed cumulative liquid volume reaches its effective liquid holding capacity.

[0069] The cumulative liquid volume is compared with the effective liquid holdup of the target pipe section to obtain the normalized displacement progress value:

[0070]

[0071] in, This represents the replacement progress value of the j-th target pipe segment, which is a dimensionless quantity. This indicates the effective liquid holdup of the target pipe section as determined by the aforementioned replacement criteria, in meters (m). 3 The displacement progress value is only used to indicate the degree of completion of the cumulative liquid flow relative to the effective liquid holdup, and does not indicate the actual volume fraction of the preceding liquid inside the pipe section. Perform the above calculations as needed. For pipe sections where the initial residual liquid volume has been confirmed to be 0, no division operation is performed; they remain in a state where no replacement is required.

[0072] when At that time, the corresponding target pipe section remains in an incomplete replacement state; when When the displacement is complete, it is recorded as complete according to the criteria consistent with the effective liquid holdup calibration. Here, "displacement complete" indicates that the residual liquid's impact on the outlet condition has reached the allowable range used during equipment calibration, but does not mean absolute absence of residue in a microscopic sense. For pipe sections where the previous residual liquid condition was unknown, displacement is only considered complete when... Only then does it transition to the replacement completion state.

[0073] S330. Summarize the material replacement status of each target pipe segment actually included in the current sampling path and the current liquid supply path. If a path has only one target pipe segment, directly use the replacement status of that pipe segment. If it contains multiple series or parallel target pipe segments, check each target pipe segment that actually participates in the current flow. If all target pipe segments are in the replacement completed state or do not require replacement, the material replacement status of the corresponding path is recorded as completed; if any target pipe segment has not yet met the replacement completion condition, the path is recorded as incomplete; if there is a necessary target segment that cannot be determined, the corresponding path is recorded as unknown.

[0074] When the same target pipe segment belongs to both the current sampling path and the current liquid supply path, only one set of cumulative liquid volume, replacement progress value, and replacement status is saved. Both paths reference the same result to avoid two different states for the same physical pipe segment. Only when the target pipe segment set is empty, all previous replacement requirements have been met, there are no unverified pipe segments that may receive residual liquid from upstream, and the current path has completed the required liquid filling confirmation, is the corresponding path recorded as requiring no new replacement. Its material status will be treated as meeting the requirements for this replacement in subsequent judgments. The mere fact that the pipe segment composition, liquid source, and flow direction have not changed is insufficient to resolve the previous incomplete or unknown replacement status.

[0075] If valve switching, pump status changes, or actual flow path alterations occur again during the accumulation process, the target pipe segment is redefined after the new equipment status is confirmed. Unfinished or unknown replacement tasks still participating in the current flow, along with their affected downstream pipe segments, are associated with the new path version. The accumulation starting point for these pipe segments is redefined according to the new upstream-downstream relationship. The previously unfinished accumulated liquid volume is not directly used for determining the completion of the new path. Only continuously used pipe segments that have already completed replacement, maintain consistent current source and flow direction, are still in a valid filling state, and will not receive residual liquid from the previous switch are allowed to inherit their completed status. Before inheritance, these conditions are verified, and the verification results are recorded in the current path version.

[0076] The material replacement status of the current sampling path and the current liquid supply path are associated with and saved along with the path version, status time, target pipe sections involved, and corresponding cumulative liquid volume. This information is then used to determine whether the concentration detection value can represent the actual liquid being supplied, in conjunction with common source relationships and common mixing relationships.

[0077] S4 specifically includes the following sub-steps:

[0078] S410: Read the current sampling path, current liquid supply path, common source relationship, common mixing relationship, and corresponding path version to determine whether the current concentration feedback meets the corresponding conditions for the flow path source. Both paths involved in the judgment must first be in a valid state. If any path is invalid due to unconfirmed valve position, uncertain pump operating status, or inability to establish a connection relationship, the corresponding conditions for the flow path source are directly recorded as unmet, and the concentration detection results on that path are not used to generate a new proportional adjustment amount.

[0079] When both paths are valid, verify the established common source and common mixing relationships. If both are true and correspond to the same valid path version, the flow path source condition is considered satisfied. If the two paths originate from the same storage region but have separated before common mixing is complete, or if only one path connects to an independent liquid capable of changing the state of the liquid components after separation, then this condition is not satisfied.

[0080] When multiple parallel liquid supply branches exist, each branch actually participating in the current liquid supply should be checked; if any necessary branch cannot establish the required common source and common mixing relationship with the sampled liquid, it should not be replaced by other branches that meet the conditions.

[0081] If a new flow path switching command is received after a flow path record has been created, but all relevant equipment has not yet been confirmed, the current flow path status changes to "undetermined." The original flow path version is only retained as historical data and is no longer used to assign representative status to new concentration detection values. The judgment result of the flow path source corresponding to the conditions is associated with the current flow path version and the judgment time and saved for continued verification of the material status under the same flow path version.

[0082] S420. Under the premise that the corresponding conditions of the flow path source are met, read the material replacement status of the current sampling path and the current liquid supply path, and determine whether the material status correspondence conditions are met. The verification scope includes the target pipe section directly determined in this switch, the pipe section whose previous replacement was not completed or whose status is unknown but is still participating in the current flow, and the downstream pipe section that may receive the liquid discharged from the above-mentioned pipe sections. Only when the current path does not have the above-mentioned pipe section to be verified, the previous replacement requirements have been met, and the required liquid filling confirmation has been completed, is it treated as not requiring additional replacement. When there is a target pipe section, the effective accumulated liquid volume in the upstream and downstream order, the corresponding effective liquid holding volume, and the replacement status under the current path version are used as the basis; when the current path has not completed the required liquid filling confirmation, the material status correspondence conditions are recorded as not met.

[0083] The material status condition is considered satisfied when all necessary target pipe sections involved in the current sampling path and the current liquid supply path are in a state of complete replacement or do not require replacement. This condition is not satisfied if the sampling path has completed replacement but there are still target pipe sections in an incomplete state in the liquid supply path. Similarly, it is not satisfied if the liquid supply path has completed replacement but there is still residual liquid from before the switch in the sampling path. If the status of any necessary target pipe section is unknown, it is also treated as not satisfied, and the unknown pipe section is not inferred from the completion status of other pipe sections.

[0084] The material replacement status must be consistent with the current path version used to determine common source and common mixing relationships. Cumulative liquid volume and replacement status established under an earlier path version cannot be directly used for a new path where connections have changed. When the current path changes again, the corresponding status must be re-established according to the new target pipe segment and actual liquid volume. The flow path source correspondence condition and material status correspondence condition respectively specify whether the spatial sources of the sampled liquid and the supply liquid correspond, and whether the liquid status has corresponded after the flow path switch. Only when both conditions are met simultaneously is it permissible to continue confirming the representativeness of the supply liquid for the current concentration detection value.

[0085] S430: Read the concentration detection value and its acquisition time output by the concentration detection point, and perform time mapping between them and the current path version, flow path source corresponding conditions, and material status corresponding conditions. The concentration detection value itself should be in the normal output state of the sensor. Detection results that have been clearly marked by the sensor as communication failure, over-range, invalid self-test, or abnormal calibration status will not be used to determine the qualification of the liquid supply representative.

[0086] For the current path version, first determine the moment when the conditions corresponding to the flow path source are met. If a target pipe segment exists, then determine the moment when both the current sampling path and the current liquid supply path meet the material state conditions; if there are no new replacement requirements, this moment is the moment when the current path is confirmed.

[0087] Only concentration values ​​collected after both conditions have been met will be included in the current representative qualification confirmation. Concentration data obtained before material replacement is completed must not be retroactively granted representative qualification after subsequent replacements. If the path version corresponding to the concentration value is inconsistent with the current path version, even if the collection time is later, it will not be used as the current valid concentration value.

[0088] After the conditions corresponding to the flow path source, the material state, and the effective state of the concentration detection value are simultaneously met, it should also be confirmed that the current detection value reflects the stable liquid state after the most recent liquid supply adjustment. During equipment commissioning, change the supply ratio of concentrate to diluent under each permissible flow path and liquid supply condition. After both branches have completed execution, keep the actual liquid supply state unchanged, and take samples at the mixed liquid source location, concentration detection point, and polishing liquid supply point. Record the time required from the completion of execution until the concentration at each location reaches the new stable state, and the concentration difference between the sampling and supply locations is within the permissible measurement error range. The stable state is confirmed by the difference between the maximum and minimum concentration values ​​not exceeding the permissible measurement error of the detection method used within a period not less than the sensor response time. Take the longest time required verified under the corresponding permissible condition as the response waiting time of the flow path, and save it according to the applicable flow path, liquid supply flow range, and permissible adjustment range. This time covers the mixing, conveying, and sensor response processes, and not only the sensor's own response time.

[0089] During operation, after adjustments to the liquid supply ratio, total liquid supply volume, flow path switching, or replenishment affecting the concentration of a common source, the response waiting time is recalculated from the moment the relevant action is completed and the actual liquid supply status is confirmed to be stable. During the waiting period, the confirmed liquid supply status is maintained, and no new concentration ratio correction is made; if the aforementioned changes occur again during this period, the timing is reset. Only when the current operating condition is within the calibration range, the response waiting time has elapsed, and the concentration detection value after the waiting period meets the above stability conditions for a period not less than the sensor response time, is a new valid concentration detection value confirmed, taking into account the corresponding conditions of the flow path source, the corresponding conditions of the material state, and the effective status of the sensor itself. Here, only a stable concentration representative of the current liquid supply is required; it is not required that this concentration has reached the target polishing liquid concentration. If the concentration exceeds the calibration range or a stable detection result is not obtained, the new liquid supply is not assigned representative status.

[0090] The effective concentration detection value is saved in association with the acquisition time, current sampling path identifier, current liquid supply path identifier, and path version. If the flow path source correspondence condition is not met, it indicates that the required spatial correspondence between the sampled liquid and the current actual liquid supply has not yet been established; if the material state correspondence condition is not met, it indicates that although the path has been established, residual liquid or filling state may still affect the current sampling or liquid supply state; if the response waiting and stability conditions are not met, it indicates that the current detection result cannot be used to confirm the actual liquid supply concentration after the most recent adjustment.

[0091] In any of the above situations, the concentration value will not be used to determine the new liquid supply ratio adjustment. Even if multiple unrepresentative concentration values ​​are obtained consecutively, their representativeness will not be altered by averaging these data.

[0092] S5 specifically includes the following sub-steps:

[0093] S510: Read the valid concentration detection value of the liquid supply representative that has obtained qualification under the current path version, and read the target polishing liquid concentration corresponding to the current processing stage. The target polishing liquid concentration comes from the process parameters that have been loaded into the equipment, and is used according to the process parameters within the same processing stage. When the equipment switches to other polishing stations or stations using CMP polishing process, the target polishing liquid concentration corresponding to the current station is called, without changing the aforementioned flow path and material state judgment rules.

[0094] Before calculating the concentration deviation, double-check that the path version corresponding to the effective concentration detection value is still the current valid version. If a new flow path switch has occurred after the concentration value is obtained, that concentration value will no longer be used for new proportional adjustments.

[0095] When the path status remains unchanged, the current concentration deviation is calculated according to the following formula:

[0096]

[0097] Where 'e' represents the current concentration deviation, and the unit is consistent with the unit of measurement for the polishing fluid concentration; This indicates the target polishing slurry concentration for the current processing stage, derived from process parameters; This indicates a valid concentration test value indicating that the supplier has obtained the qualification to act as a liquid supply representative. When the current effective concentration is lower than the target concentration, the subsequent adjustment direction is to increase the proportion of concentrate in the total supply; when When adjusting, the direction is to decrease the proportion of concentrate; when At that time, the current ratio will not change due to concentration deviation.

[0098] In actual control, the maximum permissible measurement uncertainty given in the concentration sensor calibration record is used as the zero-deviation judgment range. When the absolute value of the concentration deviation does not exceed this range, the current concentration is considered to be within the target range that the sensor can reliably distinguish, and no new concentration ratio correction is generated. This range directly uses the current effective calibration result of the concentration sensor, without setting additional fixed thresholds independent of the device.

[0099] S520. Based on the current concentration deviation, the confirmed proportional adjustment coefficient, and the actual liquid supply status of the concentrate and diluent branches, determine the liquid supply ratio adjustment amount for this operation. The industrial control software reads the currently confirmed actual liquid supply volume of the two branches, and simultaneously reads the minimum stable liquid supply volume, the maximum allowable liquid supply volume, and the minimum adjustment resolution that the actuator can reliably achieve, all determined during the equipment commissioning phase.

[0100] The aforementioned boundaries are determined by progressively changing the pump speed, metering pump pulse frequency, or regulating valve opening, and recording the corresponding actual liquid supply results. The range within which the output can be repeated without stopping, significant instability, or exceeding the rated capacity is considered the executable range of this branch.

[0101] The proportional adjustment coefficient is determined before equipment commissioning using calibration data that has been confirmed to be representative of the liquid supply. The calibration sample must record at least the initial effective concentration, target concentration, actual liquid supply volume of both branches, the proportional adjustment amount after execution, and the effective concentration after re-stabilization. The proportional adjustment coefficient can be obtained offline using a particle swarm optimization (PSO) algorithm. The particle position represents a candidate proportional adjustment coefficient. For each calibration sample, based on the sample's concentration deviation, the concentrate ratio before adjustment, the total liquid supply volume flow rate benchmark corresponding to the sample, and the executable range of the two branches, a candidate proportional adjustment coefficient interval is determined that ensures the adjusted concentrate ratio remains within the corresponding executable ratio range. The candidate proportional adjustment coefficient intervals corresponding to each calibration sample are intersected with the range of proportional adjustment coefficients greater than 0, and the resulting intersection is used as the search range for the particle position. If the intersection is empty, the current PSO solution is not executed, and a backup proportional adjustment coefficient that has been verified through stable liquid supply calibration is used.

[0102] For each candidate coefficient, the corresponding candidate proportional adjustment amount is calculated for all calibration samples. If a candidate result causes the adjustment direction to be opposite to the concentration deviation direction, or causes the target liquid supply of any branch to exceed the feasible range, the particle is judged as infeasible and will not participate in the current optimal position update. The fitness of the remaining particles is calculated using the average of the absolute deviations between the candidate proportional adjustment amounts and the verified actual adjustment amounts in all calibration samples, with the smaller average deviation being considered the better result.

[0103] Particles are updated based on their historical best position and the current best position in the swarm, until the absolute value of the difference between the optimal proportional control coefficients obtained from two consecutive iterations is no greater than the coefficient resolution capability converted from the minimum adjustment resolution of the actuator, or the maximum number of iterations determined during the debugging phase is reached. If no candidate coefficients satisfying all execution constraints are found after reaching the maximum number of iterations, the current optimization result is not adopted, and a backup proportional control coefficient that has been verified through stable liquid supply calibration and stored in the industrial control software is used. The particle swarm algorithm is only used for offline determination of proportional control parameters and does not change the concentration feedback representative qualification judgment process.

[0104] The verified actual adjustment amount refers to the proportional adjustment amount recorded when, after the actuator completes its action according to the corresponding adjustment amount, the actual liquid supply state of both the concentrate branch and the diluent branch reaches the target state corresponding to that adjustment amount, and the re-acquired effective concentration enters the allowable measurement uncertainty range corresponding to the target polishing liquid concentration, and remains within that range for a period not less than the concentration sensor response time. Calibration records that do not meet the above conditions are not used as calibration samples for the particle swarm optimization algorithm. The re-stabilized effective concentration refers to the concentration detection value that, after the actuator completes its action and obtains representative liquid supply qualification, does not exceed the allowable measurement uncertainty range during the aforementioned holding period.

[0105] In each iteration of the particle swarm optimization algorithm, the current velocity of the particle is first multiplied by a pre-set inertia weight. Then, the individual correction amount (the difference between the particle's historical best position and current position multiplied by the individual learning coefficient and a first random number) and the population correction amount (the difference between the current population best position and current position multiplied by the population learning coefficient and a second random number) are added together to obtain the update velocity. The updated particle position is then obtained by adding the current position to the update velocity. The first and second random numbers are generated independently between 0 and 1. If the updated particle position exceeds the search range of the proportional adjustment coefficient, it is restricted to the corresponding boundary. The number of particles, inertia weight, individual learning coefficient, population learning coefficient, and maximum number of iterations are set during the equipment debugging phase and saved with the calibration results, remaining unchanged during the same calibration process.

[0106] The resolution coefficient, calculated from the minimum adjustment resolution of the actuator, is obtained by dividing the minimum proportional change that enables any supply branch to produce a verifiable change in supply by the absolute value of the maximum concentration deviation exceeding the allowable measurement uncertainty in the calibration sample. If there is no concentration deviation exceeding the allowable measurement uncertainty in the calibration sample, the particle swarm optimization is not performed, and the backup proportional adjustment coefficient, which has been verified through stable supply calibration, continues to be used.

[0107] During operation, the industrial control software reads the actual supply volume flow rate of the concentrate and the actual supply volume flow rate of the diluent, which have been confirmed by the execution feedback, and records them as follows: and Both use the same unit of volumetric flow rate. + When the concentration is greater than 0, the current concentration ratio r is obtained by dividing the actual supply volume flow rate of the concentrate by the sum of the actual supply volume flow rates of the two branches. ;when + When the ratio is 0, the current concentrate ratio is not calculated, and no new supply ratio adjustment is generated.

[0108] The proportional adjustment coefficient confirmed through offline calibration is denoted as K, and candidate changes in the concentrate ratio are determined based on the current concentration deviation. The adjusted ratio and the target supply volumetric flow rates for both branches are determined using the following formula:

[0109]

[0110] Where K represents the proportional adjustment coefficient, which is positive and is derived from the aforementioned offline calibration results; when the concentration is expressed in dimensional numerical form, the unit of K is the reciprocal of the corresponding concentration unit; when the concentration is expressed in dimensionless proportional form, K is a dimensionless quantity; e represents the current concentration deviation. The candidate change in the concentrate ratio is a dimensionless quantity; r represents the current concentrate ratio, which is also a dimensionless quantity. This indicates the proportion of the target concentrate, and is a dimensionless quantity. This means that the candidate ratio is limited to a range that can simultaneously meet the minimum stable liquid supply volume flow rate and the maximum allowable liquid supply volume flow rate of both the concentrate branch and the diluent branch under the current total liquid supply volume flow rate benchmark. This indicates the total liquid supply volume flow rate reference used in this control cycle, and its value is the current value. and The sum of these values, or the total liquid supply volume flow rate setpoint given separately by the current processing technology; and These represent the target supply volumetric flow rate of the concentrate and the target supply volumetric flow rate of the diluent, respectively, with units equal to... , and Consistent.

[0111] In the present If a ratio exists that simultaneously satisfies the executable range of both the concentrate branch and the diluent branch, the candidate ratios are restricted to that range, and the executable ratio closest to the candidate ratio is selected as the appropriate ratio. Based on this, the target liquid supply volume flow rate for the two branches is determined. In the current... If there is no ratio that can simultaneously satisfy the executable range of both branches, no new target ratio and target liquid supply volume flow rate are generated. The current total liquid supply volume flow rate benchmark is marked as unexecutable, and the current actual liquid supply state that can be confirmed by execution feedback is maintained. After the total liquid supply volume flow rate set value given by the processing technology is restored to the executable range, or the actual liquid supply state of the two branches is restored, the liquid supply ratio adjustment amount is recalculated based on the new effective concentration detection value.

[0112] S530: Convert the target supply volume of the concentrate and the target supply volume of the diluent into control commands that can be received by the existing actuators. When using a variable speed supply pump, calculate the target speed based on the correspondence between supply volume and pump speed established during the equipment commissioning phase; when using a metering pump, convert it to pulse frequency or number of pulses per unit time; when using a regulating valve, calculate the target opening degree according to the calibrated valve position and supply volume relationship. The specific command format is determined by the existing actuators of the equipment, and both branches use the target supply volume obtained in S520 as the control basis.

[0113] Before issuing control commands, recheck the current path version. If a new flow path switch has occurred after the effective concentration detection value is obtained, cancel the current proportional adjustment that has not yet been executed, and do not apply the concentration results under the old path conditions to the new path. When the path status remains valid, update the concentrate and diluent control commands within the same control cycle, and read the pump operation feedback, pulse execution feedback, or valve position feedback to confirm the actual execution status. Only when the status returned by the actuator corresponds to the command and is within the allowable range is the current proportional adjustment recorded as completed.

[0114] The completion of execution only indicates that the liquid supply actuator has reached the target state for this operation; it does not mean that the corresponding concentration change has reached the concentration detection point and the polishing liquid supply point. When this instruction causes a change in the actual liquid supply ratio or the total liquid supply volume, the previously obtained concentration detection value is suspended from being used to form a new ratio correction. The feedback is then reconfirmed according to the response waiting time and stability conditions of the current flow path. During the waiting period, the actual liquid supply state that can be confirmed by the execution feedback is maintained. The next ratio adjustment amount is only calculated after the waiting period ends and a new valid concentration detection value is obtained. The same concentration detection value is only used for one ratio adjustment calculation and is not accumulated again due to repeated operation of the control cycle. If a new flow path switch, liquid replenishment, or change in execution state occurs during the operation, the corresponding path, material state, and response waiting conditions are rechecked.

[0115] If a branch fails to reach the target state, the actual liquid supply status that can be confirmed is retained, and new concentration corrections are not accumulated according to the theoretical target value. The new proportional adjustment amount is calculated only after the execution status is restored and a representative effective concentration detection value for the current branch version is obtained again.

[0116] If the current concentration detection value fails to qualify as a representative for liquid supply due to unmet conditions related to the flow path source, material state, or sensor validity, no new concentration correction will be generated based on this detection value. If the most recently confirmed effective liquid supply ratio remains consistent with the current execution feedback, that ratio will continue to be maintained. If the actual execution status has changed, or it cannot be confirmed that the most recently confirmed effective ratio will still be maintained, the current actual liquid supply status that can be confirmed by execution feedback will remain, and no new ratio correction will be proactively added.

[0117] Once the current sampling path and the current liquid supply path meet the conditions of common source relationship, common mixing relationship and material state correspondence again, and a new effective concentration detection value is obtained, the target liquid supply volume of subsequent concentrate and diluent will be determined according to the same rules.

[0118] In the above formulas, dimensional parameters are converted to a unified unit of measurement before being used in the calculations, while dimensionless parameters are used directly according to their numerical values. Each parameter adopts the data source, determination method, and applicable conditions described in the aforementioned specific embodiments, and data that has not been verified by equipment status confirmation, sensor validity confirmation, or calibration is not used to replace the corresponding parameters.

[0119] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0121] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A metallographic polishing solution supply control method based on polishing slurry concentration feedback, characterized in that, Includes the following steps: S1. Obtain the current operating status of each switching valve, circulating pump and liquid supply branch, determine the current sampling path from the source of the mixed liquid to the concentration detection point and the current liquid supply path to the polishing liquid supply point, and determine the common source relationship and common mixing relationship of the two paths; S2. Compare the current sampling path and the current liquid supply path with the corresponding path before the switch to determine the target pipe segment that has been newly entered into the two paths due to the switch, and determine the state of the residual liquid in the previous sequence based on the effective liquid holding capacity of the target pipe segment and the state of the polishing liquid before the switch. S3. Obtain the actual liquid flow rate through the target pipe section after switching, determine the cumulative liquid volume, and update the previous residual liquid state according to the cumulative liquid volume and the effective liquid holding volume to obtain the material replacement state of the current sampling path and the current liquid supply path. S4. Based on the common source relationship, common mixing relationship and material replacement state, determine whether the sampling liquid and the supply liquid corresponding to the current concentration detection value have the same flow path source and corresponding material state, and determine the concentration detection value obtained after the common source relationship and common mixing relationship are established and the current sampling path and the current supply path meet the corresponding material replacement requirements as the effective concentration detection value.

2. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 1, characterized in that, Also includes: S5. Based on the deviation between the effective concentration detection value and the target polishing fluid concentration and the current total supply volume, determine the target supply volume of the concentrate and diluent and control the supply branch; for concentration detection values ​​that do not meet the requirements of common source relationship, common mixing relationship or corresponding material replacement, do not adjust the amount based on the formation of a new supply ratio.

3. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 1, characterized in that, S1 specifically includes: Read the actual operating status of each switching valve, circulating pump and supply branch, and call the flow path connection data that describes the connection relationship between the source location of the mixture, pipe section, switching valve, pump, concentration detection point and polishing supply point; for switching or start-stop status that has not yet obtained execution confirmation, record the current flow path status as undetermined; Starting from the source of the mixture, the concentration detection point and polishing supply point are located along the current allowed flow direction. The current sampling path and the current supply path are determined according to the actual flow order of the liquid. When there are parallel effective branches, each parallel branch that actually participates in the flow is retained in the corresponding path.

4. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 3, characterized in that, S1 also includes: comparing the upstream sources and common flow paths of the current sampling path and the current liquid supply path, determining the common source relationship, common mixing relationship and the last common node, and associating and saving the two paths, common source relationship, common mixing relationship, last common node, path version and path determination time; when the flow path is in a switching transition state, the current path is not replaced by the historical path before the switch.

5. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 1, characterized in that, S2 specifically includes: Compare the current sampling path and the current liquid supply path with the corresponding path confirmed last time before the flow path switch. The pipe segment that newly enters the corresponding path after the switch, or the pipe segment that is the same physical pipe segment but the liquid source or flow direction has changed, is identified as the target pipe segment. For each target pipe section, the effective liquid holding capacity is determined. The effective liquid holding capacity is the equivalent liquid volume parameter corresponding to the residual liquid replacement criterion. For regular pipe sections whose geometric volume is confirmed to correspond to the replacement criterion after replacement verification, their geometric volume is taken as the effective liquid holding capacity. For pipe sections with local volume or backmixing effects, the cumulative input liquid volume corresponding to the replacement criterion is taken as the effective liquid holding capacity. Based on the last actual liquid source connected to the target pipe section before the flow path switching, determine the concentration state of the preceding polishing fluid, the source of the preceding liquid, and the time when the state was formed, and associate it with the effective liquid holding capacity to form the state of the preceding residual liquid; if the state of the preceding liquid cannot be confirmed, the corresponding target pipe section is recorded as having an unknown residual state.

6. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 1, characterized in that, S3 specifically includes: After the flow path switching is completed, the effective flow data of each target pipe segment is obtained. Among them, the parallel target pipe segments use data that can reflect the actual flow of the corresponding branch. Target pipe segments connected in series and without flow splitting or liquid replenishment are allowed to share the same actual flow data. When a reliable flow source cannot be obtained, the corresponding material replacement status is recorded as unknown. The cumulative starting point is determined based on the upstream and downstream replacement status of each target pipe section. The cumulative liquid volume of each target pipe section is determined based on the effective flow data and its time interval, and the cumulative liquid volume is compared with the corresponding effective liquid holding capacity. When the cumulative liquid volume reaches or exceeds the effective liquid holding capacity, the target pipe section is recorded as having completed replacement; otherwise, the replacement is kept in an incomplete state.

7. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 6, characterized in that, S3 also includes: summarizing the status of target pipe segments in the current sampling path and the current liquid supply path respectively, and only when all target pipe segments in the corresponding path are in the state of replacement completed or no replacement is required, the material replacement status of the path is recorded as completed; when the status of any target pipe segment actually participating in the current flow is unknown, the material replacement status of the corresponding path is recorded as unknown.

8. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 1, characterized in that, S4 specifically includes: Under the condition that both the current sampling path and the current liquid supply path are valid, check the common source relationship and common mixing relationship of the two paths, and when both are true and correspond to the same current path version, determine that the current sampled liquid and the liquid supply liquid meet the flow path source correspondence condition; After the conditions corresponding to the flow path source are met, check the material replacement status of the current sampling path and the current liquid supply path. Only when the target pipe segments that actually participate in the current flow in the current sampling path and the current liquid supply path are in a state of complete replacement or no replacement is required, it is determined that the material status corresponding conditions are met; if any target pipe segment is in a state of incomplete replacement or unknown, it is determined that the material status corresponding conditions are not met. The concentration detection value and its acquisition time are determined only after the conditions corresponding to the flow path source and the material state have been met, and the concentration detection value is consistent with the current path version and is in the normal output state of the sensor. The concentration detection value obtained before the two conditions are met is not retrospectively determined as a valid concentration detection value.

9. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 2, characterized in that, S5 specifically includes: Read the effective concentration detection value corresponding to the current path version and the target polishing fluid concentration of the current processing stage, and determine the concentration deviation between the two; if the concentration deviation does not exceed the allowable measurement uncertainty corresponding to the concentration sensor calibration result, no new liquid supply ratio correction will be generated; Based on the concentration deviation exceeding the allowable measurement uncertainty and the proportional adjustment coefficient determined by the liquid supply calibration, the change in the proportion of concentrate is obtained. When the current processing technology gives a set value for the total liquid supply, this set value is used as the benchmark for the total liquid supply. Otherwise, the sum of the currently confirmed actual liquid supply of concentrate and the actual liquid supply of diluent is used as the benchmark for the total liquid supply. Based on this, the target liquid supply of concentrate and the target liquid supply of diluent are determined, and both are limited to the executable range of the corresponding liquid supply branch.

10. The metallographic polishing solution supply control method based on polishing solution concentration feedback according to claim 9, characterized in that, S5 also includes: converting the target liquid supply volume into control commands for the corresponding liquid supply actuator and verifying the actual execution status; when a new flow path switch occurs or the current concentration detection value is not determined to be an effective concentration detection value, no new concentration ratio correction is formed, and the current actual liquid supply status confirmed by the execution feedback is maintained.

Images