Electrochemical forming process parameter optimization control method for aeronautical box wall shell structure

Abstract

The application discloses an electrochemical forming process parameter optimization control method for an aviation box wall shell structure and belongs to the technical field of electrochemical machining control. Through collecting time sequence data of environmental disturbance, power supply and electrolyte parameters, characteristic nodes are screened, a machining state stability index and a state deviation amplitude model are constructed, and a machining state evolution curve is generated. After function fitting optimization coordination time delay scale, characteristic nodes are screened to generate a control sequence, a double-mode abnormality matrix and a coordinated abnormality feedback matrix are constructed, and abnormality fitting coefficients are quantified. Through time sequence proliferation dynamic updating coefficients, instruction parameter adjustment is realized. The application realizes coordinated optimization control of the power supply and the electrolyte parameters, accurately captures the machining state fluctuation law under environmental disturbance, solves the problems of lack of dynamic coordination and abnormal response lag in traditional control methods, greatly improves the machining precision, stability and parameter adjustment timeliness of the aviation box wall shell electrochemical forming, and is suitable for high-precision aviation part machining scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrochemical machining control, in particular to an electrochemical forming process parameter optimization control method for an aero-case wall shell structure. BACKGROUND

[0002] The aero-case wall shell structure is a key part of core equipment such as aero-engines and spacecraft. Its structure is complex, and the dimensional accuracy requirement is strict (usually needs to meet the wall thickness tolerance within ±0.02 mm and the surface roughness Ra≤0.8 μm), which directly affects the power performance and operation safety of the equipment. Electrochemical forming technology has become one of the core technologies for aero-case wall shell machining, with the advantages of no cutting force, excellent surface quality, and strong material adaptability. However, the electrochemical forming process of the aero-case wall shell is significantly affected by environmental disturbances (such as temperature fluctuations, humidity changes, and power grid voltage fluctuations), and there is a strong coupling relationship between the power parameters (voltage and current) and the electrolyte parameters (temperature, concentration, and flow rate), which leads to the processing state deviating from the steady state, seriously affecting the processing accuracy and consistency.

[0003] At the same time, the traditional control method mostly uses single parameter threshold judgment, only monitors the single point parameter of the power supply or the electrolyte independently, ignores the synergistic effect of the two and the time sequence cumulative influence of environmental disturbances, leads to one-sided evaluation of stability, and cannot accurately reflect the comprehensive fluctuation of the processing state; and, the existing technology lacks quantitative analysis of the processing state evolution law, and an effective state deviation characterization model has not been established, making it difficult to accurately capture the correlation between parameter fluctuation and processing quality under environmental disturbance, and the parameter adjustment is blind; and, the existing abnormal judgment and feedback mechanism is static, without considering the time sequence synergistic effect of parameter changes, the control sequence lacks pertinence, leading to abnormal response lag, and unable to realize dynamic adaptive adjustment; and, the existing model is mostly of fixed structure, cannot be dynamically updated with the processing process, and is difficult to adapt to the cumulative drift of parameters in a long time processing process, and the processing stability continues to decline.

[0004] For example, some existing technologies adjust parameters by presetting fixed thresholds of voltage and temperature, but do not consider the comprehensive influence of voltage variance and temperature variance on processing stability, when environmental disturbances cause small fluctuations in parameters but do not exceed the threshold, it will still cause implicit deviation of the processing state, ultimately leading to uneven wall thickness of the aero-case wall shell; another technology only judges based on parameter data of a single time sequence node, without digging the synergistic law between feature nodes, the parameter adjustment time and amplitude are unreasonable, leading to repeated fluctuations in processing accuracy. These problems make it difficult for the existing technology to meet the processing needs of high precision and high stability of the aero-case wall shell, and restrict the further application of electrochemical forming technology in the field of aviation high-end manufacturing. Therefore, developing a control method that can accurately quantify the processing state fluctuation and realize the synergistic dynamic optimization of power supply and electrolyte parameters has become a technical problem to be solved. SUMMARY

[0005] The present application aims to provide an electrochemical forming process parameter optimization control method for an aero case wall shell structure to solve the problems raised in the background art.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] The electrochemical forming process parameter optimization control method for an aero case wall shell structure comprises the following steps:

[0008] Step S1: Collecting environmental disturbance time series data, power supply parameter time series data and electrolyte parameter time series data in the electrochemical forming process of the aero case wall shell, and performing feature node screening and marking, the feature node being in a corresponding relationship with the time series node; taking the steady-state working condition of the electrochemical forming process under no environmental disturbance as a reference benchmark, recording the state deviation amplitude relative to the reference benchmark under environmental disturbance, to form a processing state evolution curve;

[0009] Constructing a processing state stability index model, the feature sample being composed of the power supply parameter set and the electrolyte parameter set at the feature node, and the power supply stability index and the electrolyte stability index under environmental disturbance or no environmental disturbance being quantitatively obtained through the mean and square values of voltage and temperature;

[0010] Constructing a state deviation amplitude model, and evaluating the state deviation amplitude through the power supply stability index and the electrolyte stability index;

[0011] Step S2: Function fitting is performed on the processing state evolution curve, the independent variable of the fitting function being the feature node, and the dependent variable being the state deviation amplitude; based on the fitting function, the processing field coordination degree is evaluated to optimize the coordination time delay scale as the cycle period;

[0012] Step S3: Based on the cycle period, the feature nodes are screened to generate a control sequence; based on the control sequence, a bimodal matrix model is constructed, including a first modal matrix being a power supply polarization state abnormal matrix for recording the power supply polarization state, and a second modal matrix being an electrolyte chemical state abnormal matrix for recording the electrolyte chemical state;

[0013] Step S4: Based on the first modal matrix and the second modal matrix, a coordination abnormal feedback matrix model of the power supply polarization state and the electrolyte chemical state is built to quantitatively obtain the power supply abnormal fitting coefficient and the electrolyte abnormal fitting coefficient.

[0014] Step S5: The numerical values of the power supply polarization state and the electrolyte chemical state at the current time series node are collected, the bimodal matrix model is time series augmented, and is substituted into the coordination abnormal feedback matrix model to generate new power supply abnormal fitting coefficients and electrolyte abnormal fitting coefficients to instruct the adjustment of the power supply parameters and the electrolyte parameters.

[0015] Preferably, the processing state evolution curve comprises:

[0016] The power supply parameter time series data and the electrolyte parameter time series data are aligned at time nodes based on the environmental disturbance time series data in the aviation case shell electrochemical forming process, to mark a characteristic node triggered by the power supply parameter and the electrolyte parameter under the environmental disturbance, which makes the aviation case shell electrochemical forming process reach the processing state stability, and the characteristic node is a time node at which the aviation case shell electrochemical forming process reaches the processing state stability.

[0017] The state deviation amplitude of the aviation case shell electrochemical forming process relative to the reference datum is recorded, and the state deviation amplitude is fitted into a processing state two-dimensional coordinate system, with the horizontal coordinate of the processing state two-dimensional coordinate system corresponding to the characteristic node and the vertical coordinate of the processing state two-dimensional coordinate system corresponding to the state deviation amplitude, to obtain a processing state evolution curve, with the steady-state working state of the electrochemical forming process under no environmental disturbance as the reference datum.

[0018] Preferably, the processing state stability index model is constructed in step S1 as follows:

[0019] The power supply parameter set and the electrolyte parameter set at each characteristic node constitute a group of characteristic samples, and the power supply parameter and the electrolyte parameter generated at the characteristic node are recorded in the power supply parameter set and the electrolyte parameter set, respectively, and the group number of the characteristic sample corresponds to the serial number of the characteristic node.

[0020] According to the time sequence arrangement of the characteristic nodes, sequentially arranged groups of characteristic samples are constructed, and the groups of characteristic samples are parameter-corrected.

[0021] The power supply parameter correction value of the i-th group of characteristic samples is The electrolyte parameter correction value of the i-th group of characteristic samples is , wherein and are the voltage mean value and the temperature mean value of the i-th group of characteristic samples, respectively, and are the voltage variance and the temperature variance of the i-th group of characteristic samples, respectively.

[0022] The power supply stability index is quantitatively obtained based on the power supply parameter correction value The electrolyte stability index is quantitatively obtained based on the electrolyte parameter correction value , wherein max{} and min{} are maximum and minimum value functions, respectively, for selecting the maximum value and the minimum value in the power supply parameter correction value and the electrolyte parameter correction value, respectively.

[0023] Preferably, the construction state deviation amplitude model in step S1 is as follows:

[0024] The power supply parameters and electrolyte parameters under environmental disturbance are substituted into the processing state stability index model to obtain the processing state index characteristic point under environmental disturbance , wherein and are the power supply stability index and the electrolyte stability index obtained under environmental disturbance, respectively;

[0025] The power supply parameters and electrolyte parameters under no environmental disturbance are substituted into the processing state stability index model to obtain the processing state index characteristic point under no environmental disturbance , wherein and are the power supply stability index and the electrolyte stability index obtained under no environmental disturbance, respectively;

[0026] The state deviation amplitude is evaluated ;

[0027] It should be noted that no environmental disturbance is an ideal state, and the processing state of electrochemical forming is determined by the power supply parameters (core is voltage) and the electrolyte parameters (core is temperature). The stability of the voltage on the power supply side directly affects the current density uniformity of the electrolytic reaction, and the temperature stability of the electrolyte affects the ion migration efficiency and reaction rate. Both of them together constitute the "double core influence factor" of the processing state. The core calculation formula of the state deviation amplitude is the two-dimensional Euclidean distance, which can reflect the degree of deviation of the "double core factor" from the ideal state, that is, the "comprehensive fluctuation degree" quantitative index of the processing state relative to the ideal steady state. The greater the state deviation amplitude, the worse the stability of the power supply and / or the electrolyte caused by environmental disturbance, the weaker the uniformity and controllability of the electrolytic reaction, and the processing precision (such as the wall thickness and surface quality of the cartridge wall shell) is difficult to guarantee. When the state deviation amplitude is 0, the processing state completely fits the ideal steady state, which is the optimal working state of electrochemical forming.

[0028] Preferably, the optimization of the coordination time delay scale in step S2 includes:

[0029] The processing state evolution curve fitting function is , x is the characteristic node number, and ;

[0030] The coordination time delay scale k between the power supply stability index and the electrolyte stability index is initialized, and the processing field coordination degree under the coordination time delay scale k is , I is the total number of characteristic nodes;

[0031] The value of the coordination time delay scale k is adjusted, k=k+1, to obtain the processing field coordination degree under the coordination time delay scale k+1 , selecting the synchronization delay scale that maximizes the degree of coordination of the processing field ;

[0032] It should be noted that the fitting function of the processing state evolution curve (x is the feature node number, the dependent variable is the state deviation amplitude) is essentially a time sequence variation model of the state deviation amplitude. This function smooths the fluctuations of discrete feature nodes and can present the trend of the deviation amplitude changing with time (feature node time sequence) (such as periodic fluctuations, monotonic changes, etc.). In electrochemical forming, there is a time sequence coordination effect between the changes of power supply parameters and electrolyte parameters (such as after voltage adjustment, the electrolyte temperature needs to go through a certain time sequence node to reach a new stable state). The fitting function can reflect this periodicity. With the optimal synchronization delay scale k as the cycle period, the feature nodes are screened to generate a control sequence, ensuring that the subsequent bimodal matrix model and synchronization anomaly feedback matrix model can "focus on key time sequence nodes", so that the adjustment of power supply parameters and electrolyte parameters has "time sequence synchronicity". For example, within a synchronization delay cycle, only the key feature nodes in the cycle need to be monitored to accurately capture the synchronization anomaly of the two, thereby improving control efficiency and accuracy.

[0033] Preferably, constructing the bimodal matrix model in step S3 comprises:

[0034] Screening the feature nodes with the synchronization delay scale as the cycle period to obtain a control sequence of the feature nodes;

[0035] Triggering the numerical collection instructions of the power supply polarization state and the electrolyte chemical state at each feature node in the control sequence, and constructing the bimodal matrix model, wherein the first modal matrix is the power supply polarization state anomaly matrix, the second modal matrix is the electrolyte chemical state anomaly matrix, and the column index of the bimodal matrix is the serial number of the feature node, the row index of the first modal matrix is the serial number of the power supply polarization state type, and the row index of the second modal matrix is the serial number of the electrolyte chemical state type;

[0036] In the first modal matrix, if at the i-th feature node , the power supply polarization state value corresponding to the e-th power supply polarization state type is detected to be abnormal, the matrix position of the e-th row and the i-th column in the first modal matrix is set to 1; if at the i-th feature node , the power supply polarization state value corresponding to the e-th power supply polarization state type is detected to be normal, the matrix position of the e-th row and the i-th column in the first modal matrix is set to 0;

[0037] In the second modal matrix, if at the i-th feature node If the electrolyte chemical state value corresponding to the rth electrolyte chemical state type is detected to be abnormal at the ith feature node, the matrix position of the rth row and the ith column in the second modal matrix is set to 1. If the electrolyte chemical state value corresponding to the rth electrolyte chemical state type is detected to be normal at the ith feature node, the matrix position of the rth row and the ith column in the second modal matrix is set to 0.

[0038] Preferably, the step S4 of building the cooperative abnormal feedback matrix model comprises:

[0039] Based on the first modal matrix and the second modal matrix, a cooperative abnormal feedback matrix model of the power source polarization state and the electrolyte chemical state is built , wherein, the first modal matrix is represented by, the second modal matrix is represented by, T is a matrix transposition symbol, E represents the total number of power source polarization state types, R represents the total number of electrolyte chemical state types, and S represents the total number of feature nodes contained in the control sequence:

[0040] The sum of each matrix element value in the cooperative abnormal feedback matrix model is obtained to obtain a cooperative abnormal feedback value; the sum of the matrix element values in the e th row of the cooperative abnormal feedback matrix model is obtained to obtain the power source polarization state abnormal feedback value of the e th power source polarization state type varying with the feature node, and the ratio of the power source polarization state abnormal feedback value to the cooperative abnormal feedback value is taken as a power source abnormal fitting coefficient, which is used to represent the power source abnormal probability; the sum of the matrix element values in the r th column of the cooperative abnormal feedback matrix model is obtained to obtain the electrolyte chemical state abnormal feedback value of the r th electrolyte chemical state type varying with the feature node, and the ratio of the electrolyte chemical state abnormal feedback value to the cooperative abnormal feedback value is taken as an electrolyte abnormal fitting coefficient, which is used to represent the electrolyte abnormal probability.

[0041] It should be noted that each feature node in the control sequence records the polarization state of the power supply (abnormal / normal) and the chemical state of the electrolyte (abnormal / normal), and the abnormal / normal state of each node in the control sequence is recorded in 0 / 1 coding; the cooperative abnormal feedback matrix is essentially a quantitative model of the "abnormal coupling relationship" between the polarization state of the power supply and the chemical state of the electrolyte, which is used to represent the "cooperative occurrence frequency" of the e-th type of power supply abnormality and the r-th type of electrolyte abnormality in the S nodes of the control sequence; by summing the elements of O(E, R), the "total number of cooperative abnormalities" (cooperative abnormal feedback value) is obtained, and then the abnormal fitting coefficient is calculated by the "cooperative occurrence frequency / total frequency" of a certain type of abnormality, which is essentially a frequency estimation probability. In the cooperative abnormal feedback matrix O(E, R), the larger the value of the matrix element O(e, r), the higher the "coincidence probability" of the e-th type of power supply abnormality and the r-th type of electrolyte abnormality, indicating that there is a strong coupling relationship between the two (such as a certain type of voltage abnormality causing electrolyte temperature abnormality), and the power supply abnormality fitting coefficient (the sum of the elements in the e-th row and the total element sum) and the electrolyte abnormality fitting coefficient (the sum of the elements in the r-th column and the total element sum) respectively quantify the "overall occurrence probability" of a certain type of power supply abnormality and the "overall occurrence probability" of a certain type of electrolyte abnormality.

[0042] Preferably, in step S5, the instructions for adjusting the power supply parameters and the electrolyte parameters include:

[0043] Collect the values of the polarization state of the power supply and the chemical state of the electrolyte under the current timing node, return to step SS, and perform column indexing on the bimodal matrix model. Determine whether the polarization state value or the chemical state value is normal or abnormal, and record the determination result in the added column;

[0044] After the augmentation is completed, step S4 is executed to evaluate the new power supply abnormality fitting coefficient and the electrolyte abnormality fitting coefficient through the cooperative abnormal feedback matrix model;

[0045] If the new power supply abnormality fitting coefficient corresponding to the e-th power supply polarization state type is greater than the power supply abnormality fitting coefficient before augmentation, the e-th power supply polarization state type is traced and locked, and a power supply parameter adjustment instruction is sent to the staff port. If the new power supply abnormality fitting coefficient is less than or equal to the power supply abnormality fitting coefficient before augmentation, no power supply parameter adjustment instruction is generated;

[0046] If the new electrolyte abnormality fitting coefficient corresponding to the r-th electrolyte chemical state type is greater than the electrolyte abnormality fitting coefficient before augmentation, the r-th electrolyte chemical state type is traced and locked, and an electrolyte parameter adjustment instruction is sent to the staff port. If the new electrolyte abnormality fitting coefficient is less than or equal to the electrolyte abnormality fitting coefficient before augmentation, no electrolyte parameter adjustment instruction is generated;

[0047] It should be noted that the "time sequence proliferation" of the dual-mode matrix is to add the polarization state of the power supply and the chemical state data of the electrolyte at the current time sequence node based on the original control sequence, which is equivalent to "updating the historical sample library". The new abnormal fitting coefficient is the probability value calculated based on the "original sample + current sample". The difference between the new coefficient and the coefficient before proliferation reflects the "influence of the current state on the probability of abnormality". If the new coefficient > the coefficient before proliferation, it means that the state of the current time sequence node leads to an increase in the probability of this type of abnormality, and the processing state tends to be unstable, so the parameters need to be adjusted to suppress the abnormality. If the new coefficient ≤ the coefficient before proliferation: it means that the current state does not aggravate the abnormality, and the processing state remains stable, so there is no need to adjust.

[0048] The electrochemical forming process parameter optimization control system for the aerospace box wall and shell structure comprises:

[0049] a data acquisition module, a data processing and fitting optimization module, a matrix model construction module, and a parameter adjustment instruction generation module;

[0050] The data acquisition module is used to acquire various time sequence data in the electrochemical forming process of the aerospace box wall and shell and to screen and mark characteristic nodes.

[0051] The data processing and fitting optimization module is used to construct a processing state stability index model and a state deviation amplitude model, to perform function fitting and optimization on the processing state evolution curve, and to determine the optimal cooperative time delay scale.

[0052] The matrix model construction module is used to generate a control sequence, to construct a dual-mode matrix model, and to construct a cooperative abnormality feedback matrix model.

[0053] The parameter adjustment instruction generation module is used to acquire relevant state values at the current time sequence node, to perform time sequence proliferation on the dual-mode matrix model, to calculate a new abnormal fitting coefficient, and to generate a corresponding parameter adjustment instruction.

[0054] Preferably, the data acquisition module comprises a time sequence data acquisition unit and a characteristic node marking unit. The time sequence data acquisition unit is used to acquire time sequence data of environmental disturbances, power supply parameters, and electrolyte parameters. The characteristic node marking unit is used to align time sequence nodes and mark characteristic nodes that make the processing state stable.

[0055] The data processing and fitting optimization module comprises a stability index calculation unit, a state deviation amplitude evaluation unit, and a cooperative time delay scale optimization unit. The stability index calculation unit is used to quantify power supply stability indexes and electrolyte stability indexes. The state deviation amplitude evaluation unit is used to evaluate the state deviation amplitude relative to a reference benchmark under environmental disturbances. The cooperative time delay scale optimization unit is used to determine the optimal cooperative time delay scale based on the processing field cooperation degree.

[0056] The matrix model construction module comprises a dual-mode matrix generation unit and a collaborative abnormal feedback matrix building unit, the dual-mode matrix generation unit is used for recording abnormal conditions of power supply polarization state and electrolyte chemical state, and the collaborative abnormal feedback matrix building unit is used for quantifying power supply abnormal fitting coefficients and electrolyte abnormal fitting coefficients.

[0057] The parameter adjustment instruction generation module comprises a matrix time sequence generation unit, an abnormal fitting coefficient updating unit and an adjustment instruction sending unit, the matrix time sequence generation unit is used for expanding column indexes of the dual-mode matrix and recording current state judgment results, the abnormal fitting coefficient updating unit is used for recalculating abnormal fitting coefficients, and the adjustment instruction sending unit is used for sending adjustment instructions or not sending instructions to a staff port according to coefficient changes.

[0058] Compared with the prior art, the beneficial effects achieved by the present application are:

[0059] The present application constructs a processing state stability index model, and realizes quantitative evaluation of power supply and electrolyte stability through mean and variance of voltage and temperature, breaks through the limitation of traditional single parameter threshold judgment, can comprehensively capture the comprehensive fluctuation of the processing state, and solves the problem of one-sidedness of the stability evaluation in the prior art.

[0060] The present application accurately reveals the time sequence change law of the processing state under environmental disturbance by means of the state deviation amplitude model (two-dimensional Euclidean distance quantization) and the processing state evolution curve fitting, realizes targeted selection of the control sequence by combining with collaborative time delay scale optimization, makes the parameter adjustment have time sequence synchronism, and overcomes the defects of blindness and lack of collaboration of the parameter adjustment in the prior art.

[0061] The combination of the dual-mode matrix model and the collaborative abnormal feedback matrix can quantize the abnormal coupling relationship between the power supply polarization state and the electrolyte chemical state, record the abnormal state through 0 / 1 coding, realize fine identification of the abnormal type, and greatly improve the accuracy and reliability of abnormal positioning compared with the traditional abnormal judgment method.

[0062] The dual-mode matrix time sequence generation mechanism realizes dynamic updating of the model, recalculates the abnormal fitting coefficients based on the "historical sample + current sample", makes the parameter adjustment capable of responding to the processing state change in real time, solves the problem that the existing fixed model cannot adapt to parameter cumulative drift, realizes adaptive control, and further helps to improve the processing precision and stability of the aviation box wall and shell. BRIEF DESCRIPTION OF DRAWINGS

[0063] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, for explaining the present application, and do not constitute a limitation on the present application.

[0064] Figure 1 is a schematic diagram of the steps of the method for optimizing and controlling the electrochemical forming process parameters of the aviation case wall shell structure according to the present application. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0066] Please refer to Figure 1 In the first embodiment, the method for optimizing and controlling the electrochemical forming process parameters of the aviation case wall shell structure is provided. The first embodiment is directed to the electrochemical forming processing of a TC4 titanium alloy case wall shell of a certain type of aviation engine. The inner diameter of the case wall shell is φ300mm, the wall thickness is 5mm, the wall thickness tolerance is ±0.02mm, the surface roughness is Ra≤0.8μm, the processing equipment is a numerical control electrochemical forming machine tool, the electrolyte is a NaCl solution with a mass fraction of 15%, the power supply is a pulse direct current power supply, and the processing environment has ±5℃ temperature fluctuation and ±0.5V power grid voltage fluctuation.

[0067] The method comprises the following steps:

[0068] Step S1: Collecting the environmental disturbance time series data, power supply parameter time series data and electrolyte parameter time series data in the electrochemical forming process of the aviation case wall shell, and performing feature node screening and marking, the feature node and the time series node being in a corresponding relationship; taking the steady state of the electrochemical forming process under no environmental disturbance as a reference benchmark, recording the state deviation amplitude relative to the reference benchmark under environmental disturbance, so as to form a processing state evolution curve;

[0069] Constructing a processing state stability index model, the feature sample is composed of the power supply parameter set and the electrolyte parameter set at the feature node, and the power supply stability index and the electrolyte stability index under environmental disturbance or no environmental disturbance are quantitatively obtained through the mean and square values of voltage and temperature;

[0070] Constructing a state deviation amplitude model, the state deviation amplitude is evaluated through the power supply stability index and the electrolyte stability index;

[0071] Exemplarily, the processing state evolution curve comprises:

[0072] The power supply parameter time series data and the electrolyte parameter time series data are aligned in time nodes based on the environmental disturbance time series data in the aviation case wall shell electrochemical forming process, so as to mark the characteristic node of the aviation case wall shell electrochemical forming process reaching the processing state stability triggered by the power supply parameter and the electrolyte parameter under the environmental disturbance, and the characteristic node is the time node of the aviation case wall shell electrochemical forming process reaching the processing state stability.

[0073] The state deviation amplitude of the aviation case wall shell electrochemical forming process relative to the reference datum is recorded, and the state deviation amplitude is fitted into a processing state two-dimensional coordinate system, wherein the abscissa of the processing state two-dimensional coordinate system corresponds to the characteristic node, and the ordinate of the processing state two-dimensional coordinate system corresponds to the state deviation amplitude, so as to obtain a processing state evolution curve.

[0074] The processing state stability index model is constructed as follows:

[0075] The power supply parameter set and the electrolyte parameter set at each characteristic node constitute a group of characteristic samples, and the power supply parameter and the electrolyte parameter generated at the characteristic node are recorded in the power supply parameter set and the electrolyte parameter set respectively, and the grouping number of the characteristic sample corresponds to the serial number of the characteristic node.

[0076] According to the time sequence arrangement of the characteristic nodes, a plurality of groups of characteristic samples in sequence are constructed, and the groups of characteristic samples are parameter corrected:

[0077] The power supply parameter correction value of the i-th group of characteristic samples The electrolyte parameter correction value of the i-th group of characteristic samples , wherein and are the voltage mean value and the temperature mean value of the i-th group of characteristic samples, and are the voltage variance and the temperature variance of the i-th group of characteristic samples.

[0078] Based on the power supply parameter correction value, the power supply stability index is quantitatively obtained, and based on the electrolyte parameter correction value, the electrolyte stability index is quantitatively obtained, wherein max{} and min{} are maximum and minimum value functions, respectively, for selecting the maximum value and the minimum value in the power supply parameter correction value and the electrolyte parameter correction value.

[0079] The state deviation amplitude model is constructed as follows:

[0080] The power supply parameter and the electrolyte parameter under the environmental disturbance are substituted into the processing state stability index model to obtain the processing state index characteristic point under the environmental disturbance wherein, and are power supply stability index and electrolyte stability index obtained under environmental disturbance respectively;

[0081] Substitute power supply parameters and electrolyte parameters under no environmental disturbance into the processing state stability index model to obtain processing state index characteristic points under no environmental disturbance wherein, and are power supply stability index and electrolyte stability index obtained under no environmental disturbance respectively;

[0082] Evaluate the state deviation amplitude ;

[0083] For example, collect time series data through temperature sensors, voltage sensors, and electrolyte concentration sensors, with a collection frequency of 10 Hz and a collection time of 120 min, a total of 72,000 groups of time series data; align the power supply and electrolyte parameter time series nodes based on the environmental disturbance time series data, mark the characteristic nodes of the processing state stability compliance (voltage fluctuation ≤0.1 V, temperature fluctuation ≤0.3 ℃) and the continuous time length ≥3 s, a total of 120 characteristic nodes (serial numbers 1-120);

[0084] Take the 5th group of characteristic samples as an example, the voltage mean U=12 V, the voltage variance σ(U)=0.08 V, the power supply parameter correction value U'=12×(12 / (12+0.08))≈11.92 V; the electrolyte temperature mean W=30 ℃, the temperature variance σ(W)=0.2 ℃, the electrolyte parameter correction value W'=30×(30 / (30+0.2))≈29.80 ℃; the power supply stability index U(5)=0.82 and the electrolyte stability index W(5)=0.85 are calculated;

[0085] Under no environmental disturbance, , , the state deviation amplitude , fitted to a two-dimensional coordinate system to obtain a processing state evolution curve.

[0086] Step S2: Function fitting is performed on the processing state evolution curve, the independent variable of the fitting function is the characteristic node, and the dependent variable is the state deviation amplitude. Based on the fitting function, the processing field coordination degree is evaluated to optimize the coordination time delay scale as the cycle period;

[0087] Exemplarily, the optimization of the coordination time delay scale includes:

[0088] The processing state evolution curve fitting function is , x is the serial number of the characteristic node, and ;

[0089] Initialize the coordination time delay scale k between the power supply smoothness index and the electrolyte smoothness index, and the process field coordination degree under the coordination time delay scale k , I is the total number of feature nodes;

[0090] Adjust the value of the coordination time delay scale k, let k=k+1, and obtain the process field coordination degree under the coordination time delay scale k+1 , select the coordination time delay scale that makes the process field coordination degree maximum .

[0091] Step S3: Based on the cycle period, the feature nodes are screened to generate a control sequence; based on the control sequence, a bimodal matrix model is constructed, including a first modal matrix being a power supply polarization state anomaly matrix for recording power supply polarization states, and a second modal matrix being an electrolyte chemical state anomaly matrix for recording electrolyte chemical states;

[0092] Exemplarily, constructing the bimodal matrix model includes:

[0093] The feature nodes are screened with the coordination time delay scale as the cycle period to obtain the control sequence of the feature nodes;

[0094] The numerical collection instructions of the power supply polarization states and the electrolyte chemical states are triggered at each feature node in the control sequence, and a bimodal matrix model is constructed, wherein the first modal matrix is a power supply polarization state anomaly matrix, the second modal matrix is an electrolyte chemical state anomaly matrix, and the column indexes of the bimodal matrix are all the serial numbers of the feature nodes, the row indexes of the first modal matrix are the serial numbers of the power supply polarization state types, and the row indexes of the second modal matrix are the serial numbers of the electrolyte chemical state types;

[0095] In the first modal matrix, if the power supply polarization state value corresponding to the e-th power supply polarization state type is detected to be abnormal at the i-th feature node , the matrix position of the e-th row and the i-th column in the first modal matrix is set to 1; if the power supply polarization state value corresponding to the e-th power supply polarization state type is detected to be normal at the i-th feature node , the matrix position of the e-th row and the i-th column in the first modal matrix is set to 0;

[0096] In the second modal matrix, if the electrolyte chemical state value corresponding to the r-th electrolyte chemical state type is detected to be abnormal at the i-th feature node , the matrix position of the r-th row and the i-th column in the second modal matrix is set to 1; if the electrolyte chemical state value corresponding to the r-th electrolyte chemical state type is detected to be normal at the i-th feature node , the matrix position of the r-th row and the i-th column in the second modal matrix is set to 0;

[0097] For example, taking k=5 as the cycle period, screening feature nodes 1, 6, 11…116, and generating a control sequence of 24 feature nodes;

[0098] The power polarization state type is set to two (current fluctuation anomaly, polarization resistance anomaly), and the electrolyte chemical state type is set to three (PH anomaly, concentration anomaly, ion activity anomaly); the first modal matrix has a dimension of 2x24, and the second modal matrix has a dimension of 3x24; for example, if the 6th feature node detects a polarization resistance anomaly (power polarization state type 2), the first modal matrix has 1 in the 2nd row and the 2nd column; and if the electrolyte concentration is normal, the second modal matrix has 0 in the 2nd row and the 2nd column.

[0099] Step S4: based on the first modal matrix and the second modal matrix, a synergistic abnormal feedback matrix model of the power polarization state and the electrolyte chemical state is built to quantify the power abnormal fitting coefficient and the electrolyte abnormal fitting coefficient;

[0100] Exemplarily, building the synergistic abnormal feedback matrix model includes:

[0101] Based on the first modal matrix and the second modal matrix, a synergistic abnormal feedback matrix model of the power polarization state and the electrolyte chemical state is built , wherein, the first modal matrix is represented by, the second modal matrix is represented by, T is a matrix transposition symbol, E represents the total number of power polarization state types, R represents the total number of electrolyte chemical state types, and S represents the total number of feature nodes contained in the control sequence:

[0102] The synergistic abnormal feedback matrix model is summed up for each matrix element value to obtain a synergistic abnormal feedback value; the synergistic abnormal feedback matrix model is summed up for the e-th row matrix element value to obtain a power polarization state abnormal feedback value of the e-th power polarization state type varying with the feature nodes, and the ratio of the power polarization state abnormal feedback value to the synergistic abnormal feedback value is taken as a power abnormal fitting coefficient, which is used to represent the power abnormal probability; the synergistic abnormal feedback matrix model is summed up for the r-th column matrix element value to obtain an electrolyte chemical state abnormal feedback value of the r-th electrolyte chemical state type varying with the feature nodes, and the ratio of the electrolyte chemical state abnormal feedback value to the synergistic abnormal feedback value is taken as an electrolyte abnormal fitting coefficient, which is used to represent the electrolyte abnormal probability.

[0103] Step S5: Collect the values of the power polarization state and the electrolyte chemical state at the current time sequence node, perform time sequence augmentation on the bimodal matrix model, and substitute it into the collaborative anomaly feedback matrix model to generate new power anomaly fitting coefficients and electrolyte anomaly fitting coefficients to instruct the adjustment of power parameters and electrolyte parameters;

[0104] Illustratively, the instruction for adjusting the power parameters and the electrolyte parameters includes:

[0105] Collect the values of the power polarization state and the electrolyte chemical state at the current time sequence node, return to step SS, perform column index augmentation on the bimodal matrix model, determine whether the power polarization state value or the electrolyte chemical state value is normal or abnormal, and record the determination result to the augmented column;

[0106] After the augmentation is completed, step S4 is executed to evaluate the new power anomaly fitting coefficients and electrolyte anomaly fitting coefficients through the collaborative anomaly feedback matrix model;

[0107] If the new power anomaly fitting coefficient corresponding to the e-th power polarization state type is greater than the power anomaly fitting coefficient before augmentation, the e-th power polarization state type is traced and locked, and a power parameter adjustment instruction is sent to the staff port, and if the new power anomaly fitting coefficient is less than or equal to the power anomaly fitting coefficient before augmentation, no power parameter adjustment instruction is generated;

[0108] If the new electrolyte anomaly fitting coefficient corresponding to the r-th electrolyte chemical state type is greater than the electrolyte anomaly fitting coefficient before augmentation, the r-th electrolyte chemical state type is traced and locked, and an electrolyte parameter adjustment instruction is sent to the staff port, and if the new electrolyte anomaly fitting coefficient is less than or equal to the electrolyte anomaly fitting coefficient before augmentation, no electrolyte parameter adjustment instruction is generated;

[0109] For example, the new power anomaly fitting coefficient > before augmentation, the power parameter adjustment instruction is sent, the voltage adjustment is adjusted in the direction of “reducing the voltage variance and improving the stability of the voltage mean value”, the ideal voltage mean value under no disturbance can be referred to, the current voltage is corrected, the power parameter correction value is closer to the ideal range, and finally the power stability index is improved; the new electrolyte anomaly fitting coefficient > before augmentation, the electrolyte parameter adjustment instruction is sent, the temperature adjustment is adjusted in the direction of “reducing the temperature variance and improving the stability of the temperature mean value”, the ideal temperature mean value under no disturbance can be referred to, the current electrolyte temperature is corrected, the electrolyte parameter correction value is closer to the ideal range, and finally the electrolyte stability index is improved.

[0110] In the second embodiment: an electrochemical forming process parameter optimization control system for an aviation case wall shell structure is provided, which comprises a data acquisition module, a data processing and fitting optimization module, a matrix model construction module, and a parameter adjustment instruction generation module;

[0111] The data acquisition module is used to acquire various types of time sequence data in the electrochemical forming process of the aviation case wall shell and to screen and mark characteristic nodes.

[0112] The data acquisition module comprises a time sequence data acquisition unit and a characteristic node marking unit. The time sequence data acquisition unit is used to acquire time sequence data of environmental disturbances, power supply parameters, and electrolyte parameters. The characteristic node marking unit is used to align time sequence nodes and mark characteristic nodes that make the processing state stable.

[0113] The data processing and fitting optimization module is used to construct a processing state stability index model and a state deviation amplitude model, to perform function fitting and optimization of the coordinated time delay scale of the processing state evolution curve.

[0114] The data processing and fitting optimization module comprises a stability index calculation unit, a state deviation amplitude evaluation unit, and a coordinated time delay scale optimization unit. The stability index calculation unit is used to quantify power supply stability indexes and electrolyte stability indexes. The state deviation amplitude evaluation unit is used to evaluate the state deviation amplitude relative to a reference benchmark under environmental disturbances. The coordinated time delay scale optimization unit is used to determine the optimal coordinated time delay scale based on the processing field coordination degree.

[0115] The matrix model construction module is used to generate a control sequence, construct a bimodal matrix model, and construct a collaborative abnormal feedback matrix model.

[0116] The matrix model construction module comprises a bimodal matrix generation unit and a collaborative abnormal feedback matrix building unit. The bimodal matrix generation unit is used to record abnormal situations of power supply polarization states and electrolyte chemical states. The collaborative abnormal feedback matrix building unit is used to quantify power supply abnormal fitting coefficients and electrolyte abnormal fitting coefficients.

[0117] The parameter adjustment instruction generation module is used to acquire relevant state values at the current time sequence node, to perform time sequence augmentation on the bimodal matrix model, to calculate new abnormal fitting coefficients, and to generate corresponding parameter adjustment instructions.

[0118] The parameter adjustment instruction generation module comprises a matrix time sequence augmentation unit, an abnormal fitting coefficient updating unit, and an adjustment instruction sending unit. The matrix time sequence augmentation unit is used to expand the column index of the bimodal matrix and to record the current state judgment result. The abnormal fitting coefficient updating unit is used to recalculate the abnormal fitting coefficients. The adjustment instruction sending unit is used to send adjustment instructions or no instructions to the staff port according to the coefficient change.

[0119] It should be noted that the relationship terms, such as first and second, and the like, are used only to differentiate one entity or action from another, and do not necessarily require or imply any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0120] Finally, it should be noted that the above-described embodiments are merely possible implementations of the present application, and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, the technical solutions recorded in the foregoing embodiments can still be modified or replaced by other technically equivalent or similar technical features by those skilled in the art, and the modifications, equivalent replacements, improvements, and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for process parameter optimization control of electrochemical forming of an aero case wall shell structure, characterized in that, The method comprises the following steps: Step S1: Collecting environmental disturbance time series data, power supply parameter time series data and electrolyte parameter time series data in the process of electrochemical forming of the aviation case shell, and performing feature node screening and marking, the feature node being in a corresponding relationship with the time series node; taking the steady state working condition of the electrochemical forming process under no environmental disturbance as a reference benchmark, recording the state deviation amplitude relative to the reference benchmark under environmental disturbance, and constructing a processing state evolution curve; A processing state stability index model is constructed, a feature sample is formed by the power supply parameter set and the electrolyte parameter set at the feature node, and the power supply stability index and the electrolyte stability index under environmental disturbance or no environmental disturbance are quantitatively obtained through the mean and square values of voltage and temperature; A state deviation amplitude model is constructed, and the state deviation amplitude is evaluated through the power supply stability index and the electrolyte stability index; Step S2: Function fitting is performed on the processing state evolution curve, the independent variable of the fitting function is the feature node, and the dependent variable is the state deviation amplitude. Based on the fitting function, the processing field coordination degree is evaluated to optimize the coordination time scale as the cycle period; Step S3: Based on the cycle period, the feature nodes are screened to generate a control sequence; based on the control sequence, a bimodal matrix model is constructed, including a first modal matrix being a power supply polarization state abnormal matrix for recording the power supply polarization state, and a second modal matrix being an electrolyte chemical state abnormal matrix for recording the electrolyte chemical state; Step S4: Based on the first modal matrix and the second modal matrix, a synergistic abnormal feedback matrix model of the power supply polarization state and the electrolyte chemical state is built to quantize the power supply abnormal fitting coefficient and the electrolyte abnormal fitting coefficient; Step S5: The values of the power supply polarization state and the electrolyte chemical state at the current time series node are collected, the bimodal matrix model is time series augmented, and the new power supply abnormal fitting coefficient and the electrolyte abnormal fitting coefficient are generated by substituting into the synergistic abnormal feedback matrix model, so as to instruct the adjustment of the power supply parameter and the electrolyte parameter.

2. The electrochemical forming process parameter optimization control method for an aerospace case wall enclosure structure of claim 1, wherein, The processing state evolution curve comprises: Based on the environmental disturbance time series data in the process of electrochemical forming of the aviation case shell, the power supply parameter time series data and the electrolyte parameter time series data are time series node aligned to mark the feature nodes that make the aviation case shell electrochemical forming process reach the processing state stability under the coordination of the power supply parameter and the electrolyte parameter, the feature nodes being the time series nodes that make the aviation case shell electrochemical forming process reach the processing state stability; Taking the steady state working condition of the electrochemical forming process under no environmental disturbance as a reference benchmark, the state deviation amplitude of the aviation case shell electrochemical forming process relative to the reference benchmark is recorded, and the state deviation amplitude is fitted into a processing state two-dimensional coordinate system, wherein the horizontal coordinate of the processing state two-dimensional coordinate system corresponds to the feature node, and the vertical coordinate of the processing state two-dimensional coordinate system corresponds to the state deviation amplitude, to obtain the processing state evolution curve.

3. The electrochemical forming process parameter optimization control method for an aerospace case wall enclosure structure of claim 1, wherein, The process of constructing the processing state stability index model in step S1 is as follows: The power supply parameter set and the electrolyte parameter set at each feature node constitute a group of feature samples, and the power supply parameter set and the electrolyte parameter set record the power supply parameters and the electrolyte parameters generated at the feature node respectively, and the group number of the feature sample corresponds to the sequence number of the feature node; According to the time sequence arrangement of the feature nodes, sequentially arranged groups of feature samples are formed, and the groups of feature samples are parameter corrected: The power supply parameter correction value of the i-th group of characteristic samples The electrolyte parameter correction value of the i-th group of characteristic samples In the formula, And The voltage mean value and the temperature mean value of the i-th group of characteristic samples, And The voltage variance and the temperature variance of the i-th group of characteristic samples; The power supply stability index is quantified based on the power supply parameter correction value The electrolyte stability index is quantified based on the electrolyte parameter correction value In the formula, max{} and min{} are maximum and minimum value functions, respectively, for selecting the maximum and minimum values in the power supply parameter correction value and the electrolyte parameter correction value, respectively.

4. The electrochemical forming process parameter optimization control method for an aerospace case wall enclosure structure of claim 3, wherein, The state deviation amplitude model is constructed in step S1 as follows: The power supply parameter and the electrolyte parameter under the environmental disturbance are substituted into the processing state stability index model to obtain the processing state index characteristic point under the environmental disturbance wherein, and are the power supply stability index and the electrolyte stability index obtained under the environmental disturbance, respectively. The power supply parameter and the electrolyte parameter without environmental disturbance are substituted into the processing state stability index model to obtain the processing state index characteristic point without environmental disturbance wherein, and are the power supply stability index and the electrolyte stability index obtained without environmental disturbance, respectively. Assessing state deviation magnitude .

5. The method of electrochemical forming process parameter optimization control for an aerospace pocket wall enclosure structure of claim 1, wherein, The optimization of the cooperative time delay scale in step S2 includes: The fitting function of the processing state evolution curve is , x is the serial number of the feature node, and ; The initialization of the coordination time delay scale k between the power smoothness index and the electrolyte smoothness index, and the coordination degree in the processing field under the coordination time delay scale k , I is the total number of feature nodes; Adjust the value of the coordination time delay scale k, let k=k+1, get the processing field coordination degree under the coordination time delay scale k+1 , select the coordination time delay scale that makes the processing field coordination degree maximum .

6. The method of electrochemical forming process parameter optimization control for an aerospace case wall enclosure structure of claim 1, wherein, The construction of the bimodal matrix model in step S3 includes: The feature nodes are screened with the cooperative time delay scale as the cycle period to obtain a control sequence of the feature nodes; The numerical acquisition instructions of the power supply polarization state and the electrolyte chemical state are triggered at each feature node in the control sequence, and a bimodal matrix model is constructed, wherein the first modal matrix is a power supply polarization state anomaly matrix, the second modal matrix is an electrolyte chemical state anomaly matrix, and the column index of the bimodal matrix is the sequence number of the feature node, the row index of the first modal matrix is the sequence number of the power supply polarization state type, and the row index of the second modal matrix is the sequence number of the electrolyte chemical state type; In the first modal matrix, if the power polarization state value corresponding to the e-th power polarization state type is detected to be abnormal at the i-th feature node , the matrix position of the e-th row i-th column in the first modal matrix is set to 1, and if the power polarization state value corresponding to the e-th power polarization state type is detected to be normal at the i-th feature node , the matrix position of the e-th row i-th column in the first modal matrix is set to 0. In the second modal matrix, if the electrolyte chemical state value corresponding to the rth electrolyte chemical state type is detected to be abnormal at the ith characteristic node , the matrix position of the rth row and the ith column in the second modal matrix is set to 1, and if the electrolyte chemical state value corresponding to the rth electrolyte chemical state type is detected to be normal at the ith characteristic node , the matrix position of the rth row and the ith column in the second modal matrix is set to 0.

7. The method of electrochemical forming process parameter optimization control for an aerospace pocket wall enclosure structure of claim 1, wherein, The cooperative anomaly feedback matrix model is built in step S4, and the cooperative anomaly feedback matrix model includes: Based on the first modal matrix and the second modal matrix, a coordinated abnormal feedback matrix model of the power supply polarization state and the electrolyte chemical state is built , wherein, the first modal matrix is represented by, the second modal matrix is represented by, T is a matrix transpose symbol, E represents the total number of power supply polarization state types, R represents the total number of electrolyte chemical state types, and S represents the total number of feature nodes contained in the control sequence: Summing up the matrix element values in the collaborative abnormal feedback matrix model , a collaborative abnormal feedback value is obtained; summing up the matrix element values in the e-th row of the collaborative abnormal feedback matrix model , an e-th power supply polarization state type abnormal feedback value varying with the feature node is obtained, and a ratio of the power supply polarization state abnormal feedback value to the collaborative abnormal feedback value is taken as a power supply abnormal fitting coefficient of the e-th power supply polarization state type, which is used to represent a power supply abnormal probability triggered by the e-th power supply polarization state type; summing up the matrix element values in the r-th column of the collaborative abnormal feedback matrix model , an r-th electrolyte chemical state type abnormal feedback value varying with the feature node is obtained, and a ratio of the electrolyte chemical state abnormal feedback value to the collaborative abnormal feedback value is taken as an electrolyte abnormal fitting coefficient of the r-th electrolyte chemical state type, which is used to represent an electrolyte abnormal probability triggered by the r-th electrolyte chemical state type.

8. The method of electrochemical forming process parameter optimization control for an aerospace case wall enclosure structure of claim 1, wherein, The adjustment of the power supply parameters and the electrolyte parameters in step S5 includes: The numerical values of the power supply polarization state and the electrolyte chemical state at the current time sequence node are collected, and the step SS is returned to perform column index augmentation on the bimodal matrix model, to determine whether the power supply polarization state value or the electrolyte chemical state value is normal or abnormal, and record the determination result to the augmented column; After the augmentation is completed, step S4 is executed to evaluate new power supply anomaly fitting coefficients and new electrolyte anomaly fitting coefficients through the cooperative anomaly feedback matrix model; If the new power supply anomaly fitting coefficient corresponding to the e-th power supply polarization state type is greater than the power supply anomaly fitting coefficient before the augmentation, the e-th power supply polarization state type is traced and locked, and a power supply parameter adjustment instruction is sent to the staff port, and if the new power supply anomaly fitting coefficient is less than or equal to the power supply anomaly fitting coefficient before the augmentation, no power supply parameter adjustment instruction is generated; If the new electrolyte anomaly fitting coefficient corresponding to the r-th electrolyte chemical state type is greater than the electrolyte anomaly fitting coefficient before the augmentation, the r-th electrolyte chemical state type is traced and locked, and an electrolyte parameter adjustment instruction is sent to the staff port, and if the new electrolyte anomaly fitting coefficient is less than or equal to the electrolyte anomaly fitting coefficient before the augmentation, no electrolyte parameter adjustment instruction is generated.

9. A control system for performing the electrochemical forming process parameter optimization control method for an aerospace pocket wall enclosure structure as claimed in claim 1, characterized by, The system includes a data acquisition module, a data processing and fitting optimization module, a matrix model construction module, and a parameter adjustment instruction generation module; The data acquisition module is used to collect various time sequence data in the process of the electrochemical forming of the aviation case wall shell, and to screen and mark feature nodes; The data processing and fitting optimization module is used to construct a processing state stability index model and a state deviation amplitude model, to perform function fitting on the processing state evolution curve, and to optimize the cooperative time delay scale; The matrix model construction module is configured to generate a control sequence, construct a bimodal matrix model, and construct a collaborative abnormal feedback matrix model. The parameter adjustment instruction generation module is configured to collect relevant state values at a current time sequence node, perform time sequence augmentation on the bimodal matrix model, calculate new abnormal fitting coefficients, and generate corresponding parameter adjustment instructions.

10. The control system of claim 9, wherein, The data acquisition module includes a time sequence data acquisition unit and a feature node marking unit. The time sequence data acquisition unit is configured to acquire time sequence data of environmental disturbances, power supply parameters, and electrolyte parameters. The feature node marking unit is configured to align time sequence nodes and mark feature nodes that make the processing state stable. The data processing and fitting optimization module includes a stability index calculation unit, a state deviation amplitude evaluation unit, and a collaborative time delay scale optimization unit. The stability index calculation unit is configured to quantify power stability indexes and electrolyte stability indexes. The state deviation amplitude evaluation unit is configured to evaluate state deviation amplitudes relative to a reference benchmark under environmental disturbances. The collaborative time delay scale optimization unit is configured to determine an optimal collaborative time delay scale based on processing field collaboration. The matrix model construction module includes a bimodal matrix generation unit and a collaborative abnormal feedback matrix building unit. The bimodal matrix generation unit is configured to record abnormal conditions of power polarization states and electrolyte chemical states. The collaborative abnormal feedback matrix building unit is configured to quantify power abnormal fitting coefficients and electrolyte abnormal fitting coefficients. The parameter adjustment instruction generation module includes a matrix time sequence augmentation unit, an abnormal fitting coefficient updating unit, and an adjustment instruction sending unit. The matrix time sequence augmentation unit is configured to expand column indexes of the bimodal matrix and record current state judgment results. The abnormal fitting coefficient updating unit is configured to recalculate abnormal fitting coefficients. The adjustment instruction sending unit is configured to send adjustment instructions or no instructions to a staff port according to coefficient changes.

Images