Waste acid utilization facility environment risk evaluation method and device based on multi-source data

Abstract

The invention relates to the technical field of environmental risk assessment, in particular to a waste acid utilization facility environmental risk assessment method and device based on multi-source data, and the method comprises the following steps: collecting waste acid kettle operation data, calculating electrochemical signal entropy, establishing a corrosion stress field through weighting, calculating a pitting rate, and measuring and calculating punching time in combination with the property of waste acid. According to the method, voltage fluctuation and mechanical stress data in the kettle are collected in real time, the electrochemical signal entropy is subjected to trend calculation through a least square method to quantify the corrosion activeness, the corrosion activeness is used as a weight to be fused with fluid pressure and torque to construct a corrosion stress coupling field, and a reduction material stress and chemical erosion interactive damage mechanism is used; and a threshold exceeding unit is extracted, the pitting expansion rate is measured and calculated through an index model, the punching remaining time is calculated in combination with the property of waste acid, risk assessment is converted from qualitative experience to quantitative early warning, hidden pitting leakage is avoided, and timely intervention before physical failure is ensured.

Description

Technical Field

[0001] This invention relates to the field of environmental risk assessment technology, and in particular to a method and apparatus for environmental risk assessment of waste acid utilization facilities based on multi-source data. Background Technology

[0002] The field of environmental risk assessment technology systematically identifies, analyzes, and assesses the adverse impacts of industrial activities and related facilities on soil, water, atmosphere, and ecosystems during construction and operation. Its core aspects include pollution source identification, environmental media receptor analysis, risk factor selection, and risk level evaluation. Typically, it involves analyzing the migration paths and exposure scenarios of environmental media from material flow to emission nodes, using qualitative or quantitative methods to comprehensively determine potential environmental risks, providing a foundation for environmental management and risk prevention. Among these, environmental risk assessment for traditional waste acid utilization facilities refers to a technical solution that analyzes the environmental risks generated during the reception, storage, utilization, and product disposal of inorganic waste acids such as waste sulfuric acid and waste hydrochloric acid. For risk identification and assessment under complex raw material sources, diverse process conditions, and the coexistence of multiple by-product streams, traditional methods typically rely on a single data source, such as production ledger monitoring data or emission inventories, to evaluate the waste acid reception volume, storage conditions, material conversion during utilization, and product disposal destinations, and then make judgments based on established environmental risk assessment indicator systems and empirical coefficients.

[0003] Current technologies generally rely on production ledgers, pollution discharge lists, or end-of-pipe emission monitoring data for environmental risk assessment. They focus on macro-statistical analysis of waste acid reception, total storage, and product destination. When faced with the complex and variable corrosive environment inside waste acid reactors, they often rely solely on established indicator systems or general empirical coefficients for static inference, neglecting the synergistic effect between fluid dynamics and electrochemical corrosion during equipment operation. They struggle to capture non-uniform pitting corrosion caused by stress concentration or localized medium fluctuations on the inner wall. This assessment model, based on historical data or a single dimension, cannot reflect the real degradation state of equipment materials under multi-field coupling effects in real time. This results in a lack of forward-looking prediction of sudden risks caused by structural integrity damage, often only being passively detected after physical perforation occurs. This can easily lead to high-concentration waste acid leaks and cause irreparable damage to soil and groundwater environments. Furthermore, it cannot provide maintenance strategies based on the remaining lifespan of the equipment, leaving environmental safety management in a reactive state. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide an environmental risk assessment method for waste acid utilization facilities based on multi-source data, comprising the following steps: S1: Collect data on voltage fluctuations, hydrostatic pressure, output torque of stirring motor, and geometric dimensions of waste acid reactor wall. Use the least squares method to fit and detrend the voltage fluctuation data of the waste acid reactor wall to generate a detrended voltage fluctuation sequence. S2: Statistically analyze the frequency of the detrended voltage fluctuation sequence within a preset voltage amplitude range, calculate the probability value, and calculate the electrochemical signal entropy based on the probability value; S3: Using the electrochemical signal entropy as a weighting coefficient, construct a mesh model based on the geometric dimensions of the waste acid reactor wall, and calculate the stress of the mesh model by combining the hydrostatic pressure data and the output torque data of the stirring motor to generate a corrosion-stress coupling field matrix. The corrosion-stress coupled field matrix includes spatial position indices corresponding to discrete grid cells on the waste acid reactor wall, and mechanical stress values ​​weighted by electrochemical signal entropy for each grid cell; S4: Extract the set of coupled numerical values ​​of target elements exceeding the critical damage threshold from the corrosion-stress coupled field matrix, and input them into the exponential growth model to calculate the pitting depth propagation rate; S5: Calculate the remaining time for perforation by combining the pitting depth propagation rate with the geometric dimensions of the waste acid reactor wall, obtain the physicochemical properties of the waste acid, and generate a pollution accident risk level assessment result based on the remaining time for perforation and the physicochemical properties of the waste acid.

[0005] As a further aspect of the present invention, the detrended voltage fluctuation sequence includes a time domain index, fluctuation amplitude values, and noise residual components; the electrochemical signal entropy includes information disorder, probability distribution density, and corrosion activity index; the corrosion-stress coupling field matrix includes grid node coordinates, stress intensity vector, and electrochemical weighting factor; the pitting depth propagation rate includes vertical erosion rate, pit diameter growth, and damage increment per unit time; and the pollution accident risk level assessment results include the severity of leakage consequences, environmental sensitivity level, and emergency response priority.

[0006] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Start the electrochemical monitoring probe and sensor group in the waste acid reactor, collect the voltage fluctuation data of the inner wall of the waste acid reactor, the static pressure data of the fluid and the output torque data of the stirring motor, perform multi-channel signal data cleaning and time-domain synchronization processing, unify the time reference and data format of physical quantity measurements, and establish a multi-source operating condition monitoring dataset for the waste acid reactor. S102: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the voltage fluctuation data of the inner wall of the waste acid reactor, construct a polynomial regression model using the least squares method, perform global trend fitting operation on the voltage data time series, analyze the voltage signal reference trend and calculate the fitting curve parameters, and generate the voltage trend polynomial coefficient vector. S103: Construct a time-varying trend function for the voltage signal based on the voltage trend polynomial coefficient vector, call the original voltage fluctuation data of the multi-source operating condition monitoring dataset of the waste acid reactor, calculate the theoretical trend value at time, extract the global trend component from the original observation value and retain the local fluctuation residual to obtain the detrended voltage fluctuation sequence.

[0007] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Call the detrended voltage fluctuation sequence, divide the continuous voltage amplitude interval according to the preset resolution step size, traverse the sampling points and match the corresponding interval index, count the number of samples falling into the interval, construct an array structure that reflects the data density distribution, and generate a voltage interval frequency statistics vector. S202: Based on the frequency statistics vector of the voltage interval and the total sample capacity of the sampling points within the statistical period, normalize the frequency values ​​of the interval, verify the proportion weight of the voltage amplitude interval, complete the conversion from absolute quantity to probability, and generate a voltage amplitude distribution probability sequence. S203: Call the voltage amplitude distribution probability sequence, identify the probability elements and calculate the information components, perform summation and inversion operations on all interval components, quantify the disorder of the voltage fluctuation signal in the time domain distribution, and obtain the electrochemical signal entropy.

[0008] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the geometric dimension data of the waste acid reactor wall, perform discretization segmentation on the inner wall surface according to the preset spatial step size, construct a digital geometric structure including node coordinates and unit topological relationships, map the continuous entity into a virtual computing domain, and generate a three-dimensional discrete mesh model of the waste acid reactor wall. S302: Based on the three-dimensional discrete mesh model of the waste acid reactor wall, the hydrostatic pressure and stirring motor output torque data are called to map the physical load into the normal and tangential force components of the mesh element, perform multi-axial mechanical equilibrium calculation and solve the elastic deformation tensor of the element to generate dynamic mechanical load stress data. S303: Call the electrochemical signal entropy as the corrosion sensitivity weight, combine it with the dynamic mechanical load stress data, perform weighted coupling operation on the grid cells, map the corrosion activity to the stress strengthening coefficient, calculate the field strength value under the combined action of corrosion environment and mechanical load, and generate corrosion-stress coupling field matrix.

[0009] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Call the corrosion-stress coupling field matrix, set the critical damage threshold, traverse the coupling values ​​of matrix grid nodes and perform comparison operations, filter target grid elements whose values ​​exceed the threshold, extract the stress and corrosion coupling strength values ​​corresponding to the risk elements, construct a data subset including the index of damaged node locations, and generate a set of target element coupling values. S402: Based on the coupled numerical set of the target unit, construct an exponential growth model describing the nonlinear expansion of pitting depth, substitute the coupled numerical values ​​as initial driving parameters into the model, and deduce the theoretical pitting depth at a given time step according to a preset time step, forming a data set describing the dynamic increase of depth over time, and generating a pitting depth evolution sequence. S403: Call the pitting depth evolution sequence, calculate the incremental change in pitting depth at adjacent time points in the sequence, determine the slope of depth evolution per unit time, characterize the vertical downward erosion rate of the pitting pit, quantify the dynamic expansion trend of local corrosion in the depth direction, and obtain the pitting depth expansion rate.

[0010] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Obtain the physicochemical properties data of the waste acid, call the pitting depth propagation rate and the three-dimensional discrete mesh model of the waste acid reactor wall, retrieve the structural weak point with the smallest current wall thickness, perform numerical calculations on the remaining wall thickness parameters and pitting rate, deduce the physical time required for the weak point to develop to complete penetration and rupture, and generate the predicted value of the remaining life of the waste acid reactor wall perforation. S502: Based on the predicted remaining life of the waste acid reactor wall perforation, the waste acid solution object to be evaluated is identified, the toxicity index of characteristic pollutants is analyzed, a multidimensional environmental hazard assessment weighting function is constructed, environmental sensitivity weights are assigned to physicochemical indicators, the hazard value is calculated and mapped to the standard quantification range, and the environmental hazard equivalent of waste acid leakage is generated. S503: Call the predicted remaining life of the waste acid reactor wall perforation and the equivalent environmental hazard of the waste acid leakage to construct a risk assessment decision matrix, map the two to the dimensions of time urgency and severity of consequences respectively, determine the cross coordinate position and compare the risk level area boundary, determine the safety warning level of the current operating status, and generate the pollution accident risk level evaluation result.

[0011] As a further aspect of the present invention, the meaning of the preset voltage amplitude range is derived from the amplitude segment set formed after the boundary of the amplitude statistical range of the detrended voltage fluctuation sequence is determined. The amplitude statistical range is jointly defined by the maximum amplitude and the minimum amplitude of the detrended voltage fluctuation sequence, and the interval boundary of the preset voltage amplitude range is determined by the amplitude statistical range according to the equal width division rule.

[0012] As a further aspect of the present invention, the electrochemical signal entropy is calculated based on the interval probability distribution and in combination with the interval boundary consistency condition of the preset voltage amplitude interval; The electrochemical signal entropy is normalized before being used as the weighting coefficient, so that the weighting coefficient is obtained by monotonically mapping the electrochemical signal entropy and corresponds to the interval probability distribution.

[0013] An environmental risk assessment device for waste acid utilization facilities based on multi-source data includes: The data acquisition module is used to execute S1: by collecting data on voltage fluctuations on the inner wall of the waste acid reactor, hydrostatic pressure, output torque of the stirring motor, and geometric dimensions of the waste acid reactor wall through electrochemical monitoring probes and sensor groups arranged in the waste acid reactor of the waste acid utilization facility, and calling the least squares method to perform polynomial fitting detrending processing on the voltage fluctuation data on the inner wall of the waste acid reactor to generate a detrended voltage fluctuation sequence. The signal processing module is used to perform S2: statistically analyze the frequency of occurrence of the detrended voltage fluctuation sequence within the preset voltage amplitude range and calculate the distribution probability value, and calculate the electrochemical signal entropy based on the distribution probability value; The stress and corrosion coupling calculation module is used to execute S3: using the electrochemical signal entropy as a weighting coefficient, combined with the hydrostatic pressure data and the output torque data of the stirring motor, to perform stress calculation and numerical coupling on the mesh model constructed based on the geometric dimensions of the waste acid reactor wall, and generate a corrosion-stress coupling field matrix. The pitting corrosion propagation simulation module is used to perform S4: extract the target element coupling numerical set that exceeds the critical damage threshold in the corrosion-stress coupling field matrix, input the target element coupling numerical set into the exponential growth model for depth evolution simulation, and calculate the pitting corrosion depth propagation rate. The risk assessment module is used to perform S5: acquire the physicochemical properties data of the waste acid, calculate the remaining time for the perforation of the weakest point of the waste acid reactor by combining the pitting depth propagation rate and the geometric dimensions of the waste acid reactor wall, and generate a pollution accident risk level assessment result based on the remaining time for perforation and the physicochemical properties data of the waste acid.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by collecting real-time data on voltage fluctuations and mechanical stress inside the vessel, the electrochemical signal entropy is calculated using the least squares method to quantify corrosion activity. This entropy is then used as a weight to integrate fluid pressure and torque to construct a corrosion stress coupling field, thereby revealing the interaction damage mechanism between raw material stress and chemical erosion. The pitting rate is calculated using an exponential model by extracting over-threshold units, and the remaining time for perforation is calculated by combining the properties of waste acid. This risk assessment shifts from qualitative experience to quantitative early warning, avoiding hidden pitting leaks and ensuring timely intervention before physical failure. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] Please see Figure 1 This invention provides an environmental risk assessment method for waste acid utilization facilities based on multi-source data, comprising the following steps: S1: Collect data on voltage fluctuations, hydrostatic pressure, output torque of stirring motor, and geometric dimensions of waste acid reactor wall by electrochemical monitoring probes and sensor groups arranged in the waste acid reactor of the waste acid utilization facility. Then, use the least squares method to perform polynomial fitting detrending processing on the voltage fluctuation data of the waste acid reactor wall to generate a detrended voltage fluctuation sequence. S2: Call the detrended voltage fluctuation sequence, divide the preset voltage amplitude interval according to the preset resolution step size and perform frequency statistics, generate the voltage interval frequency statistics result, normalize the frequency to obtain the voltage amplitude distribution probability value, and calculate the information component of the probability value based on the voltage amplitude distribution probability value, and perform summation and inversion operation on the probability value to obtain the electrochemical signal entropy; S3: Using the electrochemical signal entropy as a weighting coefficient, combined with the hydrostatic pressure data and the output torque data of the stirring motor, stress calculation and numerical coupling are performed on the mesh model constructed based on the geometric dimensions of the waste acid reactor wall to generate a corrosion-stress coupling field matrix. S4: Extract the target element coupling numerical set that exceeds the critical damage threshold in the corrosion-stress coupling field matrix, input the target element coupling numerical set into the exponential growth model used to characterize the nonlinear growth relationship of pitting depth with time, and calculate the pitting depth propagation rate based on the change in pitting depth at adjacent time steps. The parameters of the exponential growth model were determined by corrosion test data of specimens of the same material under the same physicochemical conditions as the waste acid to be evaluated. S5: Calculate the remaining time for perforation of the waste acid reactor wall based on the pitting depth propagation rate and the geometric dimensions of the waste acid reactor wall. Obtain the physicochemical properties data of the waste acid corresponding to the waste acid to be evaluated. Based on the remaining time for perforation and the physicochemical properties data of the waste acid, map the remaining time for perforation to a time urgency index and map the physicochemical properties data of the waste acid to an environmental hazard index. Based on the correspondence between the time urgency index and the environmental hazard index in the preset risk level matrix, generate the pollution accident risk level assessment result.

[0020] The detrended voltage fluctuation sequence includes a time-domain index, fluctuation amplitude values, and noise residual components; the voltage amplitude distribution probability sequence is used to characterize the probability distribution of the detrended voltage fluctuation sequence within a preset voltage amplitude range; the electrochemical signal entropy is a scalar entropy value calculated based on the voltage amplitude distribution probability sequence, used to characterize the disorder of the voltage fluctuation signal and reflect corrosion activity; the corrosion sensitivity weighting coefficient is a weighting parameter obtained by normalizing or hierarchically mapping the electrochemical signal entropy; the corrosion-stress coupling field matrix includes grid node coordinates, stress intensity vector, and electrochemical weighting factor; the pitting depth propagation rate includes the increase in pit depth per unit time in the vertical direction, the increase in pit diameter, and the increase in damage per unit time; the pollution accident risk level assessment results include the severity of leakage consequences, environmental sensitivity level, and emergency response priority.

[0021] Please see Figure 2 The specific steps of S1 are as follows: S101: Start the electrochemical monitoring probe and sensor group in the waste acid reactor, collect the voltage fluctuation data of the inner wall of the waste acid reactor, the static pressure data of the fluid and the output torque data of the stirring motor, perform multi-channel signal data cleaning and time-domain synchronization processing, unify the time reference and data format of physical quantity measurements, and establish a multi-source operating condition monitoring dataset for the waste acid reactor. Baseline acquisition of the inner wall voltage signal of the waste acid reactor under static conditions was performed for 30 seconds. During this period, all voltage sampling points were recorded at a frequency of 100 Hz, and the median value was calculated for the sampling sequence. The median value was used as the voltage zero-point reference value for this monitoring. Subsequently, a liquid level correspondence calibration operation was performed on the hydrostatic pressure sensor. The hydrostatic pressure value corresponding to the known liquid level height was compared. If the deviation between the real-time measured value and the theoretical value exceeded 2 kPa, the deviation was recorded and used as the pressure compensation amount. Next, the torque signal output of the stirring motor was acquired under no-load and low-load conditions. The stable torque value was recorded under no-load conditions, and then a standard resistance was applied at low speed, and the torque change amplitude was recorded. The difference between the two was used as the basis for torque reference correction for this time. After calibration, the synchronous acquisition stage began. Under normal operating conditions of the waste acid reactor, data on inner wall voltage fluctuations, hydrostatic pressure, and stirring motor output torque were acquired simultaneously. The voltage data acquisition frequency was set to 100 Hz, and the pressure and torque data acquisition frequencies were set to 10 Hz. During the data acquisition process, a unified timestamp is written to each data point, with a time precision set to milliseconds. Subsequently, a cleaning operation is performed on the acquired multi-channel data. Records with null values ​​are directly discarded, and for data with duplicate timestamps, only the last one written is retained. Outliers are removed or replaced: a sliding window is formed with n sampling points before and after sampling point i, and the median Med and median absolute deviation (MAD) of the voltage values ​​within the window are calculated. When sampling point i satisfies |Vi−Med|>3×MAD, the sampling point is identified as an outlier, and the outlier is replaced using the median of the voltage values ​​within the sliding window (n is 5 or 10), with the replacement value being the median of the adjacent time period. Finally, time synchronization processing is performed, using the voltage sampling time axis as the main time reference, and interpolation is performed on the pressure and torque data to ensure that the three types of data have a corresponding relationship at the same time point, thus forming a multi-source operating condition monitoring dataset for the waste acid reactor that can be directly used in subsequent steps.

[0022] S102: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the voltage fluctuation data of the inner wall of the waste acid reactor, construct a polynomial regression model using the least squares method, perform global trend fitting operation on the voltage data time series, analyze the voltage signal baseline trend and calculate the fitting curve parameters, and generate the voltage trend polynomial coefficient vector. Inner wall voltage fluctuation data were extracted from the multi-source operating condition monitoring dataset and segmented according to continuous operating intervals, with each segment lasting 2400 seconds. After extraction, a corresponding time series was generated for each voltage data segment, with the first sampling time set as the zero point and the remaining sampling points expressed in relative seconds. A global trend analysis was then performed. To avoid interference from long-term slow voltage drift on local fluctuation analysis, a multi-order trend fitting attempt was performed on the time series. Specifically, second-order, third-order, and fourth-order trend structures were constructed, generating a trend curve for each structure, and the deviation between the original voltage data and the trend curve was calculated. Deviation evaluation was performed by comparing the average level of the absolute values ​​of the deviations of all sampling points, while also recording the number of inflection points in the trend curve to characterize its smoothness. The order selection rule prioritized the order with the smallest average deviation; if the difference in deviation between adjacent orders was less than 0.03 mV, the order with fewer inflection points was selected. After the order was determined, a unified solution operation was performed on the trend parameters under that order, using the minimum overall deviation of all sampling points as a constraint. To ensure the comparability of parameters across different time windows, the time series is scaled before parameter calculation to normalize the maximum time value to within 1, and the corresponding scaling ratio is recorded. Finally, the obtained trend parameters of each order are combined in order of order into a voltage trend polynomial coefficient vector, which is stored together with the current operating condition segment number to provide input for subsequent trend stripping operations.

[0023] S103: Construct a time-varying trend function for voltage signal based on voltage trend polynomial coefficient vector, call the original voltage fluctuation data of the multi-source operating condition monitoring dataset of waste acid reactor, calculate the theoretical trend value at time, extract the global trend component from the original observation value and retain the local fluctuation residual to obtain the detrended voltage fluctuation sequence. For any sampling time, its relative time value is first converted according to the scaling ratio, and then substituted into the trend parameters of each order to calculate the theoretical trend voltage value corresponding to that time. Then, the original voltage fluctuation data within the same operating condition segment is retrieved again, and a trend stripping operation is performed on each valid sampling point. The specific execution method of trend stripping is to subtract the original voltage observation value from the theoretical trend voltage value at that time; the difference is used as the local voltage fluctuation residual for that sampling point. This residual value, along with the original timestamp, is written into the new sequence structure to maintain temporal continuity. After processing all sampling points, a consistency check is performed on the generated detrended voltage fluctuation sequence. The check includes two indicators: residual mean and residual fluctuation range. The residual mean is required to have an absolute value not exceeding 0.03 mV, and the residual fluctuation range is required to have a standard deviation difference of no more than 0.05 mV between the first and second halves. If either condition is not met, the process returns to the previous stage to readjust the trend order or time window length. After verification, it was confirmed that the global trend component of the sequence had been effectively removed, and only the voltage fluctuation information reflecting local corrosion and disturbance behavior was retained. This detrended voltage fluctuation sequence was then used as the direct input data for subsequent statistical analysis steps.

[0024] Please see Figure 3 The specific steps of S2 are as follows: S201: Call the detrended voltage fluctuation sequence, divide the continuous voltage amplitude interval according to the preset resolution step size, traverse the sampling points and match the corresponding interval index, count the number of samples falling into the interval, construct an array structure that reflects the data density distribution, and generate a voltage interval frequency statistics vector. First, the detrended voltage fluctuation sequence is read, and statistical analysis is performed on the overall amplitude range of the sequence. The statistical operation includes obtaining the minimum and maximum residual values ​​in the sequence, and adding 0.05 mV at each end as statistical boundary buffers to avoid extreme values ​​trunculating the interval division. Then, the voltage amplitude interval is divided according to a preset resolution step size of 0.02 mV, increasing progressively from the minimum boundary to form a continuous and non-overlapping set of amplitude intervals. After interval division, a point-by-point traversal statistical process is initiated. For each sampling point in the detrended voltage fluctuation sequence, its amplitude interval index is determined. The determination rule is that the sampling point value is greater than or equal to the lower limit of the interval and less than the upper limit; if the condition is met, the match is successful. After matching, the count value of the corresponding interval is incremented by 1. For sampling points exceeding the statistical boundaries, they are uniformly assigned to the nearest boundary interval, and the number of out-of-bounds occurrences is recorded as a data quality marker. After traversal, the count values ​​of all intervals are arranged in interval order to form a voltage interval frequency statistical vector, where each element represents the number of sampling points within the corresponding amplitude interval. This statistical vector fully reflects the density distribution of detrended voltage fluctuations in different amplitude ranges, providing a basic data structure for subsequent probabilistic processing and information calculation.

[0025] S202: Based on the frequency statistics vector of voltage intervals and the total sample capacity of sampling points within the statistical period, normalize the frequency values ​​of the intervals, verify the proportion weight of the voltage amplitude intervals, complete the conversion from absolute quantity to probability, and generate a voltage amplitude distribution probability sequence. The voltage interval frequency statistics vector is invoked, and the cumulative result of the count values ​​of all intervals in the vector is calculated to obtain the total number of sampling points participating in the statistics. This total number must be consistent with the number of valid sampling points in the original detrended voltage fluctuation sequence. If there is a discrepancy, the process returns to the previous stage to check for out-of-bounds processing and missing measurement removal records. After confirmation, normalization is performed on each amplitude interval by calculating the ratio of the number of samples in that interval to the total number of sampling points, thus obtaining the weight of that interval in the entire sequence. After all intervals are normalized in sequence, a voltage amplitude distribution probability sequence consistent with the interval order is formed. Subsequently, an integrity check is performed on the probability sequence by accumulating the probability values ​​of all intervals. The accumulated result is required to fall between 0.99 and 1.01. If it exceeds this range, it indicates that there is a statistical omission or duplication, and the frequency statistics need to be re-executed. After passing the check, the probability value corresponding to each amplitude interval and its interval boundary are saved together for subsequent information calculation. After this step is completed, the original distribution information represented by the number of samples is uniformly converted into a probability form, thereby eliminating the impact of the difference in the number of sampling points in different time windows and making the voltage fluctuation distribution under different operating conditions comparable.

[0026] S203: Call the voltage amplitude distribution probability sequence, identify the probability elements and calculate the information components, perform summation and inversion operations on all interval components, quantify the disorder of the voltage fluctuation signal in the time domain distribution, and obtain the electrochemical signal entropy. The voltage amplitude distribution probability sequence is read, and the validity of each probability element in the sequence is judged one by one. Intervals with a probability value of 0 are directly marked as not participating in subsequent calculations to avoid non-numerical cases during logarithmic calculations. Then, for each non-zero probability interval, an information component calculation operation is performed. The calculation logic is as follows: first, the probability of the interval is logarithmically processed; then, the logarithmic result is multiplied by the original probability value; finally, the negative of the result is taken. The resulting value is the contribution of that amplitude interval to the overall disorder level. After completing the calculation for a single interval, the contribution values ​​of all intervals are accumulated, and the accumulated result is used as the electrochemical signal entropy for that operating condition segment. Then, an interval determination operation is performed on the obtained entropy value, based on the historical entropy value distribution. The historical distribution consists of entropy values ​​from no fewer than 30 similar operating conditions, sorted from smallest to largest, with the 25th and 75th percentiles used as interval boundaries. If the current entropy value is below the 25th percentile, it is classified into the low disorder interval; if it is between two threshold values, it is classified into the medium disorder interval; and if it is above the 75th percentile, it is classified into the high disorder interval. After the determination is completed, the entropy value and its interval attributes are stored together with the current operating condition number as important input parameters for subsequent corrosion sensitivity weight mapping.

[0027] Please see Figure 4 The specific steps of S3 are as follows: S301: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the geometric dimension data of the waste acid reactor wall, perform discretization segmentation on the inner wall surface according to the preset spatial step size, construct a digital geometric structure including node coordinates and unit topological relationships, map the continuous entity into a virtual computing domain, and generate a three-dimensional discrete mesh model of the waste acid reactor wall. First, information on the geometric dimensions of the waste acid reactor wall was extracted from the multi-source operating condition monitoring dataset. This information came from as-built drawings and periodic wall thickness re-measurement data, including the reactor's inner diameter, cylinder height, head curvature radius, and measured wall thickness values ​​at multiple measuring points. Then, the inner wall surface of the waste acid reactor was discretized according to a preset spatial step size of 10 mm. During discretization, height layers were divided along the axial direction of the waste acid reactor according to the step size. Nodes were then evenly divided circumferentially within each height layer, ensuring the arc length between adjacent nodes was close to the set step size. For the head region, equal-arc-length rings were established based on its curvature radius, and these rings were continuously connected to the nodes in the cylinder region. After node division, three-dimensional coordinates were generated for each node, with the origin set at the center of the bottom of the waste acid reactor, the height direction as the axial direction, and the horizontal direction unfolded into planar coordinates. Subsequently, mesh elements were generated based on the node connection relationships. Adjacent nodes were combined to form surface elements, and the node index corresponding to each element was recorded. After mesh generation, the mesh quality is checked, including element angles, element area change rates, and surface fitting errors. Regions that do not meet the requirements are locally refined, with the spatial step size reduced to 5 millimeters and the mesh re-divided. The final result is a three-dimensional discrete mesh model that accurately reflects the true geometry of the waste acid reactor.

[0028] S302: Based on the three-dimensional discrete mesh model of the waste acid reactor wall, the fluid static pressure and stirring motor output torque data are called to map the physical load into the normal and tangential force components of the mesh element, perform multi-axial mechanical equilibrium calculation and solve the elastic deformation tensor of the element to generate dynamic mechanical load stress data. A three-dimensional discrete mesh model of the waste acid reactor wall is invoked, and the hydrostatic pressure data and agitator motor output torque data within the same operating condition are simultaneously read. The hydrostatic pressure is then mapped to the normal load of the mesh elements. For each mesh element, the hydrostatic pressure value at the corresponding moment is read based on its height, and this pressure value is multiplied by the element area to obtain the normal force borne by that element. This normal force is distributed to the element nodes according to the inverse proportionality between the node and the geometric center of the element. Next, the agitator motor output torque is mapped to the wall tangential load. The equivalent radius of action is calculated based on the impeller radius and liquid level, and the torque is converted into a shear force distributed circumferentially. This shear force mainly acts on the mesh zone near the impeller height and decreases layer by layer with height difference. After completing the load mapping, the normal and tangential forces from each adjacent element are summed for each node to form the resultant force. Then, considering the constraints of the waste acid reactor support position, the node displacements are solved, with the constraint condition set so that the triaxial displacement of the nodes in the support area is zero. Based on the obtained nodal displacement changes, the strain state of each mesh element is calculated, and combined with the material elastic parameters, the corresponding dynamic mechanical load stress data is obtained. This stress data is recorded in chronological order to provide input for subsequent coupled calculations.

[0029] S303: Call the electrochemical signal entropy as the corrosion sensitivity weight, combine it with dynamic mechanical load stress data, perform weighted coupling operation on the mesh element, map the corrosion activity to the stress strengthening coefficient, calculate the field strength value under the combined action of corrosion environment and mechanical load, and generate corrosion-stress coupling field matrix. The entropy of the electrochemical signal is read and mapped to corrosion sensitivity weights. The mapping rule is set based on the historical entropy value interval division results: the low disorder interval corresponds to a weight range of 0.60 to 0.80, the medium disorder interval corresponds to 0.80 to 1.00, and the high disorder interval corresponds to 1.00 to 1.30. The specific weight value is linearly determined according to the position of the current entropy value within its interval. Then, dynamic mechanical load stress data is read, and the principal stress amplitude of each grid cell is extracted. This amplitude is then compared with the allowable stress of the material to obtain a normalized mechanical load strength index. Next, a weighted coupling operation is performed. For each grid cell, its normalized mechanical load strength is multiplied by the corresponding corrosion sensitivity weight to obtain a stress intensification coefficient. This coefficient is then multiplied again by the original principal stress amplitude to calculate the corrosion-stress coupling field strength value of that cell. After the calculation is completed, the coupling field strength values ​​of all grid cells are written into a matrix structure according to their spatial position in the grid. The matrix index corresponds to the axial and circumferential discrete numbers of the waste acid reactor, thus forming a complete corrosion-stress coupling field matrix.

[0030] Please see Figure 5 The specific steps of S4 are as follows: S401: Call the corrosion-stress coupling field matrix, set the critical damage threshold, traverse the coupling values ​​of matrix grid nodes and perform comparison operations, filter target grid elements whose values ​​exceed the threshold, extract the stress and corrosion coupling strength values ​​corresponding to the risk elements, construct a data subset including the index of damaged node locations, and generate a set of target element coupling values. First, the corrosion-stress coupling field matrix is ​​read, and a critical damage threshold is set. This threshold is determined through calibration using historical maintenance data and wall thickness re-measurement records. During the calibration process, at least 20 actual maintenance samples are selected, and samples with residual wall thickness less than 30% of the nominal wall thickness are screened, and their corresponding coupling field strength values ​​are extracted. These field strength values ​​are sorted, and the median value is taken as a candidate threshold. The false positive and false negative rates of this threshold are verified across the entire sample range. Finally, the field strength value that meets the requirements of a false positive rate of no more than 10% and a false negative rate of no more than 5% is determined as the critical damage threshold. After the threshold is determined, a comparison operation is performed on all grid elements in the corrosion-stress coupling field matrix one by one. Elements with coupling field strength exceeding the threshold are marked as risk elements, and their spatial index and corresponding field strength value are recorded. Subsequently, all risk elements are sorted from high to low according to their coupling field strength values, and the top number of elements are selected as the target element set, which serves as the input object for subsequent pitting corrosion evolution simulation.

[0031] S402: Based on the coupled numerical set of the target unit, an exponential growth model describing the nonlinear expansion of pitting depth is constructed. The coupled numerical values ​​are substituted into the model as initial driving parameters. The theoretical pitting depth at a given time is deduced according to the preset time step, forming a data set describing the dynamic increase of depth over time, and generating a pitting depth evolution sequence. First, the target unit's coupled numerical set is invoked, and a projection process for the pitting depth expansion over time is established for each target unit. The projection parameters are set based on corrosion test results conducted on specimens of the same material under the same waste acid environment. During the test, the maximum pitting depth is measured at 72, 144, and 216 hours, and the test results are categorized according to the coupled field strength interval. Based on the increase in pitting depth over time in different intervals, the corresponding growth rate range is determined. Subsequently, the corresponding growth rate is selected as the projection parameter according to the interval to which the target unit's coupled field strength belongs. The initial pitting depth value is preferentially taken from the maximum pitting depth obtained by on-site endoscopic detection. If no obvious pitting depth is detected, the equivalent pitting depth corresponding to the ultrasonic echo anomaly point is taken as the initial value. During the projection process, the time step is set to 1 hour, and the pitting depth change is projected hourly from the current moment until the preset projection termination time is reached. All the projected pitting depth data are arranged in chronological order to form a pitting depth evolution sequence, and stored in one-to-one correspondence with the target unit number.

[0032] S403: Call the pitting depth evolution sequence, calculate the change increment of pitting depth at adjacent time points in the sequence, determine the depth evolution slope per unit time, characterize the vertical downward erosion rate of pitting pits, quantify the dynamic expansion trend of local corrosion in the depth direction, and obtain the pitting depth expansion rate. First, the pitting depth evolution sequence is read, and then differential processing is performed on the sequence between adjacent time steps. Specifically, for each time step, the pit depth value at the next time step is subtracted from the pit depth value at the previous time step to obtain the change in pit depth within that time period. This change is directly used as the pitting depth expansion slope per unit time. After calculating all time steps, statistical analysis is performed on the obtained slope sequence, including the maximum slope, average slope, and 90th percentile slope. Subsequently, the 90th percentile slope is used as the representative pitting depth expansion rate of the target unit and compared with the experimental calibration interval. The rate interval is divided into a slow interval, a medium-speed interval, and a fast interval, with the interval boundaries determined based on experimental data. After completing the interval determination, the rate value is used as the final pitting depth expansion rate result for the target unit and written into the result set for subsequent remaining lifetime estimation.

[0033] Please see Figure 6 The specific steps of S5 are as follows: S501: Obtain the physicochemical properties data of waste acid, call the pitting depth propagation rate and the three-dimensional discrete mesh model of the waste acid reactor wall, retrieve the structural weak point with the smallest current wall thickness, compare the remaining wall thickness parameter of the weak point with the pitting depth evolution sequence, determine the moment when the pitting depth reaches the remaining wall thickness threshold, and use the time difference between this moment and the current moment as the physical time required for the weak point to develop to complete penetration and rupture, and generate the predicted value of the remaining life of the waste acid reactor wall perforation. First, the physicochemical properties of the waste acid were acquired, sourced from batch testing records and online monitoring results, including acidity, temperature, conductivity, and concentration of major corrosive ions. Then, using a three-dimensional discrete mesh model of the waste acid reactor wall, the structural weak point with the smallest current wall thickness was identified. The search method involved traversing all mesh nodes to obtain the measured wall thickness values, selecting the node with the smallest value, and averaging the values ​​of its neighboring nodes within a certain range to reduce the impact of single-point measurement errors. Next, the pitting depth evolution sequence corresponding to this weak point was obtained. If the node was not within the target unit set, the pitting depth evolution sequences of neighboring target units were weighted and fused based on spatial distance to generate the pitting depth evolution sequence of the weak point. The remaining wall thickness was then calculated by subtracting the current maximum pitting depth from the average wall thickness of the neighboring units. Finally, the pitting depth evolution sequence was compared with a remaining wall thickness threshold to determine the moment when the pitting depth reached the remaining wall thickness threshold. The time difference between this moment and the current moment was taken as the physical time required for the weak point to develop into a fully penetrated state, and this time was used as the predicted value of the remaining perforation life of the waste acid reactor wall.

[0034] S502: Based on the predicted remaining life of the waste acid reactor wall perforation, the waste acid solution to be evaluated is identified, the toxicity index of characteristic pollutants is analyzed, a multi-dimensional environmental hazard assessment weighting function is constructed, environmental sensitivity weights are assigned to physicochemical indicators, the hazard value is calculated and mapped to the standard quantification range, and the environmental hazard equivalent of waste acid leakage is generated. Based on the predicted remaining life of perforation, the batch of waste acid solution to be assessed was identified, and the characteristic pollutant information within it was analyzed. The analysis included heavy metal content, concentration of highly corrosive ions, and proportion of volatile components. Subsequently, various pollutants were quantified, and the detected concentrations were mapped to grade scores according to preset ranges. The grade classification was determined based on historical accident monitoring data and environmental risk assessment standards. After grade quantification, environmental sensitivity weights were assigned to different pollutant grade scores, taking into account information on the surrounding environmental conditions of the plant area. Environmental condition information included the distance to the nearest surface water body, groundwater depth, frequency of prevailing wind direction, and soil permeability characteristics, all of which were converted into numerical weighting factors. Then, the grade scores of each pollutant and their corresponding weights were weighted and accumulated to obtain a comprehensive environmental hazard value. This value was then mapped to a unified standardized range to form the environmental hazard equivalent of the waste acid leak, providing a quantitative basis for subsequent risk assessment.

[0035] S503: By calling the predicted remaining life of the waste acid reactor wall perforation and the equivalent environmental hazard of waste acid leakage, a risk assessment decision matrix is ​​constructed. The two are mapped to the dimensions of time urgency and severity of consequences, respectively. The cross coordinate position is determined and compared with the boundary of the risk level area. The safety warning level of the current operating status is determined, and the pollution accident risk level evaluation result is generated. First, the predicted remaining lifespan of the waste acid reactor wall perforation and the environmental hazard equivalent of waste acid leakage are used to construct a risk assessment decision matrix. One dimension of the matrix represents the time urgency, divided into multiple intervals based on the remaining lifespan, with interval boundaries determined by historical maintenance preparation cycles and emergency response times. The other dimension represents the severity of consequences, divided into multiple levels based on the size of the environmental hazard equivalent. Then, the predicted remaining lifespan under the current operating condition is mapped to the corresponding time urgency interval, and the environmental hazard equivalent is mapped to the corresponding consequence severity interval. The risk level region is determined by the intersection of these two intervals. After mapping, the pollution accident risk level assessment result corresponding to the current operating state is output and stored along with the operating condition number, predicted lifespan, and hazard equivalent.

[0036] Please see Figure 7 An environmental risk assessment device for waste acid utilization facilities based on multi-source data includes: The data acquisition module is used to execute S1: by collecting data on voltage fluctuations on the inner wall of the waste acid reactor, hydrostatic pressure, output torque of the stirring motor, and geometric dimensions of the waste acid reactor wall through electrochemical monitoring probes and sensor groups arranged in the waste acid reactor of the waste acid utilization facility, and calling the least squares method to perform polynomial fitting detrending processing on the voltage fluctuation data on the inner wall of the waste acid reactor to generate a detrended voltage fluctuation sequence. The signal processing module is used to perform S2: statistically analyze the frequency of occurrence of the detrended voltage fluctuation sequence within the preset voltage amplitude range and calculate the distribution probability value, and calculate the electrochemical signal entropy characterizing the local corrosion activity of the waste acid reactor based on the distribution probability value; The stress and corrosion coupling calculation module is used to execute S3: using the electrochemical signal entropy as a weighting coefficient, combined with the hydrostatic pressure data and the output torque data of the stirring motor, to perform stress calculation and numerical coupling on the mesh model constructed based on the geometric dimensions of the waste acid reactor wall, and generate the corrosion-stress coupling field matrix. The pitting corrosion propagation simulation module is used to perform S4: extract the target element coupling numerical set that exceeds the critical damage threshold in the corrosion-stress coupling field matrix, input the target element coupling numerical set into the exponential growth model for depth evolution simulation, and calculate the pitting corrosion depth propagation rate. The risk assessment module is used to execute S5: acquire data on the physical and chemical properties of waste acid, calculate the remaining time for perforation of the weakest point of the waste acid reactor by combining the pitting depth propagation rate and the geometric dimensions of the waste acid reactor wall, and generate a pollution accident risk level assessment result based on the remaining time for perforation and the physical and chemical properties of waste acid.

[0037] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An environmental risk assessment method for waste acid utilization facilities based on multi-source data, characterized in that, Includes the following steps: S1: Collect data on voltage fluctuations, hydrostatic pressure, output torque of the stirring motor, and geometric dimensions of the reactor wall inside the waste acid reactor. The voltage fluctuation data is detrended by least squares to obtain a detrended voltage fluctuation sequence. S2: Perform frequency statistics on the detrended voltage fluctuation sequence within a preset voltage amplitude range, and normalize the frequency to obtain a voltage amplitude distribution probability value. Based on the voltage amplitude distribution probability value, calculate and accumulate the information content of the distribution probability value to obtain the electrochemical signal entropy. S3: Using the electrochemical signal entropy as a weighting coefficient, construct a mesh model based on the geometric dimensions of the waste acid reactor wall, and calculate the stress of the mesh model by combining the hydrostatic pressure data and the output torque data of the stirring motor to generate a corrosion-stress coupling field matrix. S4: Extract the set of coupled numerical values ​​of target units exceeding the preset critical damage threshold from the corrosion-stress coupled field matrix, and input them into the exponential growth model to obtain the pitting depth propagation rate. The parameters of the exponential growth model are determined by corrosion test data of specimens of the same material under the same physicochemical conditions as the waste acid to be evaluated. S5: Based on the pitting depth propagation rate and the geometry of the waste acid reactor wall, calculate the remaining time of perforation of the waste acid reactor wall, obtain the physicochemical property data of the waste acid corresponding to the waste acid to be evaluated, and assess the risk of pollution accident based on the remaining time of perforation and the physicochemical property data of the waste acid, and generate the pollution accident risk level evaluation result.

2. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The detrended voltage fluctuation sequence includes a time domain index, fluctuation amplitude values, and noise residual components. The electrochemical signal entropy includes information disorder, probability distribution density, and corrosion activity index. The corrosion-stress coupling field matrix includes grid node coordinates, stress intensity vector, and electrochemical weighting factor. The pitting depth propagation rate includes the increase in pitting depth per unit time in the vertical direction, the increase in pit diameter, and the increase in damage per unit time. The pollution accident risk level assessment results include the severity of leakage consequences, environmental sensitivity level, and emergency response priority.

3. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Start the electrochemical monitoring probe and sensor group in the waste acid reactor, collect the voltage fluctuation data of the inner wall of the waste acid reactor, the static pressure data of the fluid and the output torque data of the stirring motor, perform multi-channel signal data cleaning and time-domain synchronization processing, unify the time reference and data format of physical quantity measurements, and establish a multi-source operating condition monitoring dataset for the waste acid reactor. S102: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the voltage fluctuation data of the inner wall of the waste acid reactor, construct a polynomial regression model using the least squares method, perform global trend fitting operation on the voltage data time series, analyze the voltage signal reference trend and calculate the fitting curve parameters, and generate the voltage trend polynomial coefficient vector. S103: Construct a time-varying trend function for the voltage signal based on the voltage trend polynomial coefficient vector, call the original voltage fluctuation data of the multi-source operating condition monitoring dataset of the waste acid reactor, calculate the theoretical trend value at time, extract the global trend component from the original observation value and retain the local fluctuation residual to obtain the detrended voltage fluctuation sequence.

4. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Call the detrended voltage fluctuation sequence, divide the continuous voltage amplitude interval according to the preset resolution step size, traverse the sampling points and match the corresponding interval index, count the number of samples falling into the interval, construct an array structure that reflects the data density distribution, and generate a voltage interval frequency statistics vector. S202: Based on the frequency statistics vector of the voltage interval and the total sample capacity of the sampling points within the statistical period, normalize the frequency values ​​of the interval, verify the proportion weight of the voltage amplitude interval, complete the conversion from absolute quantity to probability, and generate a voltage amplitude distribution probability sequence. S203: Call the voltage amplitude distribution probability sequence, identify the probability elements and calculate the information components, perform summation and inversion operations on all interval components, quantify the disorder of the voltage fluctuation signal in the time domain distribution, and obtain the electrochemical signal entropy.

5. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Call the multi-source operating condition monitoring dataset of the waste acid reactor, extract the geometric dimension data of the waste acid reactor wall, perform discretization segmentation on the inner wall surface according to the preset spatial step size, construct a digital geometric structure including node coordinates and unit topological relationships, map the continuous entity into a virtual computing domain, and generate a three-dimensional discrete mesh model of the waste acid reactor wall. S302: Based on the three-dimensional discrete mesh model of the waste acid reactor wall, the hydrostatic pressure and stirring motor output torque data are called to map the physical load into the normal and tangential force components of the mesh element, perform multi-axial mechanical equilibrium calculation and solve the elastic deformation tensor of the element to generate dynamic mechanical load stress data. S303: Call the electrochemical signal entropy as the corrosion sensitivity weight, combine it with the dynamic mechanical load stress data, perform weighted coupling operation on the grid cells, map the corrosion activity to the stress strengthening coefficient, calculate the field strength value under the combined action of corrosion environment and mechanical load, and generate corrosion-stress coupling field matrix.

6. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Call the corrosion-stress coupling field matrix, set the critical damage threshold, traverse the coupling values ​​of matrix grid nodes and perform comparison operations, filter target grid elements whose values ​​exceed the critical damage threshold, extract the stress and corrosion coupling strength values ​​corresponding to the risk elements, construct a data subset including the index of damaged node locations, and generate a set of target element coupling values. S402: Based on the coupled numerical set of the target unit, construct an exponential growth model describing the nonlinear expansion of pitting depth, substitute the coupled numerical values ​​as initial driving parameters into the model, and deduce the theoretical pitting depth at a given time step according to a preset time step, forming a data set describing the dynamic increase of depth over time, and generating a pitting depth evolution sequence. S403: Call the pitting depth evolution sequence, calculate the incremental change in pitting depth at adjacent time points in the sequence, determine the slope of depth evolution per unit time, characterize the vertical downward erosion rate of the pitting pit, and obtain the pitting depth expansion rate.

7. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Obtain the physicochemical properties data of the waste acid, call the pitting depth propagation rate and the three-dimensional discrete mesh model of the waste acid reactor wall, retrieve the structural weak point with the smallest current wall thickness, perform numerical calculations on the remaining wall thickness parameters and pitting rate, deduce the physical time required for the weak point to develop to penetrating rupture, and generate the predicted value of the remaining life of the waste acid reactor wall perforation. S502: Based on the predicted remaining life of the waste acid reactor wall perforation, the waste acid solution object to be evaluated is identified, the toxicity index of characteristic pollutants is analyzed, a multidimensional environmental hazard assessment weighting function is constructed, environmental sensitivity weights are assigned to physicochemical indicators, the hazard value is calculated and mapped to the standard quantification range, and the environmental hazard equivalent of waste acid leakage is generated. S503: Call the predicted remaining life of the waste acid reactor wall perforation and the equivalent environmental hazard of the waste acid leakage to construct a risk assessment decision matrix, map the two to the dimensions of time urgency and severity of consequences respectively, determine the cross coordinate position and compare the risk level area boundary, determine the safety warning level of the current operating status, and generate the pollution accident risk level evaluation result.

8. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, wherein the preset voltage amplitude interval is a set of amplitude segments formed after determining the boundary of the amplitude statistical range of the detrended voltage fluctuation sequence, the amplitude statistical range is jointly defined by the maximum amplitude and the minimum amplitude of the detrended voltage fluctuation sequence, and the interval boundary of the preset voltage amplitude interval is determined by the amplitude statistical range according to the equal width division rule.

9. The environmental risk assessment method for waste acid utilization facilities based on multi-source data according to claim 1, characterized in that, The electrochemical signal entropy is calculated based on the interval probability distribution and in combination with the interval boundary consistency condition of the preset voltage amplitude interval; The electrochemical signal entropy is normalized before being used as the weighting coefficient, so that the weighting coefficient is obtained by monotonically mapping the electrochemical signal entropy and corresponds to the interval probability distribution. The critical damage threshold is determined based on the corrosion-stress coupling strength corresponding to the maintenance data or wall thickness detection data.

10. An environmental risk assessment device for waste acid utilization facilities based on multi-source data, characterized in that, The system is used to implement the environmental risk assessment method for waste acid utilization facilities based on multi-source data as described in any one of claims 1-9, and the system includes: The data acquisition module is used to execute S1: by collecting data on voltage fluctuations on the inner wall of the waste acid reactor, hydrostatic pressure, output torque of the stirring motor, and geometric dimensions of the waste acid reactor wall through electrochemical monitoring probes and sensor groups arranged in the waste acid reactor of the waste acid utilization facility, and calling the least squares method to perform polynomial fitting detrending processing on the voltage fluctuation data on the inner wall of the waste acid reactor to generate a detrended voltage fluctuation sequence. The signal processing module is used to perform S2: statistically analyze the frequency of occurrence of the detrended voltage fluctuation sequence within the preset voltage amplitude range and calculate the distribution probability value, and calculate the electrochemical signal entropy based on the distribution probability value; The stress and corrosion coupling calculation module is used to execute S3: using the electrochemical signal entropy as a weighting coefficient, combined with the hydrostatic pressure data and the output torque data of the stirring motor, to perform stress calculation and numerical coupling on the mesh model constructed based on the geometric dimensions of the waste acid reactor wall, and generate a corrosion-stress coupling field matrix. The pitting corrosion propagation simulation module is used to perform S4: extract the target element coupling numerical set that exceeds the critical damage threshold in the corrosion-stress coupling field matrix, input the target element coupling numerical set into the exponential growth model for depth evolution simulation, and calculate the pitting corrosion depth propagation rate. The risk assessment module is used to perform S5: acquire the physicochemical properties data of the waste acid, calculate the remaining time for the perforation of the weakest point of the waste acid reactor by combining the pitting depth propagation rate and the geometric dimensions of the waste acid reactor wall, and generate a pollution accident risk level assessment result based on the remaining time for perforation and the physicochemical properties data of the waste acid.

Images