341results about How to "Improve fitting accuracy" patented technology

The invention discloses a power load model parameter identification method based on improved deep reinforcement learning, and belongs to the technical field of power load identification. The method comprises the following steps: determining a to-be-identified power load parameter based on a complete comprehensive load model considering power distribution network impedance and reactive compensation; then designing a power load parameter identification model based on a bidirectional gating cycle unit depth deterministic strategy gradient algorithm of priority experience playback; circuit data are collected for model training, and the trained model is used for parameter identification. On the basis of an existing power system mechanism model and a data driving model, the load model parameters are identified by improving the deep reinforcement learning model, and verification shows that the method can accurately identify the power system load model parameters, can improve the fitting precision of the power system load model, and has a good application prospect. The operation strategy of the power grid is optimized, and the operation cost is reduced.

The invention discloses a high-speed rail system sliding layer peak displacement envelope curve and rail direction irregularity prediction method considering seismic oscillation and pier height uncertainty and application. The method comprises the following steps: constructing a multi-mode hybrid neural network model TransTCN embedded with field priori knowledge; feature extraction is carried out on multi-modal input features of a time sequence based on seismic oscillation fusion structure features and a space sequence for outputting peak displacement based on an elastic response spectrum method; the pier height features and the seismic oscillation features are subjected to dimension fusion to serve as time sequence features, and a time dependency relationship is processed; processing a spatial dependency relationship by taking an elastic response spectrum analysis result as a spatial feature; and jointly inputting the extracted features into a constructed multi-mode hybrid neural network model TransTCN, and realizing prediction of the peak displacement envelope curve and rail direction irregularity of the sliding layer of the high-speed rail system under the consideration of seismic oscillation and pier height uncertainty. The prediction precision is improved, the training sample demand is reduced, the regional high-speed rail system damage is rapidly evaluated, and the practical value is good.

The invention discloses a lithium battery state estimation method based on relaxation time scale modeling, and the method comprises the steps: 1, carrying out a sudden pulse charging test on a lithium battery, collecting voltage and current data, and calculating the open-circuit voltage, direct-current internal resistance and a time-domain impedance sequence of the lithium battery; and 2, converting the time domain impedance sequence of the lithium battery into a frequency domain impedance sequence after discrete Fourier transform, establishing a fitting equation set of a distribution function of the relaxation time scale DRT of the lithium battery and the frequency domain impedance sequence, and solving by adopting a regularized least square method. And 3, taking the solved DRT model distribution function as the input of a neural network, taking the state parameters of the lithium battery as the output of the neural network, and establishing a prediction model of the state parameters of the lithium battery. According to the method, EIS test required by traditional relaxation time scale modeling is avoided, the control complexity and design cost of test acquisition equipment are effectively reduced, and the method can be effectively applied to occasions such as battery life prediction, battery classification and screening, battery echelon utilization and the like.

The invention discloses a fixed-wing unmanned aerial vehicle predetermined time cooperative path following formation control method, and relates to an unmanned aerial vehicle technology. A multi-unmanned aerial vehicle cooperative path following control strategy based on preset time stability, neural network adaptive control and preset performance control is provided; according to the method, the multiple fixed-wing unmanned aerial vehicles can realize preset time convergence of path following errors and preset time preset performance convergence of control signal tracking errors under the conditions of external disturbance and actuator faults, so that a high-precision and quick-response cooperative path following formation control task is completed; and meanwhile, the control precision and robustness are improved.

The invention discloses a polar equipment fatigue crack propagation reliability analysis method based on an improved SVM (Support Vector Machine), which comprises the following steps: acquiring uncertain parameters influencing the fatigue crack propagation of a polar equipment structure, and generating a candidate sample population by adopting a random sampling method according to the uncertain parameters of fatigue performance; generating an initial sample by adopting an orthogonal test method based on the fatigue performance uncertain parameters; evaluating the real cycle index response of the test sample; performing normalization processing on the obtained candidate sample, the test sample, the initial training sample and the cycle index response set; constructing an IBOA-LS-SVM model based on the initial sample, and obtaining a cycle index prediction value of the test sample and the candidate sample set; carrying out reverse normalization processing on the predicted value of the cycle index of the test sample point and the predicted value of the cycle index of the candidate sample point, carrying out precision test on the IBOA-LS-SVM model by adopting the true value and the predicted value of the cycle index of the test sample, and if the precision meets the requirement, calculating the failure probability of fatigue crack propagation at different polar region service environment temperatures by utilizing an MCS method; and if the precision does not meet the condition, carrying out orthogonal test design again.

The invention discloses a mobile terminal multi-size webpage adaptive display method based on a browser viewport, which belongs to the field of browsers and comprises the following steps of: firstly, selecting iPhoneX (375px width) as a reference viewport size; and then, calculating a zoom factor according to the proportion of the viewport width of the current equipment to the reference viewport width, and dynamically adjusting the webpage content. And then, zooming the page root element by using the zooming factor. In order to adapt to screen widths and resolutions of different devices, breakpoints are set, a compact layout is applied to a narrower device, and more blanks and spaces are reserved on a wider device. And finally, recalculating the scaling in real time and updating the page. The method has the beneficial effects that a finer self-adaptive design is realized by dynamically calculating the zoom factor, adjusting the page layout in real time and optimizing display according to the breakpoint of the equipment.

The invention discloses a foundation pile horizontal deformation reliability calculation method and device and electronic equipment, and belongs to the technical field of pile foundation engineering. Factors influencing horizontal deformation of the foundation pile and a simulation calculation value range of the factors are determined; designing a foundation pile horizontal deformation three-dimensional finite element numerical simulation calculation scheme based on a uniform test design method; calculating a horizontal deformation value of the foundation pile based on three-dimensional finite element numerical simulation; establishing a training sample data set and a test sample data set; establishing a foundation pile horizontal deformation agent model based on a machine learning model; establishing a limit state equation for calculating the reliability of the horizontal deformation of the foundation pile; and calculating a foundation pile horizontal deformation failure probability and a reliability index. The method has the advantages of being good in fault tolerance, high in adaptability and high in fitting and prediction precision, and the technical problem that a high-precision foundation pile deformation agent model is difficult to construct due to too many input variables is solved.

The invention belongs to the technical field of point cloud data processing, and discloses a point cloud smooth fitting method based on local density parameter adjustment and intensity optimization, and the method comprises the steps: obtaining original point cloud data, carrying out the preprocessing, carrying out the local density calculation, carrying out the dynamic adjustment of a search radius strategy, and carrying out the dynamic extension of a neighborhood, and after boundary judgment, performing boundary protection mechanism or joint weight calculation and updating smooth fitting on boundary points to obtain a point cloud visualization result graph after fitting reconstruction. The method effectively overcomes the defects of sparse point cloud smoothing, noise robustness, intensity feature retention, boundary feature fuzziness and the like in the prior art, can dynamically adapt to sparsity and non-uniform distribution of the point cloud, and remarkably improves the precision of smooth fitting of the point cloud and the optimization capability of intensity information. A more accurate and efficient technical scheme is provided for topographic surveying and mapping, underwater target detection and environment modeling of the forward-looking three-dimensional sonar.

The invention provides a method for improving the temperature compensation precision of a pressure sensor, and the method comprises the steps: obtaining the output-pressure-temperature data of a pressure sensor sample through an experiment by employing calibrated high-precision equipment, carrying out the polynomial fitting based on a least square method, calculating the coefficient of a multi-term expression, optimizing the polynomial times of the multi-term expression, and obtaining the temperature compensation precision of the pressure sensor. And the polynomial piecewise function is combined to obtain the highest fitting precision, the function piecewise of the polynomial piecewise function is optimized, and the optimized polynomials are combined into the polynomial piecewise function to serve as a temperature compensation function for improving the temperature compensation precision of the pressure sensor. According to the method provided by the invention, the temperature compensation precision can be further improved, the fitting precision can be effectively improved, the fitting mean square error can be improved from 4.993 e-05 to 1.013 e-05, and the precision is improved by about 80%.

The invention discloses a Raman spectrum fluorescence baseline correction method and device, and the method comprises the steps: collecting a time domain signal of a to-be-detected sample, and obtaining an IRF function of a to-be-detected sample collection system; pre-processing the time domain signal, and carrying out deconvolution on the pre-processed time domain signal and an IRF function to obtain a deconvolution result; fitting the second half data of the deconvolution result by using a support vector regression model to obtain a fitted curve; wherein a radial basis function is adopted as a kernel function, and hyper-parameters of the support vector regression model are optimized through a grid search and cross validation method; performing difference operation on the deconvolution result and the fitting curve to obtain a Raman attenuation curve; and windowing the Raman attenuation curve and averaging to obtain a Raman spectrum. According to the method, a fluorescence attenuation model and parameters thereof do not need to be preset, the IRF deconvolution result is processed and the time domain baseline is predicted by utilizing the characteristics of the time domain signal data and the support vector regression model, the signal-to-noise ratio of the extracted Raman signal can be greatly improved, a good effect is achieved, and the method is suitable for practical application.

The invention discloses a chamfer defect detection method, device and equipment and a storage medium, and relates to the technical field of defect detection.The method comprises the steps that an obtained to-be-detected device image is preprocessed, and chamfer contour points and chamfer contour point pixel coordinates are determined in the preprocessed image; correcting the pixel coordinates of the chamfer contour points, and fitting a chamfer edge straight line by using the corrected pixel coordinates of the chamfer contour points; according to the chamfer edge straight line, screening the corrected chamfer contour point pixel coordinates, and determining a target chamfer end point; and calculating the actual chamfer length based on the target chamfer endpoint, and comparing the actual chamfer length with the preset chamfer length to judge whether the chamfer has defects or not. According to the method, the detection requirements of chamfers of different shapes and sizes can be met by preprocessing and correcting the images, fatigue and subjective judgment errors possibly existing in traditional manual detection are avoided through automatic image processing and data analysis processes, and the detection efficiency is improved.

A straight line fitting method for image detection in a complex interference environment comprises the following steps: firstly, carrying out interference point deletion on a feature point set of an environment image, then carrying out multiple times of random sampling, forming a total point set by using associated points through distance operation, obtaining a plurality of straight line models through initial fitting, and carrying out linear fitting on the linear models; and then calculating an interior point proportion in the total point set according to the linear model, finally selecting the total point set with the maximum number of interior points, and continuously reducing a residual threshold in multiple iterative calculations to improve an interior point proportion threshold by dynamically adjusting a feedback mechanism of the residual threshold when the interior points are calculated. In the initial stage of iteration, more points can enter the model to ensure the diversity and robustness of the model, and the threshold value is gradually tightened along with the convergence of the model, so that the interior points conforming to the model are screened out more accurately. And finally, performing secondary sampling on the inner points and performing straight line fitting on an inner point set subjected to secondary sampling again to eliminate potential error inner points and further improve the fitting precision.

The invention discloses a clean type linear motor module, and particularly relates to the technical field of linear motors, the clean type linear motor module comprises a motor base and a sliding seat, two sides of the motor base are provided with guide rail grooves, two sides of the sliding seat are fixedly provided with side plates, the inner sides of the guide rail grooves are provided with guide rod assemblies, and the inner sides of the side plates are provided with guide rod assemblies extending into the guide rail grooves. The guide rod assembly comprises a lower guide rod and an upper guide rod, the lower guide rod is installed in the bottom wall of the guide rail groove in a sliding mode, and a height adjusting structure is arranged at the bottom of the lower guide rod. The guide structure is arranged on the outer side of the motor base, cannot be interfered by the working temperature of the inner cavity of the motor base, can keep a good working state, cannot interfere with the upper guide rod to cause abrasion even if the module is influenced by the external temperature and the guide wheel expands, and can effectively ensure the working safety of the module; the service life of the module is prolonged.

The invention relates to the technical field of sound insulation material proportioning parameter regulation and control, in particular to a sound insulation material proportioning parameter adaptability regulation and control method and device based on environmental characteristics, and the method comprises the steps: constructing an initial sound intensity fitting curve according to an initial center sound intensity set, carrying out the fitting precision verification of the initial sound intensity fitting curve, and if the precision verification value is not greater than a fitting precision threshold value, determining that the sound insulation material proportioning parameter is not greater than the fitting precision threshold value; if the precision check value is greater than a fitting precision threshold value, constructing an iteration center sound intensity set according to the initial center sound intensity set and the supplementary center sound intensity set, and reconstructing an initial sound intensity fitting curve, if the precision check value is greater than the fitting precision threshold value, performing sound intensity sound insulation calculation to obtain a target sound insulation curve set, and extracting an optimal sound insulation curve from the target sound insulation curve set, according to the optimal material matching parameters, sound insulation material matching parameters are regulated and controlled. The problems that an existing sound insulation material is low in noise reduction adaptability and poor in sound insulation effect can be solved.

The invention relates to the technical field of computer aided design, and discloses an artificial intelligence fitting method and device for device parameters, electronic equipment, a computer readable storage medium and a computer program product. The method is used for solving the technical problems that in the prior art, when simulation calculation is conducted on a semiconductor device, fitting precision is insufficient, steps are complex, and time cost is high. Comprising the following steps: constructing a device simulation case and obtaining a first input parameter and a first output parameter; determining the influence type of the first input parameter on the first output parameter according to the device type of the semiconductor device, and generating a mechanism adjustment function of the device simulation case according to the influence type; correcting the initial artificial intelligence model based on a mechanism adjustment function to obtain a corrected artificial intelligence fitting model, and fitting the device simulation case to obtain a first input parameter combination enabling the first output parameter to conform to the experimental output parameter; and constructing a correlation model between the second input parameter and the second output parameter of the device simulation case.

The invention discloses an analysis and detection method and device based on multi-exposure and multi-brightness double parameters, and belongs to the field of immunochromatography analysis and detection. The analysis and detection method comprises the steps of obtaining a multi-exposure multi-brightness image set under a plurality of multi-exposure time sequences and different illumination intensities, and in the process of restoring a camera response function of an image acquisition system, using an HDR algorithm based on multi-exposure and multi-brightness fusion. The weights of images with different exposures and different current values are adjusted by adopting a segmented weighting function, the characteristics of each concentration section in immunochromatography are reflected more finely by finely regulating and controlling parameters of the segmented weighting function, and the optimal weighting function is established to enhance the fitting accuracy of a response function of a two-parameter camera, so that the accuracy of the response function fitting is improved. Therefore, the immunochromatography HDR image with the highest quality is recovered, and the linear range of quantitative detection is widened.

The present invention discloses a friction compensation control method for an electro-hydraulic servo system. The method first collects signals such as system flow, oil supply pressure, acceleration, etc. of the electro-hydraulic servo system, models the system friction using an improved LuGre friction model, identifies its parameters offline using a genetic algorithm, and uses the estimated value generated by the identification model for feedforward compensation; then, a hyperbolic sine function with better smoothness is introduced to design the nonlinear state error feedback and extended state observer part of the variable coefficient self-disturbance rejection controller; and on this basis, a fuzzy adaptive control method is combined to adjust the nonlinear state feedback coefficient through error and error differential; the "contradiction between rapidity and overshoot" is solved, and the stability and finite time convergence of the system are guaranteed. This method effectively improves the system's compensation capability for friction nonlinearity, improves the system's motion performance, and improves the system's tracking accuracy and anti-disturbance capability.

The invention discloses a turbulence model for underwater wireless optical communication and a communication system performance calculation method, and the method comprises the steps: building a new turbulence statistical distribution model under the condition of ocean turbulence, and obtaining an expression of a flicker index of the model in combination with model parameters; and a Meijer-G function and a Fox-H function are utilized to obtain a closed expression of the outage probability and the bit error rate under the model condition, so that the influence on the transmission performance of the system under different scene parameters can be researched. Compared with an existing turbulence model, the method is closer to the complex turbulence environment of the ocean, and especially under the complex turbulence scene of bimodal distribution, simulation research on the real ocean environment can be better achieved.

The invention provides a parameter-limited Monte Carlo method for determining a solar proton event energy spectrum, which comprises the following steps of: S1, acquiring a proton time-varying flux of each energy channel of a space proton detector, and judging the beginning and the end of a solar proton event; s2, calculating the total flux of protons in each energy channel after the background in the solar activity calm period is removed during the occurrence period of the solar proton event; s3, performing logarithmic transformation on parameters of the proton double-power-law energy spectrum function according to the total flux obtained in the S2 to determine a sampling range, sampling the parameters in the determined sampling range by using a uniform and random Monte Carlo method, and fitting the energy spectrum by using an integral method; and S4, obtaining a group of optimal parameters of the proton continuous energy spectrum, and drawing an energy spectrum curve in combination with an energy spectrum function. According to the Monte Carlo method provided by the invention, the calculation efficiency is improved in the fitting process, and the fitting precision is improved. By using the integral method, the calibration requirement on the satellite observation energy channel and energy section is avoided, and the problem of error accumulation caused by calibration is solved.

The invention provides a heat source tower heat pump applicability evaluation model construction and applicability evaluation method and system. The heat source tower heat pump applicability evaluation model construction method comprises the steps that historical state data of a heat source tower heat pump, all index values of the historical state data under a preset applicability evaluation index system and real applicability evaluation results corresponding to all the index values are obtained; taking each index value of the historical state data under the applicability evaluation index system as input data of training data; taking a real applicability evaluation result corresponding to each index value as output data of the training data; based on the input data and the output data, training a neural network model by using a particle swarm algorithm to obtain an applicability evaluation model; the constructed applicability evaluation model is based on the applicability evaluation index system, the subjectivity of manual judgment can be reduced, the neural network model is trained by adopting the particle swarm optimization, the fitting precision of the applicability evaluation model can be improved, and the accuracy of model prediction can be improved.

The invention relates to the technical field of numerical control system information processing, in particular to a curved surface numerical control machining information extraction method and device based on B-spline curved surface refactor.The method comprises the steps that after three-dimensional space position information of a curved surface machining cutter location point is obtained, cutter location point parameterization is carried out to determine B-spline curved surface parameters of a cutter positioning curved surface; determining a precision index according to the sum of squares of deviations of all cutter location points and the cutter positioning curved surface to calculate a regression matrix, determining a fairing energy index to calculate a fairing energy matrix, combining the two indexes to determine a comprehensive index, generating a curved surface reconstruction problem, solving the problem to obtain a reconstruction structure, and extracting curved surface numerical control machining information. Therefore, the problem that in the related technology, when a curved surface reconstruction algorithm processes disordered spatial discrete cutter location points, the requirements for rapidity, fairness and precision are generally difficult to consider is solved.

The invention discloses a novel power system output regulation and control strategy generation method and device and a control system, and belongs to the technical field of power system regulation and control. The output regulation and control strategy generation method comprises the following steps: representing net load historical data corresponding to net load climbing demand uncertainty characteristics by adopting a Gaussian mixture model; the uncertainty of the net load climbing requirement is reduced; inputting the net load day-ahead prediction data corresponding to the net load historical data and the net load uncertainty scene set into a two-stage risk avoidance model based on the lowest operation cost and the worst condition risk probability assessment theory; the two-stage risk avoidance model divides the whole strategy generation process into day-ahead optimization scheduling plan generation and intra-day real-time scheduling plan generation; the day-ahead optimal scheduling plan is optimized by using the obtained intra-day real-time scheduling plan to obtain a safe and economic output scheduling strategy; and furthermore, the novel power system is controlled in different scenes to realize efficient, stable and economical operation.

The invention provides a magnetic element magnetic core loss prediction method and device, equipment and a storage medium. The method comprises the steps of obtaining a corresponding real magnetic core loss value when a magnetic element works in a working temperature range; the working temperature range of the magnetic core of the magnetic element is partitioned, and a piecewise function representing magnetic core loss in each temperature interval is constructed; constructing a magnetic core loss model based on the piecewise function; estimating parameters of the magnetic core loss model by taking the minimum error sum of squares between the predicted value of the magnetic core loss and the true value of the magnetic core loss as a target to obtain a prediction model of the magnetic core loss; and realizing prediction of the magnetic core loss of the magnetic element based on the magnetic core loss prediction model. By using a piecewise function, the temperature is converted into a qualitative variable, and the influence of the qualitative variable on the magnetic core loss is divided into different intervals. A magnetic core loss model is constructed based on a piecewise function, and the model can better capture local changes of experimental data in different temperature ranges, so that the fitting precision is improved.

The invention relates to a deformation measuring device and method capable of deforming along with the rear edge of a variable-camber wing, and belongs to the field of wing deformation measurement. The key component of the deformation measuring device is a deformation beam provided with a fiber grating sensor, the deformation beam is connected with the wing through a sliding support, a positioning pin and other structures, according to the deformation characteristics of the rear edge of the variable camber wing, a sliding connection mode is adopted in a large deformation area, and a fixed connection mode is adopted in the rest area. The deformation beam can deform along with a wing after being connected with the wing structure, and the length change of a sliding connection area can be measured through a stay wire displacement sensor. The problems that in an existing wing deformation measurement technology, large deformation generated by a flexible structure is difficult to directly measure, large deformation measurement and small deformation measurement are difficult to consider, errors exist in the sensor manufacturing and installing process and the like can be solved, and efficient and accurate wing trailing edge profile measurement is achieved.

The utility model discloses a hot forging die for forging two parts, which relates to the technical field of forging dies and is used for solving the problems of forging die thermal mechanical damage and poorer utilization effect of excess materials of forgings caused by demolding damping of the forgings. According to the utility model, the supporting auxiliary die sleeve, the lower die base body and the lower die piece are mainly arranged, so that the total amount of contact heat enrichment of the forging die is reduced, and the thermal mechanical damage of demolding damping generated by forging to the forging die is reduced. By arranging the first separation piston rod, the lower core die, the separation cavity and the lower die piece, the bearing pressure of the forge piece can be discretely distributed, the utilization input cost of excess materials of the forge piece is reduced, and the possibility of thermal mechanical damage of the forging die is reduced. By arranging the lower core die, the lower die piece and the wad die sleeve, the forging quality and the machining manufacturability of forging forming are improved, and the thermal contact temperature peak value of the forging die is reduced.

The invention provides a multi-feature fusion-based black box attack detection method (MFBA), device and system, different features are extracted by using a multi-model fusion strategy, the effectiveness and quality of opposite samples are improved, and the MFBA comprises the following three key steps: firstly, obtaining an initial data set from a target model; a decision boundary is approached by using a multi-feature fusion model; secondly, the stability of the model is enhanced through multiple rounds of training application data expansion and incremental learning technologies; and finally, generating opposite examples by using the optimized fusion model. By constructing the comprehensive attack framework, how to combine various model features to improve the performance against the resistance attack is demonstrated; compared with a single substitution model, the antagonism examples generated by the MFBA have higher mobility under different disturbance levels and have better attack effects; according to the method, an effective solution is provided for black box antagonism attacks, and the model in the generation of the advantage opposite paradigm of multi-feature fusion is verified.

The invention discloses a multi-channel sampling circuit hybrid error modeling and cyclic calibration method, and belongs to the technical field of signal sampling. The method comprises the following steps: constructing a system error model, and introducing polynomial series to realize model simplification; generating a reference signal participating in error cycle calibration according to the simplified hybrid error model; initializing a frequency response function of each sampling channel, and calculating an initial tap coefficient of an error correction module; a cascade multi-order error correction structure is used for completing reference signal error iteration correction, so that the reference signal error continuously approaches to an input signal; correcting the signal to calculate a loss function, and completing estimation and updating of a tap coefficient; and judging whether the loss function reaches a convergence condition or not until hybrid error cyclic calibration is realized. On the basis of the reference signal constructed by the hybrid error model, low-complexity online blind cycle calibration is realized, the precision and completeness of the system error model are improved, the quantity of parameters to be estimated and the number of sample points for calibration are remarkably reduced, and the calibration efficiency is improved.

The invention relates to the technical field of remote sensing, and discloses a forestry loss value evaluation method, device and equipment and a storage medium. A second time sequence image data set and a first fitting model are combined to calculate a first difference image; and the change condition of the vegetation in the cloud-free area of the remote sensing image after typhoon disaster relative to the non-disaster state can be intuitively and quantitatively displayed. Furthermore, model training is carried out by using the environmental influence factor set and the first difference image, and a deep relationship rule of the influence of the typhoon disaster hidden behind the complex data on the forest can be mined. Furthermore, the normalized vegetation index difference value of the cloud coverage area is predicted by using the constructed model to generate a second difference value image, so that the spatial representation information of the disaster condition of the whole typhoon disaster area can be perfected. And finally, in combination with the real forestry loss value, the first difference image and the second difference image, rapid evaluation of the forestry loss value in the whole typhoon affected area can be realized.

The utility model provides a single-station air bag chip mounter. The single-station air bag chip mounter comprises a machine body, an operation panel, a heating carrier seat, a lifting driving device, a mounting seat, a rubber mat, a soft air pipe and an air pump, the operation panel is installed on the machine body and connected with the heating carrier, the lifting driving device and the air pump through electric signals. The heating carrier is fixed at the lower part of the machine body; the lifting driving device is mounted at the upper part of the machine body and a power output end is connected with the mounting seat; the mounting seat is positioned above the heating carrier seat; a downward opening is formed in the mounting seat; the outer periphery of the rubber mat is fixed on the mounting seat; an air chamber is formed between the rubber mat and the mounting seat; an air pipe connector communicated with the air chamber is arranged above the mounting seat; and the air pipe interface is connected with an air pump through a flexible air pipe. According to the design, the wafer can be tightly attached to the ceramic plate.

The invention relates to a battery spare part prediction method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring various data of a plurality of in-use battery packs to obtain total data; wherein the plurality of in-use battery packs are arranged in different electric equipment; determining a failure change relationship according to the total data; the failure change relationship comprises a relationship that the failure probability of the battery pack changes along with the first survival duration; according to the failure change relationship, predicting the failure number of the plurality of in-use battery packs in a preset future time period; and determining the number of battery pack spare parts required in the preset future time period according to the failure number. According to the invention, the accuracy of battery spare part prediction can be improved, so that spare parts can be better supplied.