CN114420586A

CN114420586A - Parameter anomaly detection method and semiconductor process equipment

Info

Publication number: CN114420586A
Application number: CN202111583182.6A
Authority: CN
Inventors: 杨浩; 王馨梦; 任志豪; 申震
Original assignee: Xi'an North Huachuang Microelectronic Equipment Co ltd; Beijing Naura Microelectronics Equipment Co Ltd
Current assignee: Xi'an North Huachuang Microelectronic Equipment Co ltd; Beijing Naura Microelectronics Equipment Co Ltd
Priority date: 2021-12-22
Filing date: 2021-12-22
Publication date: 2022-04-29

Abstract

The embodiment of the invention provides a parameter anomaly detection method and semiconductor process equipment, wherein the process of the semiconductor process equipment comprises a plurality of process steps which are sequentially carried out, and the parameter anomaly detection method comprises the following steps: acquiring actual parameter values of target parameters at preset time points in the process step; dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter changing along with the preset time precision at the preset time point and the normal parameter value of the corresponding preset time point in the parameter relation graph meet preset abnormal conditions, wherein the parameter relation graph is a relation graph of the normal parameter value of the target parameter changing along with each preset time point in the process step; adjusting the process according to an adjustment scheme corresponding to a preset abnormal condition; and outputting prompt information corresponding to the preset abnormal conditions. By adopting the parameter abnormity detection method, the parameter abnormity detection efficiency can be improved.

Description

Parameter anomaly detection method and semiconductor process equipment

Technical Field

The invention relates to the technical field of semiconductors, in particular to a parameter abnormality detection method and semiconductor process equipment.

Background

The process is also called "processing flow" or "production flow". The whole process from raw material input to finished product output is carried out sequentially and continuously through certain production equipment or pipelines. A complete process usually includes several steps. In the process flow related to the manufacturing of process equipment, the fluctuation of some target parameters (such as temperature, flow, radio frequency, pressure and the like) directly affects the process result, and the rework and the equipment capacity are seriously affected.

The existing parameter anomaly detection scheme has the advantages that the detection of the anomaly parameters is simple and solidified, the anomaly condition is often found from the produced process result after the process is finished, the detection efficiency is not high, and the ever-increasing production requirement cannot be met.

Disclosure of Invention

The technical problem to be solved by the embodiment of the invention is that the abnormal parameters are not detected in time at present and the efficiency is not high.

In order to solve the above problems, an embodiment of the present invention discloses a parameter anomaly detection method, which is applied to semiconductor process equipment, wherein a process of the semiconductor process equipment includes a plurality of process steps performed in sequence, and the method includes:

acquiring actual parameter values of target parameters at preset time points in the process step;

dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter changing along with the preset time precision at the preset time point and the normal parameter value of the corresponding preset time point in the parameter relation graph meet preset abnormal conditions, wherein the parameter relation graph is a relation graph of the normal parameter value of the target parameter changing along with each preset time point in the process step;

adjusting the process according to an adjustment scheme corresponding to a preset abnormal condition;

and outputting prompt information corresponding to the preset abnormal conditions.

Another embodiment of the present invention discloses a semiconductor process apparatus, wherein a process of the semiconductor process apparatus includes a plurality of process steps performed in sequence, and the semiconductor process apparatus includes:

the controller is used for acquiring actual parameter values of the target parameters at all preset time points in the process step;

According to the semiconductor process equipment provided by the invention, the actual parameter values of the target parameters at all preset time points in the process steps are collected; dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter at a preset time point, which changes along with the preset time precision, and the normal parameter value of the corresponding preset time point in the parameter relation graph meet a preset abnormal condition, wherein whether the parameter is abnormal in the process step can be accurately and quickly detected by comparing the actual parameter value of the target parameter in the process step with the normal parameter value of the target parameter in the process step; therefore, the corresponding adjusting scheme can be triggered quickly and efficiently to adjust the process progress and output the prompt information corresponding to the preset abnormal condition, and therefore, the productivity of the equipment can be improved, the capacity of the equipment for processing the abnormality can be improved, and the safety is improved by improving the detection speed of the semiconductor process equipment on the parameter abnormality.

Drawings

FIG. 1 is a flow chart of a process provided in the present example;

FIG. 2 is a flow chart of a parameter anomaly detection method provided in this embodiment;

FIG. 3 is a schematic diagram illustrating a parameter relationship diagram and parameter values provided in this embodiment;

FIG. 4 is a diagram illustrating a parameter exception provided by the present embodiment;

FIG. 5 is a schematic diagram illustrating another parameter anomaly provided by the present embodiment;

FIG. 6 is a schematic diagram illustrating an exception of another parameter provided in the present embodiment;

FIG. 7 is a schematic diagram illustrating a frequency spectrum curve of an abnormal parameter provided in the present embodiment;

FIG. 8 is a diagram illustrating a spectrum curve of a normal parameter provided by the present embodiment;

FIG. 9 is a flowchart illustrating a method for detecting parameter anomalies according to this embodiment;

FIG. 10 is a flow chart illustrating another method for implementing parameter anomaly detection according to the present embodiment;

fig. 11 shows a schematic structural diagram of a semiconductor processing apparatus provided in this embodiment.

Detailed Description

Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

First, technical terms related to the embodiments of the present invention are described.

The solar cell is a photoelectric semiconductor sheet which directly generates electricity by using sunlight, is also called as a solar chip or a photovoltaic cell, and can output voltage instantly and generate current under the condition of a loop as long as the solar cell is illuminated under a certain illumination condition.

The most important two directions of solar cell research are high efficiency and low cost, and in the cell production process, because the graphite boat loading quantity of the coating equipment is far less than that of the diffusion and annealing equipment, and the capacity bottleneck of a workshop is always the coating equipment, the reduction of the process failure boat and the reduction of reworked wafers become important research points for improving the production capacity of the coating equipment.

The parameter anomaly detection method provided by the embodiment of the invention can be at least applied to the following application scenarios, which are explained below.

As shown in figure 1, the method relates to a deposition step process part in a coating process section in the process of manufacturing a crystalline silicon solar cell, and target parameters of temperature, flow, radio frequency and pressure in the deposition step directly influence the film thickness of a coated silicon wafer. Changes in both hardware and environment cause fluctuations in the parameter values of these four parameters, such as: the connection line of the radio frequency power supply is connected in a virtual mode, the pumping speed of the pump is reduced after long-term use, the flow of a factory building is fluctuated, the temperature of a cavity is fluctuated due to radio frequency discharge, and the like, or the set value of a process skipping parameter is changed.

The fluctuation of the parameter value has two conditions, one is instantaneous jump, if the occurrence frequency is less, the filtering can be directly carried out, and when the frequency is more, the process result can be influenced; the other is continuous fluctuation, the fluctuation is within a certain range or has small influence on the process result, when the fluctuation range is large or the fluctuation duration is long, the process result is influenced, the silicon wafer reworking is seriously caused, and the equipment productivity is influenced.

Currently, in the process of semiconductor equipment, only one parameter fluctuation reference value exists in each process step, and the reference value does not change along with the process time of the process step. By comparing the actual parameter value with the reference parameter value, timing is started when the fluctuation of the actual parameter value relative to the reference parameter value is large, and alarm processing is performed when the fluctuation is large for a certain time. And if the fluctuation returns to normal within a certain time, clearing the timing, and restarting the timing after waiting for the next larger fluctuation. However, each process step in the process only has one reference value, and in a plurality of process steps in the process, the tolerable fluctuation ranges of various conditions at various times are different, so that the prior art cannot meet the judgment of the conditions; moreover, only continuous abnormal fluctuation can be detected, and other fluctuation influencing the process result cannot be detected and judged; in addition, the judgment process cannot realize early warning on the abnormality of the equipment hardware, and the judgment can be realized only when the abnormality really occurs.

Based on the above problems and application scenarios, the parameter anomaly detection method provided by the embodiment of the present invention is described in detail below.

The embodiment of the invention provides a parameter abnormality detection method which is applied to semiconductor process equipment, wherein the process of the semiconductor process equipment comprises a plurality of process steps which are sequentially carried out.

Fig. 2 is a flowchart of a parameter anomaly detection method according to an embodiment of the present invention.

As shown in fig. 2, the parameter anomaly detection method may include steps 210 to 240, which are specifically as follows:

and step 210, acquiring actual parameter values of the target parameters at each preset time point in the process step.

Step 220, dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter at the preset time point, which changes along with the preset time precision, and the normal parameter value of the corresponding preset time point in the parameter relation graph meet preset abnormal conditions, wherein the parameter relation graph is a relation graph of the normal parameter value of the target parameter in the process step, which changes along with each preset time point.

Step 230, adjusting the process according to the adjustment scheme corresponding to the preset abnormal condition.

And 240, outputting prompt information corresponding to the preset abnormal conditions.

In the parameter anomaly detection method provided by the invention, the parameter value of the target parameter in the preset time period of each process step in the process is collected; determining that the target parameter in the process step is abnormal under the condition that the relation between the detected parameter value and the parameter relation graph obtained in advance meets the preset abnormal condition; the parameter relation graph is used for indicating the relation between time in the process and a target parameter value; the acquired actual parameter values can be compared with the target parameter values at corresponding moments in the parameter relation diagram, and whether parameter abnormality occurs in the technological process can be accurately and quickly detected; therefore, the corresponding adjusting scheme can be triggered to adjust the process rapidly and efficiently, and the prompt message corresponding to the preset abnormal condition is output, so that the yield of the equipment can be improved, the abnormal handling capacity of the equipment can be improved, and the safety is improved by increasing the detection speed of the parameter abnormality.

The contents of steps 210-240 are described below:

and 210, acquiring actual parameter values and target parameters of the target parameters at all preset time points in the process step.

Wherein, the parameter values of the target parameters mentioned above may include: temperature, flow, radio frequency and pressure and butterfly valve angle.

As shown in fig. 3, the relationship between the target parameter values of each process step in the process as time goes on is represented in the parameter relationship diagram. The parameter relation graph can be determined according to the relation between the time set in each process step of the normal process and the target parameter value. The first dimension (abscissa) of the parameter relationship diagram may be time, the time is generally the process time of the process step, the numerical value of the first dimension represents preset time points, the interval of each preset time point is preset time precision, the preset time point is 0 at the beginning and is step time of the process step at the end, the unit is second, illustratively, the preset time points are respectively 0-23, and the preset time precision is 1 s. The second dimension (ordinate) of the parameter map may be the target parameter value, i.e. the curve of the target parameter value over time in the corresponding process step. And the point corresponding to the change curve of each preset time point is the normal parameter value of the target parameter value at the preset time point. The accuracy and unit of the second dimension are determined according to the corresponding target parameters. Specifically, the parameter relationship diagram is determined according to the process recipe and the parameter type of the process step, and the required time precision and the parameter precision requirement. The higher the precision and the higher the complexity, the higher the precision of calculating the fluctuation of the actual parameter value and the target parameter value, and the fewer the false alarm and missed alarm conditions.

And when the technological process is skipped, namely the technological process is carried out from one technological step to the next technological step, the parameter relation graph of the corresponding technological step is obtained again.

In an embodiment, in step 220, when the actual parameter value at the preset time point, which is changed along with the preset time precision, of the target parameter and the normal parameter value at the preset time point corresponding to the parameter relation diagram satisfy the preset abnormal condition, determining that the target parameter in the process step is abnormal may specifically include the following steps:

determining a preset parameter fluctuation interval corresponding to each preset time point by the normal parameter value according to the parameter relation graph;

counting to obtain abnormal times in response to the fact that the actual parameter value corresponding to the preset time point exceeds the parameter fluctuation interval corresponding to the preset time point;

if the abnormal times are larger than a preset first threshold value, determining that the target parameters in the process steps are abnormal, and the target parameters are in a first abnormal state;

and if the abnormal times are less than a preset first threshold value, determining the deviation between the actual parameter value corresponding to each preset time point and the corresponding normal parameter value, and if the deviation meets a preset abnormal condition, determining that the target parameter in the process step is abnormal.

And determining a parameter fluctuation interval corresponding to each preset time point of the normal parameter value according to the parameter relation graph, wherein the parameter fluctuation interval can be specifically shown as a black curve in fig. 3. Each preset time point has an allowable deviation, namely an 'i' -shaped solid line in the vertical direction in fig. 3, and if the actual parameter value appears in the parameter fluctuation interval, it indicates that the target parameter at the preset time point is normal; and if the actual parameter value exceeds the parameter fluctuation interval, indicating that the target parameter at the preset time point is abnormal.

Generally, according to the existing detection technology, when the parameter value suddenly jumps abnormally and then returns to the normal range in the process, the parameter is not considered to be abnormal, and the target parameter abnormality in the process step can be found only when the process finishes detecting the silicon wafer process result.

The parameter abnormity detection method provided by the embodiment of the invention processes the parameter abnormity detection method. As shown in fig. 4, when the frequency of the parameter value exceeding the parameter fluctuation interval is very high, the process result is also affected. If the abnormality is detected in the process and the abnormality processing is carried out in advance, much time can be saved and the productivity of the equipment can be improved.

For the situation, counting is carried out in response to the fact that the actual parameter value corresponding to the preset time point exceeds the parameter fluctuation interval corresponding to the preset time point, and abnormal times are obtained; and if the abnormal times are larger than a preset first threshold value, determining that the target parameters in the process steps are abnormal, and the target parameters are in a first abnormal state.

Specifically, a mode of an exception counter may be adopted, the counter is valid in the process step, and the counter is cleared after the step skipping. When one abnormity occurs in the process step, adding 1 to an abnormity counter, and accumulating to obtain the abnormity times; determining that the target parameter in the process step is abnormal under the condition that the abnormal times are greater than a preset first threshold value; and determining that the target parameter in the process step is abnormal under the condition that the abnormal times are less than or equal to the first threshold value.

As shown in fig. 4, a certain target parameter in the process step jumps 7 times in total, and if the preset first threshold is 5 times, it indicates that the number of times of abnormality is greater than the preset first threshold, and the target parameter in the process step is abnormal.

Generally, according to the existing parameter detection technology, the fluctuation of the actual parameter value at each time point is within a reasonable range, or the number of abnormalities is smaller than a preset first threshold, and then it is considered to be normal. The target parameter abnormality in the process step can only be found when the process is finished and the silicon wafer process result is detected.

In view of the above situation, the embodiment of the present application designs the detection method, and determines a deviation between an actual parameter value and a corresponding normal parameter value corresponding to each preset time point if the number of abnormal times is less than a preset first threshold, and determines that a target parameter in a process step is abnormal if the deviation satisfies a preset abnormal condition.

As shown in fig. 5, before the process step skipping, the calculated accumulated deviation of the process step exceeds the tolerable parameter fluctuation interval of the process step. If the abnormity is detected before jumping to the next step and then abnormity processing is carried out, much time can be saved, and the equipment productivity is improved.

Specifically, the deviation between the actual parameter value corresponding to each preset time point and the corresponding normal parameter value may be determined, and if the deviation satisfies a preset abnormal condition, it is determined that the target parameter in the process step is abnormal.

In one possible embodiment, if the target parameter is in the first abnormal state, adjusting the process according to the adjustment scheme corresponding to the preset abnormal condition includes: terminating the process;

outputting prompt information corresponding to the preset abnormal condition, comprising: and outputting first prompt information, wherein the first prompt information is used for prompting the wiring condition of the inspection equipment and/or inspecting the air source parameters of the factory building.

And determining that the target parameter in the process step is abnormal under the condition that the abnormal times are greater than a preset first threshold value. The abnormity is mostly caused by hardware or environmental factors such as line virtual connection, unstable pressure of each air source of a factory building and the like, process compensation can not be carried out according to fluctuation deviation due to unpredictability and quantization, and once the abnormity processing mode is triggered, the process is directly terminated no matter what parameter types.

And inputting first prompt information for prompting equipment personnel to check the wiring condition of the equipment or to investigate the air source of a workshop and the like.

It should be noted that fig. 5 shows a case where the number of anomalies of the parameter fluctuation interval corresponding to the actual parameter value exceeding the preset time point is 0, and also a case where the number of anomalies of the parameter fluctuation interval corresponding to the actual parameter value exceeding the preset time point is smaller than a preset first threshold, where the preset first threshold is a positive integer.

In one embodiment, determining a deviation between an actual parameter value corresponding to each preset time point and a corresponding normal parameter value, and if the deviation satisfies a preset abnormal condition, determining that a target parameter in the process step is abnormal, including:

under the condition that the actual parameter value at the preset time point is larger than the normal parameter value corresponding to the preset time point, the accumulated larger value between the actual parameter value and the normal parameter value is calculated;

under the condition that the actual parameter value at the preset time point is smaller than the normal parameter value corresponding to the preset time point, the accumulative minimum value between the actual parameter value and the normal parameter value is counted;

if the difference value between the accumulated larger value and the preset larger threshold value is larger than a preset second threshold value, determining that the target parameter in the process step is abnormal; and/or

And if the difference value between the accumulated smaller value and the preset smaller threshold value is larger than a preset third threshold value, determining that the target parameter in the process step is abnormal.

Illustratively, the target parameter is temperature, for a given preset time point, the parameter value of the preset time point is 20 degrees celsius, the target parameter value corresponding to the preset time point in the parameter relation graph is 15 degrees celsius, the larger value of the preset time point is 5 degrees celsius, and so on, the larger value corresponding to each time point in the preset time period is accumulated to obtain an accumulated larger value. And determining that the target parameter in the process step is abnormal under the condition that the difference value (20 ℃) between the accumulated larger value (such as 50 ℃) and a preset larger threshold (such as 30 ℃) is larger than a preset second threshold (10 ℃).

Illustratively, the target parameter is pressure, for a given preset time point, the parameter value of the preset time point is 20 pa, the target parameter value corresponding to the preset time point in the parameter relation graph is 30 pa, the minimum value of the preset time point is 10 pa, and so on, the minimum value corresponding to each preset time point in the preset time period is accumulated to obtain an accumulated minimum value. And determining that the target parameter in the process step is abnormal under the condition that the difference value (20 Pa) between the accumulated smaller value (such as 50 Pa) and the preset smaller threshold value (such as 30 Pa) is greater than the preset third threshold value (10 Pa).

In a possible embodiment, determining a deviation between an actual parameter value corresponding to each preset time point and a corresponding normal parameter value, and if the deviation satisfies a preset abnormal condition, determining that a target parameter in the process step is abnormal, including:

determining the difference value between the actual parameter value of each preset time point and the normal parameter value corresponding to each preset time point, and summing the difference values to obtain a total deviation value;

if the total deviation value is larger than zero and the difference value between the total deviation value and the preset fourth threshold value is larger than zero, determining that the target parameter is abnormal in the process step and the target parameter is in a second abnormal state;

and if the total deviation value is smaller than zero and the difference value between the absolute value of the total deviation value and the preset fifth threshold value is larger than zero, determining that the target parameter is abnormal in the process step and the target parameter is in a third abnormal state.

And when the requirement of the process on the parameter stability is high and the large abnormality and the small abnormality cannot be mutually offset, namely the difference value between the total deviation value obtained by accumulating the deviation values corresponding to the preset time points of each acquired parameter value and the preset fourth threshold value is greater than zero, determining that the target parameter is abnormal in the process step and the target parameter is in a second abnormal state.

Illustratively, the cumulative maximum value is 50 degrees Celsius, and the cumulative minimum value is-20 degrees Celsius; the total deviation value is 30 degrees celsius, the 30 degrees celsius (total deviation value) is greater than zero, and a difference between the 30 degrees celsius (total deviation value) and the 10 degrees celsius (preset fourth threshold value) is greater than zero (30 degrees celsius-10 degrees celsius is 20 degrees celsius), it is determined that the target parameter is abnormal in the process step and the target parameter is in the second abnormal state.

Illustratively, a cumulative upper value of 50 degrees Celsius and a cumulative lower value of-80 degrees Celsius are obtained; the total deviation value is-30 degrees centigrade, the-30 degrees centigrade (total deviation value) is less than zero, and the difference value between the total deviation value and the preset fifth threshold value is as follows: and if the temperature is 20 ℃ or more, determining that the target parameter in the process step is abnormal and the target parameter is in a third abnormal state.

In a possible embodiment, the semiconductor processing equipment is a coating equipment, wherein if the target parameter is a temperature parameter and the processing step is a deposition step, and the target parameter is in a second abnormal state, the process is adjusted according to an adjustment scheme corresponding to a preset abnormal condition, including:

terminating the process;

if the target parameter is a radio frequency parameter and the process step is a deposition step, and the target parameter is in a second abnormal state, adjusting the process according to an adjustment scheme corresponding to a preset abnormal condition, including:

carrying out process compensation treatment on the process;

if the target parameter is a temperature parameter, the process step is a deposition step or a heating step, and the target parameter is in a third abnormal state, adjusting the process according to an adjustment scheme corresponding to a preset abnormal condition, including:

and carrying out process compensation treatment on the process progress.

If the target parameter is a temperature parameter and the process step is a deposition step, the condition that the target parameter is in a second abnormal state can cause the film coating result of the silicon chip to be thicker, and the process is directly stopped;

under the condition that the target parameter is a radio frequency parameter and the process step is a deposition step, performing compensation treatment on the process; because only the deposition step generates radio frequency arc, the condition that the target parameter is in the second abnormal state can lead the film coating result of the silicon chip to be thinner, and at this time, a processing mode of compensating the process step for a certain time before the step jump can be adopted to ensure the film coating thickness. Wherein the compensation time may be determined based on the number of radio frequency arcs that occur.

And if the target parameter is a temperature parameter, the process step is a deposition step or a heating step, and the target parameter is in a third abnormal state, performing process compensation treatment on the process progress.

If the process step is a temperature rise step or a deposition step and the target parameter is in a third abnormal state, the film coating result of the silicon wafer is thin, and a compensation measure for compensating the process step for a certain time before the jump step can be adopted to ensure the film coating thickness. If the other process steps are adopted, the influence is small and can be ignored.

In a possible embodiment, determining that the target parameter in the process step is abnormal when the normal parameter value at the preset time point corresponding to the relationship between the actual parameter value at the preset time point and the normal parameter value at the preset time point in the parameter relationship diagram, where the target parameter varies with the preset time accuracy, meets a preset abnormal condition includes:

comparing the actual parameter value corresponding to each preset time point with the corresponding normal parameter value;

if the actual parameter value corresponding to the preset time point is larger than the corresponding normal parameter value and the larger value is larger than a preset sixth threshold value, counting the larger times to obtain the total larger times;

if the actual parameter value corresponding to the preset time point is smaller than the corresponding normal parameter value and the smaller value is smaller than a preset seventh threshold value, counting the smaller times to obtain the total smaller times;

under the condition that the total large times and the total small times meet a preset time condition, carrying out Fourier transform on the actual parameter value to obtain a frequency spectrum curve corresponding to the actual parameter value;

and determining that the target parameters in the process steps are abnormal under the condition that the frequency spectrum curve meets the preset discrete condition.

In general, the conventional parameter detection technology is considered to be normal when the fluctuation of the periodic waveform of the sawtooth shape is abnormal, that is, when the parameter value becomes abnormally large and small. Such anomalies generally do not affect the process results, but are indicative of a problem with the corresponding hardware control capabilities of the equipment.

If the hardware is processed and maintained in time, the abnormity affecting the process can be avoided. Can save much time and improve the productivity of the equipment.

Aiming at the situation, the actual parameter values corresponding to the preset time points and the corresponding normal parameter values can be compared; as shown in fig. 6, if the actual parameter value corresponding to the preset time point is greater than the corresponding normal parameter value, and the large value is greater than the preset sixth threshold, counting the large number of times to obtain the total large number of times; and if the actual parameter value corresponding to the preset time point is smaller than the corresponding normal parameter value and the small value is smaller than a preset seventh threshold value, counting the small times to obtain the total small times.

Under the condition that the total large number of times and the total small number of times satisfy the preset number of times condition, performing fourier transform on the actual parameter value to obtain a spectrum curve corresponding to the actual parameter value, which may specifically include:

if the total number of larger times N1> the expected larger threshold; and the total number of partial minima N2> the expected minimum threshold; and | N1-N2| < δ 1(δ 1 is close to 0).

Namely, the total large times and the total small times exceed the expected threshold, and the difference between the total large times and the total small times is not much, the Fourier transform is performed on the parameter value to obtain a frequency spectrum curve corresponding to the parameter value, so as to start to judge whether the target parameter has periodicity.

The Fourier transform is one of the most basic methods in time domain and frequency domain transform analysis, and the target parameter abnormality in the process step is determined under the condition that a frequency spectrum curve corresponding to the parameter value obtained by the Fourier transform meets a preset discrete condition.

Wherein, under the condition that the frequency spectrum curve meets the preset discrete condition, determining that the target parameter in the process step is abnormal comprises the following steps:

determining the number of actual parameter values of which the frequency values are smaller than a preset eighth threshold value from the frequency spectrum curve;

and if the ratio of the number of the actual parameter values of which the frequency values are smaller than the preset eighth threshold value to the number of the preset time points is larger than the preset ninth threshold value, determining that the target parameter in the process step is abnormal.

If the parameter value changes periodically, the spectrum curve is discrete, as shown in fig. 7, and the parameter value only appears in a limited number of frequencies; the target parameter is only in a limited number of frequencies, and is in abnormal sawtooth periodic fluctuation, so that the target parameter in the process step can be determined to be abnormal, and abnormal treatment needs to be triggered.

If the actual parameter values vary non-periodically, the spectral curve is continuous, as shown in FIG. 8, and the actual parameter values occur randomly within each frequency. The target parameters are randomly appeared in each frequency and do not belong to abnormal sawtooth periodic fluctuation, and the target parameters in the process steps can be determined not to be abnormal.

In fig. 7 and 8, the abscissa represents frequency, and the ordinate represents parameter values.

Comparing each point on the spectrum curve N ' (f) with a value of 0, if N ' (f) -0< δ 2(δ 2 is close to 0), the point may be defaulted to 0, determining the number N3 of parameter values having frequency values smaller than the preset eighth threshold value (δ 2) to be N3+1, and if N ' (f) -0> δ 2, defaulted to be a non-0 value;

when the ratio of the number (N3) to the sampling times (N4) is greater than a preset ninth threshold (90%), it indicates that the target parameter only appears in a limited number of frequencies, and is an abnormal sawtooth periodic fluctuation, and it can be determined that the target parameter in the process step is abnormal, and an abnormal process needs to be triggered.

When the ratio of the number (N3) to the sampling times (N4) is smaller than a preset ninth threshold (90%), it indicates that the target parameter randomly appears in each frequency and does not belong to abnormal sawtooth periodic fluctuation, and it can be determined that the process is not abnormal.

In one possible embodiment, the target parameter is a temperature parameter, and the adjusting the process according to the adjustment scheme corresponding to the preset abnormal condition includes:

under the condition that the process is finished, outputting an adjusting instruction to the temperature control meter, wherein the instruction is used for adjusting the temperature parameter by the temperature control meter;

the target parameter is an angle parameter of the butterfly valve, and the process is adjusted according to an adjustment scheme corresponding to a preset abnormal condition, and the process comprises the following steps:

and outputting a learning instruction to the butterfly valve under the condition that the process is finished, wherein the learning instruction is used for learning and adjusting the angle parameters of the butterfly valve.

On one hand, under the condition that the process is finished, an adjusting instruction is output to the temperature control meter, and the instruction is used for the temperature control meter to adjust the temperature parameters.

The parameters that are typically required to handle the periodic, anomalous fluctuations in the sawtooth shape are temperature and butterfly valve angle. After the equipment runs for a long time, the stability of the equipment is poor, and the equipment is indicated to be required to be maintained when the abnormality occurs, so that the current process is not processed, but the temperature and the butterfly valve maintenance mode are automatically touched by software after the process is finished, and the process can be continued after the maintenance.

Aiming at the target parameter being the temperature parameter, the temperature control table has a self-tuning function, and after the spectral curve is detected to meet the preset discrete condition, after the process is finished, an adjustment instruction can be automatically issued to the temperature control table so that the temperature control table can complete the adjustment of a temperature control PID (proportional-Integral-Differential) controller. Among them, a PID controller that controls according to a proportion (P), an integral (I), and a derivative (D) of a deviation in process control is one of the most widely used automatic controllers.

On the other hand, when the process is finished, a learning instruction is output to the butterfly valve, and the learning instruction is used for the butterfly valve to learn and adjust the angle parameters.

A butterfly valve, also called a flap valve, is a regulating valve with simple structure, and can be used for the on-off control of low-pressure pipeline media, namely a valve which is opened and closed by rotating around a valve shaft, wherein a closing member (a valve clack or a butterfly plate) is a disk. The valve can be used for controlling the flow of various types of fluids such as air, water, steam, various corrosive media, slurry, oil products, liquid metal, radioactive media and the like. The pipe mainly plays a role in cutting off and throttling. The butterfly valve opening and closing piece is a disc-shaped butterfly plate and rotates around the axis of the butterfly plate in the valve body, so that the opening and closing or adjusting purpose is achieved.

Aiming at the condition that the target parameter is the angle parameter, the butterfly valve has a learning function, and after the frequency spectrum curve is detected to meet the preset discrete condition, a learning command can be automatically issued to the butterfly valve after the process is finished, so that the adjustment of PI (proportion (P) and integral (I) of deviation) is completed.

In the process, the acquired actual parameter values can be compared with the normal parameter values at the corresponding moments in the parameter relation diagram, so that whether the target parameters in the process steps are abnormal or not can be detected more accurately; the abnormity can be detected earlier, and the time consumed for processing the abnormity is shorter; the abnormal state corresponding to the parameter can be detected more accurately, and the corresponding abnormal processing scheme can be triggered more accurately, so that the capacity of the equipment can be improved, and the abnormal processing capability of the equipment can also be improved.

According to the embodiment of the invention, the actual parameter values of the target parameters at all preset time points in the process steps are collected; dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter at a preset time point, which changes along with the preset time precision, and the normal parameter value of the corresponding preset time point in the parameter relation graph meet a preset abnormal condition, wherein whether the parameter is abnormal in the process step can be accurately and quickly detected by comparing the actual parameter value of the target parameter in the process step with the normal parameter value of the target parameter in the process step; therefore, the corresponding adjusting scheme can be triggered to adjust the process rapidly and efficiently, and the prompt message corresponding to the preset abnormal condition is output, so that the yield of the equipment can be improved, the abnormal handling capacity of the equipment can be improved, and the safety is improved by increasing the detection speed of the parameter abnormality.

In addition, based on the parameter abnormality detection method shown in fig. 2 and the abnormal situation shown in fig. 6, the present invention further provides a method for implementing parameter abnormality detection, and fig. 9 is a flowchart of a method for implementing parameter abnormality detection according to an embodiment of the present invention. The details are as follows:

1. when the process starts or jumps, firstly, clearing the sum of the accumulated large values S1, the sum of the accumulated small values S2 and the sum of the total accumulated deviation values S, namely S1 is 0, S2 is 0 and S3 is 0;

2. acquiring actual values of target parameters according to the time precision of the two-dimensional matrix, and dispersing the actual values of the parameters into the two-dimensional matrix;

3. comparing the actual value d (t) with a normal value b (t) at a corresponding time in the curve to obtain a deviation value n (t), wherein n (t) is d (t) -b (t);

4. if n (t) >0, calculating the cumulative maximum S1, S1 ═ S1+ n (t);

5. if n (t) <0, calculating the cumulative minimum S2, S2 ═ S2+ | n (t) |, where | n (t) | represents the absolute value of the deviation;

6. if the process step time is over, the step 7 is carried out, and if the process step time is not over, the

steps

2, 3, 4 and 5 are repeated

7. After the process step time is finished, calculating the total accumulated deviation value S of the step, namely S1-S2;

8. judging whether the accumulated larger value S1 is abnormal: comparing S1 with the expected threshold P1, if S1-P1 is greater than 0, it belongs to (fluctuation exception 1) larger exception, and the exception handling module may be triggered. If the S1-P1 is less than or equal to 0, the operation is normal;

9. judging the cumulative minimum value S2_NWhether or not abnormal is detected S2_NComparing with the expected threshold P2 if S2_N-P2 is greater than 0,then it is a (fluctuation exception 2) very small exception and the exception handling module may be triggered. If S2_N-P2 is less than or equal to 0, then normal;

10. judging whether the total accumulated deviation S is abnormal, comparing S with an expected threshold value P, and if S is abnormal_N>0 and S_NIf P is greater than 0, it belongs to a (fluctuation exception 3) big exception, which may trigger the exception handling module; if S_N<0 and | S_NIf the | P is larger than 0, the (fluctuation exception 4) is a small exception, and an exception processing module can be triggered; the others are normal.

Here, by determining a deviation value between the parameter value and the target parameter value in response to detecting that the parameter value does not match the target parameter value; the time point corresponding to the parameter value is consistent with the time point corresponding to the target parameter value; and determining that the target parameter in the process step is abnormal under the condition that the deviation value meets a preset abnormal condition. In the technical process, whether the abnormality occurs in the technical process can be detected more accurately; the abnormity can be detected earlier, and the time consumed for processing the abnormity is shorter; the parameter exception type can be detected more accurately, and the corresponding exception handling scheme can be triggered more accurately, so that the equipment productivity can be improved, and the exception handling capacity of the equipment can also be improved.

In addition, based on the parameter abnormality detection method shown in fig. 2, for the abnormal condition shown in fig. 6, the present invention further provides another method for implementing parameter abnormality detection, and fig. 10 is a flowchart of another method for implementing parameter abnormality detection provided in the embodiment of the present invention. The details are as follows:

1. the process step starts, clearing an accumulated large number counter N1, an accumulated small number counter N2, a period detection counter N3 and a sampling counter N4, namely N1 is equal to 0; n2 ═ 0; n3 ═ 0; n4 ═ 0; wherein:

cumulative number of times counter N1: every time a big exception occurs, the counter is increased by 1;

cumulative number of small counts counter N2: every time a small exception occurs, the counter is increased by 1;

cycle detection counter N3: when detecting whether the actual value of the parameter has periodicity, adding 1 to the counter every time an approximate 0 value appears in the spectrogram;

sample counter N4: the process step has a total of how many samples, and the counter is incremented by 1 for each sample.

2. Acquiring actual values of target parameters according to the time precision T of the two-dimensional matrix, dispersing the actual values of the parameters into the two-dimensional matrix, and acquiring data once, wherein a sampling counter N4 is N4+ 1;

3. comparing the target parameter actual value D (t) with a normal value B (t) at a corresponding moment in the curve to obtain a deviation value N (t), wherein N (t) is D (t) -B (t);

4. judging the fluctuation range of N (t), if N (t) is greater than 0 and greater than an expected threshold value, judging that the fluctuation is larger and abnormal, and setting an accumulated larger number counter N1 to be N1+ 1; if N (t) is less than 0 and less than the expected threshold, the fluctuation is small and abnormal, and the cumulative small number counter N2 is N2+ 1;

5. if the process step time is over, performing the step 6, and if the process step time is not over, repeating the

steps

2, 3 and 4;

6. after the process step time is finished, whether periodic detection is needed or not is judged, if N1 is larger than an expected larger threshold value, N2 is larger than an expected smaller threshold value, and | N1-N2| < delta 1 (delta 1 is close to 0), namely, the accumulated larger and accumulated smaller number counters exceed the expected threshold value and have little difference, whether the target parameter has periodic judgment or not is judged;

7. and carrying out Fourier transformation on the actual value curve D (t) of the target parameter to obtain a frequency spectrum curve N' (f) of the actual value of the target parameter. In the case of a periodic function, the spectral curve N' (f) is discrete, as shown in fig. 9, with the parameter values only occurring within a limited number of frequencies; if it is a non-periodic function, the N' (f) spectrum is continuous, as shown in FIG. 10, with the parameter values occurring randomly at various frequencies. In fig. 9 and 10, the abscissa represents frequency, and the ordinate represents parameter values.

8. Comparing each point on the spectrum curve N ' (f) with a value of 0, if N ' (f) -0< δ 2(δ 2 is close to 0), the point may be defaulted to 0, the cycle detection counter N3 ═ N3+1, if N ' (f) -0> δ 2, the point may be defaulted to a non-0 value;

9. if the proportion of the periodic detection counter N3 in the sampling counter N4 exceeds 90%, namely N3/N4 is more than 90%, the target parameter is considered to be only within a limited number of frequencies and to belong to abnormal sawtooth periodic fluctuation, and abnormal processing needs to be triggered;

10. if the proportion of the period detection counter N3 in the sampling counter N4 does not exceed 90%, namely N3/N4 is less than 90%, the target parameter is considered to randomly appear in each frequency and not belong to the abnormal sawtooth periodic fluctuation, and the abnormal condition is ignored and not triggered.

It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.

Referring to fig. 11, there is shown a block diagram of a semiconductor processing apparatus 1110, including,

the controller 1111 is used for acquiring parameter values of target parameters within a preset time period in a process; determining that the target parameter in the process step is abnormal under the condition that the relation between the detected parameter value and the parameter relation graph obtained in advance meets the preset abnormal condition; the parameter relation graph is used for indicating the time in each process step of the process and the change relation of the parameter value of the target parameter; adjusting the process according to an adjustment scheme corresponding to a preset abnormal condition; and outputting prompt information corresponding to the preset abnormal conditions.

In an optional embodiment of the present invention, the controller 1111 is specifically configured to determine, according to the parameter relationship map, a preset parameter fluctuation interval corresponding to each preset time point for the normal parameter value;

counting to obtain abnormal times in response to detecting that the actual parameter value corresponding to the preset time point exceeds the parameter fluctuation interval corresponding to the preset time point;

if the abnormal times are larger than a preset first threshold value, determining that the target parameter in the process step is abnormal, and the target parameter is in a first abnormal state;

if the abnormal times are smaller than the preset first threshold, determining the deviation between the actual parameter value corresponding to each preset time point and the corresponding normal parameter value, and if the deviation meets the preset abnormal condition, determining that the target parameter in the process step is abnormal.

In an alternative embodiment of the present invention, if the target parameter is in the first abnormal state, the controller 1111 is specifically configured to terminate the process; and outputting first prompt information, wherein the first prompt information is used for prompting the wiring condition of the inspection equipment and/or inspecting the air source parameters of the factory building.

In an optional embodiment of the present invention, the controller 1111 is specifically configured to, when the actual parameter value at the preset time point is greater than the normal parameter value corresponding to the preset time point, count an accumulated larger value between the actual parameter value and the normal parameter value;

under the condition that the actual parameter value at the preset time point is smaller than the normal parameter value corresponding to the preset time point, calculating an accumulated small value between the actual parameter value and the normal parameter value;

if the difference value between the accumulated larger value and a preset larger threshold value is larger than a preset second threshold value, determining that the target parameter in the process step is abnormal; and/or

And if the difference value between the accumulated smaller value and a preset smaller threshold value is larger than a preset third threshold value, determining that the target parameter in the process step is abnormal.

In an alternative embodiment of the present invention, the controller 1111 is specifically configured to: determining difference values of the actual parameter values of the preset time points and the normal parameter values corresponding to the preset time points, and summing the difference values to obtain a total deviation value;

if the total deviation value is larger than zero and the difference value between the total deviation value and a preset fourth threshold value is larger than zero, determining that the target parameter in the process step is abnormal and the target parameter is in a second abnormal state;

and if the total deviation value is smaller than zero and the difference value between the absolute value of the total deviation value and a preset fifth threshold value is larger than zero, determining that the target parameter in the process step is abnormal and the target parameter is in a third abnormal state.

In an optional embodiment of the present invention, the semiconductor processing equipment is a coating equipment, wherein if the target parameter is a temperature parameter and the process step is a deposition step, the controller 1111 is specifically configured to: terminating the process;

if the target process parameter is a radio frequency parameter and the process step is a deposition step, and the target parameter is in the second abnormal state, the controller 1111 is specifically configured to: carrying out process compensation treatment on the process progress;

if the target process parameter is a temperature parameter, the process step is a deposition step or a temperature rise step, and the target parameter is in a third abnormal state, the controller 1111 is specifically configured to: and carrying out process compensation treatment on the process progress.

In an alternative embodiment of the present invention, the controller 1111 is specifically configured to: comparing the actual parameter values corresponding to the preset time points with the corresponding normal parameter values;

counting the small times to obtain the total small times if the actual parameter value corresponding to the preset time point is smaller than the corresponding normal parameter value and the small value is smaller than a preset seventh threshold value;

under the condition that the total large times and the total small times meet a preset time condition, performing Fourier transform on the actual parameter value to obtain a frequency spectrum curve corresponding to the actual parameter value;

and determining that the target parameter in the process step is abnormal under the condition that the frequency spectrum curve meets the preset discrete condition.

In an alternative embodiment of the present invention, the controller 1111 is specifically configured to: determining the number of actual parameter values of which the frequency values are smaller than a preset eighth threshold value from the frequency spectrum curve;

and if the ratio of the number of the actual parameter values of which the determined frequency values are smaller than the preset eighth threshold value to the number of the preset time points is larger than a preset ninth threshold value, determining that the target parameter in the process step is abnormal.

In an alternative embodiment of the present invention, the controller 1111 is specifically configured to: under the condition that the process is finished, outputting an adjusting instruction to a temperature control meter, wherein the instruction is used for the temperature control meter to adjust the temperature parameter; the target parameter is an angle parameter of a butterfly valve, and the process is adjusted according to an adjustment scheme corresponding to the preset abnormal condition, and the process comprises the following steps:

and outputting a learning instruction to a butterfly valve under the condition that the process is finished, wherein the learning instruction is used for the butterfly valve to learn and adjust the angle parameter.

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

An embodiment of the present invention further provides an electronic device, including: the processor, the memory, and the computer program stored in the memory and capable of running on the processor, when executed by the processor, implement the processes of the above-mentioned parameter anomaly detection method embodiment, and can achieve the same technical effects, and are not described herein again to avoid repetition.

The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when executed by a processor, the computer program implements each process of the above-mentioned parameter anomaly detection method embodiment, and can achieve the same technical effect, and is not described herein again to avoid repetition.

The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

The parameter anomaly detection method and the semiconductor process equipment provided by the invention are described in detail, specific examples are applied in the method for explaining the principle and the implementation mode of the invention, and the description of the examples is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. A parameter anomaly detection method is applied to semiconductor process equipment, the process of the semiconductor process equipment comprises a plurality of process steps which are carried out in sequence, and the parameter anomaly detection method is characterized by comprising the following steps:

dispersing the actual parameter value of the target parameter into a pre-acquired parameter relation graph, and determining that the target parameter in the process step is abnormal under the condition that the actual parameter value of the target parameter at a preset time point, which changes along with preset time precision, and the normal parameter value of the corresponding preset time point in the parameter relation graph meet preset abnormal conditions, wherein the parameter relation graph is a relation graph of the normal parameter value of the target parameter in the process step, which changes along with each preset time point;

adjusting the process according to an adjustment scheme corresponding to the preset abnormal condition;

2. The method according to claim 1, wherein the determining the abnormal parameter relationship diagram of the target parameter in the process step when the actual parameter value at a preset time point, which varies with the preset time precision, and the normal parameter value at the preset time point corresponding to the parameter relationship diagram satisfy a preset abnormal condition comprises:

determining a preset parameter fluctuation interval of the normal parameter value corresponding to each preset time point according to the parameter relation chart of the parameter relation chart;

3. The method of claim 2, wherein if the target parameter is in the first abnormal state, the adjusting the process recipe according to the adjustment scenario corresponding to the predetermined abnormal condition comprises: terminating the process;

the outputting of the prompt information corresponding to the preset abnormal condition includes: and outputting first prompt information, wherein the first prompt information is used for prompting the wiring condition of the inspection equipment and/or inspecting the air source parameters of the factory building.

4. The method according to claim 2, wherein the determining a deviation between the actual parameter value corresponding to each of the predetermined time points and the corresponding normal parameter value, and if the deviation satisfies the predetermined abnormal condition, determining that the target parameter in the process step is abnormal comprises:

under the condition that the actual parameter value at the preset time point is larger than the normal parameter value corresponding to the preset time point, calculating an accumulated larger value between the actual parameter value and the normal parameter value;

5. The method according to claim 2, wherein determining a deviation between the actual parameter value corresponding to each of the predetermined time points and the corresponding normal parameter value, and if the deviation satisfies the predetermined abnormal condition, determining that the target parameter in the process step is abnormal comprises:

determining difference values of the actual parameter values of the preset time points and the normal parameter values corresponding to the preset time points, and summing the difference values to obtain a total deviation value;

6. The method of claim 5, wherein the semiconductor processing equipment is a coating equipment, and wherein if the target parameter is a temperature parameter and the processing step is a deposition step, and the target parameter is in a second abnormal state, the adjusting the process according to the adjustment scheme corresponding to the preset abnormal condition comprises:

terminating the process;

if the target parameter is a radio frequency parameter and the process step is a deposition step, and the target parameter is in a second abnormal state, adjusting the process according to an adjustment scheme corresponding to the preset abnormal condition, including:

carrying out process compensation treatment on the process progress;

if the target parameter is a temperature parameter, the process step is a deposition step or a temperature rise step, and the target parameter is in a third abnormal state, adjusting the process according to an adjustment scheme corresponding to the preset abnormal condition, including:

and carrying out process compensation treatment on the process progress.

7. The method according to claim 1, wherein the determining that the target parameter in the process step is abnormal in the case that the actual parameter value at a preset time point, which varies with a preset time accuracy, of the target parameter and the normal parameter value at the preset time point, which corresponds to the parameter relation diagram, satisfy a preset abnormal condition comprises:

comparing the actual parameter values corresponding to the preset time points with the corresponding normal parameter values;

8. The method according to claim 7, wherein the determining that the target parameter in the process step is abnormal in the case that the spectrum curve satisfies the preset dispersion condition comprises:

9. The method of claim 7, wherein the target parameter is a temperature parameter, and wherein adjusting the process recipe according to the adjustment profile corresponding to the predetermined abnormal condition comprises:

under the condition that the process is finished, outputting an adjusting instruction to a temperature control meter, wherein the instruction is used for the temperature control meter to adjust the temperature parameter; the target parameter is an angle parameter of a butterfly valve, and the process is adjusted according to an adjustment scheme corresponding to the preset abnormal condition, and the process comprises the following steps:

10. A semiconductor processing apparatus, wherein a process of the semiconductor processing apparatus comprises a plurality of process steps performed in sequence, the semiconductor processing apparatus comprising:

the controller is used for acquiring actual parameter values of the target parameters at all preset time points in the process steps;